Pardis Sabeti
/
enMachine learning guides researchers to new synthetic genetic switches
/news/machine-learning-guides-researchers-new-synthetic-genetic-switches
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Wed, 23 Oct 2024 15:05:19 +0000tulrich@broadinstitute.org5557621 at Six scientists named as new core institute members at 映画传媒
/news/six-scientists-named-new-core-institute-members-broad
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Wed, 07 Aug 2024 12:56:43 +0000Corie Lok5557161 at Simple test for flu could improve diagnosis and surveillance
/news/simple-test-flu-could-improve-diagnosis-and-surveillance
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Tue, 18 Jun 2024 15:00:00 +0000adicorat5557056 at A new viral surveillance system in West Africa is showing the world how to prevent the next pandemic
/news/new-viral-surveillance-system-west-africa-showing-world-how-prevent-next-pandemic
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Thu, 25 Apr 2024 15:00:00 +0000adicorat5556791 at Researchers uncover genetic factors for severe Lassa fever
/news/researchers-uncover-genetic-factors-severe-lassa-fever
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Thu, 08 Feb 2024 14:00:00 +0000adicorat5556161 at Gene-delivering viruses reach the brain in a step toward gene therapy for neurological diseases
/news/gene-delivering-viruses-reach-brain-step-toward-gene-therapy-neurological-diseases
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Tue, 22 Nov 2022 23:54:31 +0000aviveros@broadinstitute.org1262191 at 映画传媒 Discovery Center in Cambridge opens to the public this October
/news/broad-discovery-center-cambridge-opens-public-october
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Mon, 19 Sep 2022 14:53:09 +0000kzusi@broadinstitute.org1195701 at Deployment of outbreak-simulating app on college campuses reveals hidden transmission pathways
/news/deployment-outbreak-simulating-app-college-campuses-reveals-hidden-transmission-pathways
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Mon, 15 Aug 2022 13:31:36 +0000adicorat1163211 at Computer model predicts dominant SARS-CoV-2 variants
/news/computer-model-predicts-dominant-sars-cov-2-variants
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Tue, 24 May 2022 13:24:02 +0000adicorat1133351 at Machine learning could help scientists design better viral diagnostics
/news/machine-learning-could-help-scientists-design-better-viral-diagnostics-0
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
<span class="field field--name-uid field--type-entity-reference field--label-hidden">
<span>By Tom Ulrich</span>
</span>
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" class="datetime">October 23, 2024</time>
</span>
<div class="hero-section container">
<div class="hero-section__row row">
<div class="hero-section__content hero-section__content_left col-6">
<div class="hero-section__breadcrumbs">
<div class="block block-system block-system-breadcrumb-block">
<nav class="breadcrumb" role="navigation" aria-labelledby="system-breadcrumb">
<h2 id="system-breadcrumb" class="visually-hidden">Breadcrumb</h2>
<ol>
<li>
<a href="/">Home</a>
</li>
<li>
<a href="/news">News</a>
</li>
</ol>
</nav>
</div>
</div>
<div class="hero-section__title">
<div class="block block-layout-builder block-field-blocknodelong-storytitle">
<span class="field field--name-title field--type-string field--label-hidden"><h1>Machine learning guides researchers to new synthetic genetic switches</h1>
</span>
</div>
</div>
<div class="hero-section__description">
<div class="block block-layout-builder block-field-blocknodelong-storybody">
<div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A new method allows precise activation or repression of genes in specific cells and tissues.</p>
</div>
</div>
</div>
<div class="hero-section__author">
<div class="block block-layout-builder block-extra-field-blocknodelong-storyextra-field-author-custom">
By Sarah C.P. Williams
</div>
</div>
<div class="hero-section__date">
<div class="block block-layout-builder block-field-blocknodelong-storycreated">
<span class="field field--name-created field--type-created field--label-hidden"><time datetime="2024-10-23T11:05:19-04:00" title="Wednesday, October 23, 2024 - 11:05" class="datetime">October 23, 2024</time>
</span>
</div>
</div>
</div>
<div class="hero-section__right col-6">
<div class="hero-section__image">
<div class="block block-layout-builder block-field-blocknodelong-storyfield-image">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item">
<article class="media media--type-image media--view-mode-multiple-content-types-header">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop_xl/public/longstory/CRE-Main.jpg?h=6d684022&itok=wUmIvEEv 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="754" height="503">
<source srcset="/files/styles/multiple_ct_header_desktop/public/longstory/CRE-Main.jpg?h=6d684022&itok=05a7B_H4 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="736" height="520">
<source srcset="/files/styles/multiple_ct_header_laptop/public/longstory/CRE-Main.jpg?h=6d684022&itok=ghhFBJ8s 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="641" height="451">
<source srcset="/files/styles/multiple_ct_header_tablet/public/longstory/CRE-Main.jpg?h=6d684022&itok=2__mXLQz 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="706" height="417">
<source srcset="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l 1x" media="all and (max-width: 539px)" type="image/jpeg" width="499" height="294">
<img loading="eager" width="499" height="294" src="/files/styles/multiple_ct_header_phone/public/longstory/CRE-Main.jpg?h=6d684022&itok=BUIZYr4l" alt="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " title="An illustration of four circular diagrams representing plasmids, each with colored segments indicating different pieces of regulatory DNA. " typeof="foaf:Image">
</picture>
</div>
<div class="media-caption">
<div class="media-caption__credit">
Credit: Natalie Velez, 映画传媒 Communications, from Gosai and Castro et al 2024.
</div>
<div class="media-caption__description">
</div>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block block-better-social-sharing-buttons block-social-sharing-buttons-block">
<div style="display: none"><link rel="preload" href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg" as="image" type="image/svg+xml" crossorigin="anonymous"></div>
<div class="social-sharing-buttons">
<a href="https://www.facebook.com/sharer/sharer.php?u=/taxonomy/term/625/feed&title=" target="_blank" title="Share to Facebook" aria-label="Share to Facebook" class="social-sharing-buttons-button share-facebook" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#facebook" />
</svg>
</a>
<a href="https://twitter.com/intent/tweet?text=+/taxonomy/term/625/feed" target="_blank" title="Share to X" aria-label="Share to X" class="social-sharing-buttons-button share-x" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#x" />
</svg>
</a>
<a href="mailto:?subject=&body=/taxonomy/term/625/feed" title="Share to Email" aria-label="Share to Email" class="social-sharing-buttons-button share-email" target="_blank" rel="noopener">
<svg aria-hidden="true" width="32px" height="32px" style="border-radius:100%;">
<use href="/modules/contrib/better_social_sharing_buttons/assets/dist/sprites/social-icons--no-color.svg#email" />
</svg>
</a>
</div>
</div>
<div class="block block-layout-builder block-field-blocknodelong-storyfield-content-paragraphs">
<div class="field field--name-field-content-paragraphs field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--text-with-sidebar text-with-sidebar">
<div class="field field--name-field-sidebar field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-menu sidebar-menu">
<div class="sidebar-menu__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related people</p>
</div>
<div class="field field--name-field-links field--type-link field--label-hidden field__items">
<div class="field__item"><a href="/node/7055">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--sidebar-articles sidebar-articles">
<div class="sidebar-articles__col">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Related news</p>
</div>
<div class="field field--name-field-content-reference field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><article about="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=zb7wS9Iz 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=ef9j8BfC 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=vWgkpcUk 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/blog/images/2016/MPRA_TesselatedPeaFlower_SigridKnemeyer_325x337_72dpi.jpg?itok=U93JLRDo" alt typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/blog/massive-approach-finding-whats-real-genome-wide-association-data" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">A massive approach to finding what's "real" in genome-wide association data</span>
</a>
</div>
</article>
</div>
<div class="field__item"><article about="/news/exploring-autoimmunitys-regulatory-roots" class="node">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-multiple-ct-sidebar-link-with-image">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <a href="/news/exploring-autoimmunitys-regulatory-roots"><picture>
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1921px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop_xl/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=WQ3RXfb0 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/jpeg" width="104" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_desktop/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=aVmSwkCK 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/jpeg" width="87" height="104">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_tablet/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=uc4wAwII 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/jpeg" width="285" height="186">
<source srcset="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg 1x" media="all and (max-width: 539px)" type="image/jpeg" width="220" height="186">
<img loading="eager" width="220" height="186" src="/files/styles/multiple_ct_sidebar_link_with_image_phone/public/news/images/2022/NIAID_T_lymphocyte_16760076824_8fab97208d_c_800x596.jpg?h=d3e04ee7&itok=jRkxOIYg" alt="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." title="Micrograph of a T cell, an immune cell type often implicated in autoimmune disorders." typeof="foaf:Image">
</picture></a>
</div>
</article>
</div>
<div class="node__content">
<a href="/news/exploring-autoimmunitys-regulatory-roots" class="node__title"><span class="field field--name-title field--type-string field--label-hidden">Exploring autoimmunity's regulatory roots </span>
</a>
</div>
</article>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Researchers at The Jackson Laboratory (JAX), the 映画传媒 of MIT and Harvard, and Yale University, have used artificial intelligence to design thousands of new DNA switches that can precisely control the expression of a gene in different cell types. Their new approach opens the possibility of controlling when and where genes are expressed in the body, for the benefit of human health and medical research, in ways never before possible. </p>
<p>"What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for," said Ryan Tewhey, an associate professor at The Jackson Laboratory and co-senior author of the work with Steven Reilly of Yale, and <a href="/node/7055">Pardis Sabeti</a> of the 映画传媒. "This creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."</p>
<p>In recent years, genetic editing technologies and other gene therapy approaches have given scientists the ability to alter the genes inside living cells. However, affecting genes only in selected cell types or tissues, rather than across an entire organism, has been difficult. That is in part because of the ongoing challenge of understanding the DNA switches, called cis-regulatory elements (CREs), that control the expression and repression of genes. </p>
<p>In a paper published in <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank"><em>Nature</em></a>, Tewhey, Reilly, Sabeti, and their collaborators not only designed new, never-before-seen synthetic CREs, but used the CREs to successfully activate genes in brain, liver or blood cells without turning on those genes in other cell types. </p>
<p>"The more we learn about the genome, the more we see evidence of the deep influence elements like CREs have on biological function," said Sabeti, who is a core institute member at 映画传媒 and a professor at Harvard University and the Harvard T. H. Chan School of Public Health. "By applying machine learning and molecular biology to the logic of when and where CREs work, we can leverage that knowledge using generative AI to build tools for modulating gene expression in new ways experimentally and, perhaps one day, therapeutically."</p>
<h2>Tissue- and time-specific instructions</h2>
<p>Although every cell in an organism contains the same genes, not all the genes are needed in every cell, or at all times. CREs help ensure that genes needed in the brain are not used by skin cells, for instance, or that genes required during early development are not activated in adults. CREs themselves are not part of genes, but are separate, regulatory DNA sequences 鈥� often located near the genes they control. </p>
<p>Scientists know that there are thousands of different CREs in the human genome, each with slightly different roles. But the grammar of CREs has been hard to figure out, "with no straightforward rules that control what each CRE does," explained Rodrigo Castro, a computational scientist in the Tewhey lab at JAX and co-first author of the new paper. "This limits our ability to design gene therapies that only effect certain cell types in the human body."</p>
<p>"This project essentially asks the question: 'Can we learn to read and write the code of these regulatory elements?'" said Reilly, who is an assistant professor of genetics at Yale and one of the senior authors of the study. "If we think about it in terms of language, the grammar and syntax of these elements is poorly understood. And so, we tried to build machine learning methods that could learn a more complex code than we could do on our own."</p>
</div>
</div>
</div>
<div class="field__item">
<div class="paragraph paragraph--type--image-with-caption image-with-caption">
<div class="image-with-caption__col">
<div class="image-with-caption__col-row">
<div class="field field--name-field-image field--type-entity-reference field--label-hidden field__item"><article class="media media--type-image media--view-mode-pt-image-with-caption">
<div class="field field--name-field-media-image field--type-image field--label-hidden field__item"> <picture>
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1921px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1601px) and (max-width: 1920px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_desktop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=_co_xTGs 1x" media="all and (min-width: 1340px) and (max-width: 1600px)" type="image/png" width="1256" height="689">
<source srcset="/files/styles/pt_image_with_caption_laptop/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=k0o6avtr 1x" media="all and (min-width: 800px) and (max-width: 1339px)" type="image/png" width="1221" height="584">
<source srcset="/files/styles/pt_image_with_caption_tablet/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=YWqFGOKp 1x" media="all and (min-width: 540px) and (max-width: 799px)" type="image/png" width="707" height="466">
<source srcset="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT 1x" media="all and (max-width: 539px)" type="image/png" width="499" height="329">
<img loading="eager" width="499" height="329" src="/files/styles/pt_image_with_caption_phone/public/2024-10/CRE-Graphic_JAX_dimensions-buffer4_0.png?itok=na2Rt2VT" alt="/news/test-5" title="/news/test-5" typeof="foaf:Image">
</picture>
</div>
</article>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Credit: Natalie Velez, 映画传媒 Communications; and Sager Gosai</p>
</div>
</div>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--text-narrow paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Using a form of artificial intelligence (AI) called deep learning, the group trained a model using hundreds of thousands of DNA sequences from the human genome that they measured in the laboratory for CRE activity in three types of cells: blood, liver and brain. The AI model allowed the researchers to predict the activity for any sequence from the almost infinite number of possible combinations. By analyzing these predictions, the researchers discovered new patterns in the DNA, learning how the grammar of CRE sequences in the DNA impact how much RNA would be made 鈥� a proxy for how much a gene is activated. </p>
<p>The team then developed a platform called CODA (Computational Optimization of DNA Activity), which used their AI model to efficiently design thousands of completely new CREs with requested characteristics, like activating a particular gene in human liver cells but not activating the same gene in human blood or brain cells. Through an iterative combination of 鈥榳et鈥� and 鈥榙ry鈥� investigation, using experimental data to first build and then validate computational models, the researchers refined and improved the program鈥檚 ability to predict the biological impact of each CRE and enabled the design of specific CREs never before seen in nature. </p>
<p>"Natural CREs, while plentiful, represent a tiny fraction of possible genetic elements and are constrained in their function by natural selection," said study co-first author Sager Gosai, a postdoctoral fellow in Sabeti's lab. "These AI tools have immense potential for designing genetic switches that precisely tune gene expression for novel applications, such as biomanufacturing and therapeutics, that lie outside the scope of evolutionary pressures."</p>
<h2>Pick-and-choose your organ</h2>
<p>Castro, Gosai, Reilly, Sabeti, Tewhey, and their team tested the new, AI-designed synthetic CREs by adding them into cells and measuring how well they activated genes in the desired cell type, as well as how good they were at avoiding gene expression in other cells. The new CREs, they discovered, were even more cell-type-specific than naturally occurring CREs known to be associated with the cell types. </p>
<p>"The synthetic CREs semantically diverged so far from natural elements that predictions for their effectiveness seemed implausible," said Gosai. "We initially expected many of the sequences would misbehave inside living cells."</p>
<p>"It was a thrilling surprise to us just how good CODA was at designing these elements," said Castro. </p>
<p>Tewhey and his collaborators studied why the synthetic CREs were able to outperform naturally occurring CREs and discovered that the cell-specific synthetic CREs contained combinations of sequences responsible for expressing genes in the target cell types, as well as sequences that repressed or turned off the gene in the other cell types.</p>
<p>Finally, the group tested several of the synthetic CRE sequences in zebrafish and mice, with good results. One CRE, for instance, was able to activate a fluorescent protein in developing zebrafish livers but not in any other areas of the fish. </p>
<p>"This technology paves the way toward the writing of new regulatory elements with pre-defined functions," said Tewhey. "Such tools will be valuable for basic research but also could have significant biomedical implications where you could use these elements to control gene expression in very specific cell types for therapeutic purposes."</p>
<p><em>Adapted from <a href="https://www.jax.org/news-and-insights/2024/october/researchers-flip-genes-on-and-off-with-ai-designed-dna-switches" target="_blank">a press release</a> issued jointly with The Jackson Laboratory.</em></p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro paragraph--view-mode--default">
<div class="field field--name-field-paragraph field--type-entity-reference-revisions field--label-hidden field__items">
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Funding</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Support for this study came from the National Human Genome Research Institute. Pardis Sabeti is an Investigator with the Howard Hughes Medical Institute.</p>
</div>
</div>
</div>
<div class="field__item"> <div class="paragraph paragraph--type--table-outro-row paragraph--view-mode--default">
<div class="clearfix text-formatted field field--name-field-heading field--type-text field--label-hidden field__item"><p>Paper cited</p>
</div>
<div class="clearfix text-formatted field field--name-field-text field--type-text-long field--label-hidden field__item"><p>Gosai SJ, Castro RI, et al. <a href="https://www.nature.com/articles/s41586-024-08070-z" target="_blank">Machine-guided design of cell type-specific cis-regulatory elements</a>. <em>Nature</em>. Online October 23, 2024. DOI: 10.1038/s41586-024-08070-z.</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="content-section container">
<div class="content-section__main">
<div class="block-node-broad-tags block block-layout-builder block-field-blocknodelong-storyfield-broad-tags">
<div class="block-node-broad-tags__row">
<div class="block-node-broad-tags__title">Tags:</div>
<div class="field field--name-field-broad-tags field--type-entity-reference field--label-hidden field__items">
<div class="field__item"><a href="/broad-tags/genome-regulation" hreflang="en">Genome regulation</a></div>
<div class="field__item"><a href="/broad-tags/machine-learning-0" hreflang="en">Machine Learning</a></div>
<div class="field__item"><a href="/broad-tags/pardis-sabeti" hreflang="en">Pardis Sabeti</a></div>
</div>
</div>
</div>
</div>
</div>
Thu, 03 Mar 2022 16:06:55 +0000adicorat1127651 at