Cardiometabolic genetics, bringing functional genetic data together, looking at diabetes genes' other effects, and more
Research Roundup: February 25, 2022
Welcome to the February 25, 2022 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.
Genetic ties to cardiometabolic diseases
Last year, the UK Biobank released whole-exome sequencing data for over 200,000 participants. In a study published in Nature Genetics, Sean Jurgens, Seung Hoan Choi, Valerie Morrill, institute member and Cardiovascular Disease Initiative director Patrick Ellinor, and colleagues plumbed those data to study rare genetic variants’ contributions to 83 different cardiometabolic diseases and traits, including heart failure, diabetes, kidney disease, and high cholesterol levels. The team found 57 significant associations — some previously known and some newly discovered. They also showed that 1 to 2.4 percent of the participants carry genetic variants linked to certain cardiometabolic diseases. Read more about the study here.
Exploring sub-Saharan Africa's ancient social networks
Genetics and archeology suggest that southern and eastern sub-Saharan Africa saw major demographic changes during the late Pleistocene and early Holocene epochs, but later shifts have blurred any picture of the region's population structure at the time. By comparing genome-wide DNA data from 34 people from as far back as 20,000 years ago, Mark Lipson (HMS), Elizabeth Sawchuk (University of Alberta), Jessica Thompson (Yale), associate member David Reich in the Program in Medical and Population Genetics, Mary Prendergast (Rice University), and colleagues found evidence of three broad, stable populations that mingled throughout southern and eastern Africa 50,000 to 20,000 years ago, with localized subgroups appearing after that. Read more in Nature, a Nature research briefing, the Harvard Gazette, and an essay in The Conversation.
Mitochondrial barcodes enhance single-cell RNA-seq
Single-cell transcriptomics can be helpful in gathering information about cell states in health and disease, but its use for studying clonal lineages has been largely limited to model systems that can accommodate artificially engineered molecular barcodes. In a recent study published in Nature Biotechnology, Tyler Miller, Caleb Lareau, Julia Verga, Cell Circuits Program associate member Peter van Galen, and colleagues utilized naturally-occurring barcodes in mitochondrial DNA combined with single-cell RNA-sequencing protocols to create a new method called MAESTER. MAESTER and its accompanying computational toolkit, maegatk, captures the mitochondrial variants from high-throughput RNA-sequencing platforms and will enable future research of clonal relationships within complex human tissues.
Complex picture of diabetes loci and metabolic outcomes
Patients with type 2 diabetes (T2D) have a variety of clinical features. In previous work on the genetic basis of this heterogeneity, associate member Miriam Udler (MGH) in the Metabolism Program and colleagues identified five clusters of T2D genetic loci associated with diabetes-related traits. Now the team has built partitioned polygenic scores (pPS) for these clusters for more than 450,000 individuals across 13 cohorts. They found the clusters had different associations with metabolic outcomes beyond T2D, such as one cluster pPS that was associated with both decreased coronary artery disease risk and reduced renal function. The findings suggest genetic pathways for T2D predispose differently to other clinical outcomes. Read more in Diabetes Care.
Shape-shifting protein "cages" help lipid droplets grow
The protein seipin helps lipid storage organelles called lipid droplets form in cells, but precisely how it does so has been unclear. Using cryo-electron microscopy and deep learning-based structural modeling, Henning Arlt (HSPH), associate members Robert Farese and Tobias Walther of the Metabolism Program, and colleagues generated a full-length model of seipin, and found that seipin proteins gather into cage-like structures that span the membrane of the endoplasmic reticulum. These cages switch back and forth between two conformations — open or closed — that facilitate different stages of lipid droplet growth. Learn more in Nature.
Epigenetic control of T cell differentiation
CD8+ T cells undergo transcriptional changes and chromatin remodeling when they bind to antigens and differentiate into effector cells. The mechanisms underlying these shifts are not well understood. In Science Immunology, Hsiao-Wei Tsao of the Cancer Program, James Kaminski, W. Nicholas Haining, Nir Yosef (Berkeley), and colleagues describe the role of the basic leucine zipper activating transcription factor-like transcription factor (Batf) in these processes. The team studied the transcriptome and epigenome of differentiating CD8+ T cells in mice and identified key transcription factors, including Irf4, Runx3, and T-bet, that cooperated with Batf. Expressing these transcription factors in fibroblasts recapitulated chromatin and transcriptional properties of CD8+ T cells.
Predicting functional variants with MACIE
Deciphering genetic variants' functional roles is an ongoing challenge in biology. Researchers have generated many databases with varying categories of information, such as chromatin structure, biochemistry, and evolutionary conservation, for specific DNA elements. Some tools exist to aggregate information from these datasets, but they are limited to summarizing different aspects of variant function with a one-dimensional rating. A team supervised by Broad associate member Xihong Lin now describes a new framework called MACIE that is capable of integrating these categories of information and producing multi-dimensional assessments of variant functionality. Read more in the American Journal of Human Genetics.