You are here

NEWS

News / 05.8.20

Research Roundup: May 8, 2020

Len Rubenstein
Credit : Len Rubenstein
By Broad Communications

One-pot COVID test, machine learning goes to physiology class, putting anthrax to good use, and more

 

Welcome to the May 8, 2020 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.

Building a new diagnostic for COVID-19

A team led by core institute member Feng Zhang, Omar Abudayyeh, Jonathan Gootenberg, Julia Joung, and Alim Ladha has released a new SARS-CoV-2 diagnostics assay called STOP (SHERLOCK Testing in One Pot) COVID. The STOPCovid test can be run in an hour as a single-step reaction with minimal handling, advancing CRISPR-based diagnostics closer to a point-of-care tool. The team has posted the protocol and prepared reagents to share freely with collaborators around the world. Visit STOPCovid.science for the latest, and check out more in the news release and coverage in The New York Times, NPR, STAT, and Boston 25 News.

Revealing the population history of the Andes 

Due to lack of ancient DNA studies in the past, there were many unanswered questions about how large-scale societies such as the Moche, Wari, Tiwanaku, and Inca impacted the population history of the Central and South Central Andes (present-day Peru, Bolivia, and North Chile). Senior associate member David Reich of the Program in Medical and Population Genetics and collaborators assembled genome-wide data from 89 individuals from these regions over the past ~9,000 years. Reporting in Cell, they were able to throw light on the changes to the genetic landscape in the Central Andes, revealing large-scale gene flow and cosmopolitanism in the heartlands of the Tiwanaku and Inca societies.

Bridging the gap between medicine and machine learning

The intersection of medicine and machine learning (ML) has the potential to transform healthcare. However, the complexities of one field often perplex experts from the other, thus creating a need for a shared vocabulary between the two fields. Gopal Sarma, Erik Reinertsen, and the members of the Machine Learning for Cardiovascular Disease (ML4CVD) group describe in Patterns how physiology, a foundational discipline of medical training and practice with a rich quantitative history, could serve as a starting point for the development of a common language between clinicians and ML experts, thereby accelerating real-world impact and advancing diagnosis and treatment of disease.

Baby's beta cells adjust

Birth is a metabolic challenge for newborns, a transition from constant maternal nutrient supply to intermittent feeding. A team led by Ronny Helman at Harvard, Andrew Cangelosi at MIT, senior associate member David Sabatini of the Cell Circuits Program and the Whitehead Institute, and associate member Douglas Melton of Harvard Stem Cell Institute discovered that insulin-producing beta cells adapt by altering the nutrient sensitivity of the mTORC1 pathway, which controls their insulin secretion in response to nutrients. Reporting in Cell Metabolism, the researchers also manipulated the nutrient sensitivity of mTORC1 in stem-cell derived beta cells in vitro, strongly enhancing their ability to secrete insulin in response to glucose and potentially enhancing their utility in regenerative medicine applications.

Retracing yeast emergence

Candida auris is an emerging dangerous yeast that can lead to outbreaks of invasive and multidrug-resistant infections in hospitals. Through a global collaboration, a team including Nancy Chow of the CDC, postdoctoral associate Jose Muñoz in the Infectious Disease and Microbiome Program, Anastasia Litvintseva (CDC), and Fungal Genomics Group leader Christina Cuomo surveyed the genomes of 304 C. auris isolates from 19 countries. Reporting in mBio, the researchers found four predominant clades, which have distinct evolutionary histories, genome-wide patterns of variation, and clade-specific drug mutations. Molecular clock estimates showed that nearly all outbreak-causing clusters originated less than 40 years ago, suggesting that anthropogenic factors, such as increased use of azole antifungals, led to drug resistance.

Using anthrax for good

The Cancer Dependency Map project has uncovered many potential therapeutic targets in cancer. For example, roughly 10 percent of cancers are missing one copy of the gene SF3B1, leaving them vulnerable to approaches that interfere with the remaining copy.  Brenton Paolella and associate members Rameen Beroukhim, and Bradley Pentelute (representing the Cancer and Chemical Biology and Therapeutics Science programs) worked with Zeyu Lu and Nicholas Truex in Pentelute's MIT lab to develop an approach to target SF3B1 that uses parts of the anthrax toxin to deliver antisense peptide nucleic acids (PNAs) to cancer cells. Their findings, reported in ACS Chemical Biology, lay the groundwork for a generalizable method utilizing antisense PNAs as targeted cancer therapies.

GWAS straight from the heart

Identifying common genetic variants underlying heart failure has been challenging, but James Pirruccello, Krishna Aragam, and their colleagues in the Cardiovascular Disease Initiative took a new approach, focusing on imaging measures of heart structure and function. The team performed a genome-wide association study, using cardiac MRI-derived left ventricular measurements in 36,041 participants of the UK Biobank to look for genetic loci associated with key structural and functional changes in the heart that define dilated cardiomyopathy (DCM), which damages the organ's muscle tone and can lead to heart failure. The team found 45 novel genetic loci that contribute to risk for DCM. Read more in Nature Communications.

To learn more about research conducted at the Broad, visit broadinstitute.org/publications, and keep an eye on broadinstitute.org/news.