Research Roundup: April 19, 2019

Welcome to the April 19, 2019 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.

Calculating genetic risk for obesity

A team led by associate scientist Amit Khera, associate computational biologist Mark Chaffin, and institute member, Cardiovascular Disease Initiative director, and Program in Medical and Population Genetics (MPG) co-director Sekar Kathiresan has developed a polygenic score for obesity — a quantitative tool that predicts an individual’s inherited risk for becoming overweight. The researchers report in Cell that a genetic predisposition to obesity begins to appear in early childhood and is often clearly evident by early adulthood; adults with the highest polygenic scores weighed nearly 30 pounds more on average than those with the lowest, and were 25 times as likely to be severely obese. The team hopes that the score will enable new opportunities for biological understanding and clinical interventions. Read more in a Broad news story and coverage from STAT, Associated Press, The New York Times, WBUR, and The Wall Street Journal.

CRISPR meets chemical profiling

In Nature Chemical Biology, a team led by associate member Brian Liau in the Epigenomics Program and colleagues, with collaborators in the Genetic Perturbation Platform, reports a new method to explore interactions between proteins and small molecules. Called CRISPR-suppressor scanning, the approach combines CRISPR-Cas9 edits with small molecule screens to systematically identify mutations that alter a protein’s response to drugs. The team tested this method by profiling drug-protein interactions in acute myeloid leukemia (AML), and clarified how LSD1 inhibitor drugs work in treating AML. Learn more in a Harvard news story.

Scanning the genome with IMPACT

Despite the tremendous progress made with genome-wide association studies, researchers still don’t have a complete understanding of the non-coding region of the human genome. Locating binding sites of specific proteins called transcription factors (TFs), which regulate gene expression, is a feasible approach for identifying the regulatory regions of the non-coding genome. Reporting in American Journal of Human Genetics, graduate student Tiffany Amariuta, associate member Alkes Price, institute member Soumya Raychaudhuri of MPG, and colleagues introduce IMPACT, a mathematical model that can predict binding profiles of TFs with much greater accuracy while identifying regions contributing to the variance we observe in human disease.

To treat leukemia, two drugs may be better than one

The enzyme asparaginase, which breaks down the amino acid asparagine in the bloodstream, is used as a drug to treat leukemia. Deprived of asparagine, most leukemia cells die, while healthy cells remain mostly unaffected. However, many pediatric patients develop resistance to asparaginase treatment. Using a genome-wide CRISPR/Cas9 screen and a mouse model study, associate member Alejandro Gutierrez of MPG and colleagues identified a lethal interaction between a signaling pathway and asparaginase in acute leukemias resistant to this enzyme. The findings suggest how two drugs used in combination may help overcome resistance, thus offering hope to patients with leukemia. Read more in Cancer Cell and Boston Children’s Hospital’s Vector blog.