Dog behaviors aren't specific to breed, building a melanoma model from scratch, new insights into TB granulomas, and more
By Ó³»´«Ã½ Communications
Credit: Len Rubenstein
Welcome to the May 2, 2022 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Ó³»´«Ã½ and their collaborators.
You can't judge a dog by its breed
In addition to their physical characteristics, dog breeds (which have only been defined for ~150 years) are generally thought to have specific personality traits. A study in , however, puts that idea in the doghouse. After analyzing more than 2,000 dog genomes and 18,000+ dog owner surveys from the project, Kathleen Morrill, Vertebrate Genomics Group director Elinor Karlsson, and colleagues identified 11 regions of the genome associated with behavioral traits, none of which were specific for breed. The findings suggest that the behaviors we associate with modern breeds have their roots much further back in dog evolution. Learn more in a UMass/Ó³»´«Ã½ press release, a post by Morrill on the , and coverage in the , , and .
Building melanoma with CRISPR
Connecting specific mutations to phenotypes in cancer is challenging in part because every tumor can have many mutations. Now, Eran Hodis, Elena Torlai Triglia, institute members (on leave) Levi Garraway and Aviv Regev, and colleagues have used CRISPR to introduce cancer-associated mutations into healthy human skin cells, one by one, building cell models that connect mutations to hallmarks of melanoma (rapid growth, increased ability to invade other tissues, certain gene expression programs, etc.). The authors say this is the first time scientists have made a human cancer model using precisely controlled genetic engineering and starting from non-stem cells. Read more in , tweetorials from and , and a Ó³»´«Ã½ news story.
Correlates of tuberculosis control
Multicellular structures called granulomas are found in Mycobacterium tuberculosis-infected lungs and lymph nodes and are key sites of host-pathogen interactions. Travis Hughes, Constantine Tzouanas, Sarah Fortune, institute member Alex Shalek, and collaborators used PET and CT imaging and single-cell RNA sequencing to study immune cell activity and bacterial clearance in granulomas from cynomolgus macaques. They found that granulomas that control M. tuberculosis growth had higher levels of particular T cell subtypes engaged in pro-inflammatory signaling networks. They also found an underappreciated role for type 2 immunity in bacterial persistence. The results suggest potential host immune targets for vaccine and therapeutics development. Read more in .
Shared immune pathways in heart and lung disease
To better understand the relationships between coronary artery disease (CAD) and pneumonia and guide potential therapeutic development, a team led by Zhi Yu and associate member Pradeep Natarajan in the Program in Medical and Population Genetics performed genome-wide analyses on 450,899 participants in the UK Biobank. Zeroing in on the genes ADAMTS7 and IL6R, they found that increased ADAMTS7 expression was linked to decreased risk for CAD but increased risk for pneumonia; IL6R showed the opposite. In , they report that these findings provide new insights into ways to reduce CAD risk via immune modulation, and emphasize the need for careful prioritization of therapeutic targets.
Prying correlation apart with PDR
Most disease-associated genetic variants affect multiple traits, perhaps through shared mechanisms, but the exact processes are often unclear. To help identify potential shared components, Jenna Ballard and Schmidt Fellow Luke O’Connor in the Program in Medical and Population Genetics developed an analysis method called pleiotropic decomposition regression (PDR), and tested it on clusters of traits genetically correlated with coronary artery disease, asthma, and type 2 diabetes. In these analyses, PDR identified components that most likely have distinct underlying genetic variants, representing different mechanisms. The results help clarify the genetic relationships between risk factors that appear highly correlated. Read more in the .