Stanley Center scientists worked with Broad Clinical Labs (pictured) to develop a low-cost, high-quality sequencing approach that is helping reveal new biological insights.
Photo Credit: Kyle Klein
Scientific Frontline: Extended "At a Glance" Summary: Blended Genome Exome (BGE) Sequencing
The Core Concept: Blended Genome Exome (BGE) is a high-quality, cost-effective genetic sequencing methodology that simultaneously captures both deep exome coverage and broad whole-genome variation in a single machine run.
Key Distinction/Mechanism: Unlike traditional deep whole-genome sequencing or limited genotyping arrays, BGE utilizes two complementary genomic scans simultaneously. It performs a deep-coverage scan of the protein-coding exome to identify rare, high-impact mutations, alongside a lighter scan of the entire genome to capture common genetic variants and structural variations (such as missing or extra DNA). This single-run process instantly synchronizes the data and reduces sequencing costs by approximately 75 percent compared to deep whole-genome sequencing.
Major Frameworks/Components:
- Deep Exome Sequencing: Targeted, high-resolution scanning of protein-coding genomic regions to detect rare mutations.
- Light Whole-Genome Sequencing: Broad genomic scanning designed to identify common genetic variants and structural anomalies.
- Single-Run Synchronization: The simultaneous generation of genome and exome data within one sequencing run, which eliminates the bioinformatic challenges of merging separately generated datasets.
Branch of Science: Genomics, Medical Genetics, Psychiatric Genetics, and Bioinformatics.
Future Application: BGE is actively enabling the development of highly accurate polygenic risk scores and cost-effective clinical diagnostic testing for conditions ranging from severe mental illness to prostate cancer. Furthermore, it is positioned to vastly expand genomics research within ancestrally diverse and historically underrepresented populations.
Why It Matters: By dramatically lowering the financial barrier to comprehensive sequencing, BGE allows researchers to conduct massive-scale genetic studies involving hundreds of thousands of participants. This scale is critical for uncovering the heritable biological roots of complex diseases, identifying novel genetic loci, and ensuring global genomic datasets accurately reflect human diversity.
For researchers on the hunt for the genetic roots of disease, the cost of deep whole-genome sequencing makes it challenging to conduct large genetic studies involving thousands of participants, which are needed to reveal new genetic insights. Therefore, scientists at the Broad Institute came up with a clever approach, called the Blended Genome Exome (BGE), that lowers the cost of sequencing by 75% and is becoming one of the most commonly used sequencing methods at the Broad.
Now, those researchers have published the first scientific paper describing BGE’s utility, showing that the method generates high-quality data at a much lower cost than the current gold standard of deep whole-genome sequencing. The study, in Nature Genetics, is from scientists in the Stanley Center for Psychiatric Research at the Broad and Broad Clinical Labs.
“In this study, we’ve shown that the BGE technology works and it works at scale, and now the entire field can benefit from the method,” said co-senior author Alicia Martin, a Broad associate member, an assistant investigator in the Analytic and Translational Genetics Unit of Massachusetts General Hospital, and an assistant professor at Harvard Medical School.
Many researchers have already benefited from BGE. Since Broad Clinical Labs first began offering BGE in late 2022, it has used the low-cost method to sequence more than 400,000 human DNA samples from dozens of research studies. In 2025 alone, Broad Clinical Labs processed nearly 123,000 samples using the BGE method, representing 30% of all genomic specimens processed that year.
The method allows scientists to conduct larger studies, to better survey variation in ancestrally diverse populations, and to accelerate new genetic discoveries regarding the roots of human disease. A version of the method designed for clinical use is now enabling low-cost genetic testing for patients, including those at risk for prostate cancer.
Balancing the Blend
The idea for BGE was born from a need to cost-effectively analyze thousands of genomes. “In the Stanley Center, we want to identify the heritable basis of severe mental illnesses, and doing so requires very large sample sizes,” said Martin, who led the new work along with Broad core faculty member Ben Neale and Dan Howrigan, a group leader in the Stanley Center. “To reach the scale that we need, with a fixed budget, we need to be able to ideally capture as much of the genome as we can, but at the lowest cost possible.”
BGE takes two complementary scans of the genome that together give a full view of how it varies among people. One scan includes deep coverage of the exome, or the protein-coding parts of the genome that tend to harbor rare, high-impact mutations. The other, lighter scan looks across the entire genome, capturing the many common genetic variants that influence traits and risk for common diseases.
In the new study, the researchers describe how they developed BGE and demonstrate its utility by applying it to more than 53,000 samples from the Populations Underrepresented in Mental Illness Associations Studies (PUMAS) project, which includes people from African, African American, and Latin American populations.
They found that BGE measured far more variants than is possible with genotyping arrays, which read only specific locations in the genome. Furthermore, BGE accomplished this at roughly a quarter of the cost of using deep-coverage whole-genome sequencing.
In addition to being lower cost and unbiased, BGE can also measure structural variants—extra or missing bits of DNA—that are known to underlie some psychiatric conditions. Another advantage of BGE is that the genome and exome data are generated in a single run on the sequencing machine, so the data are already synchronized, avoiding problems faced when combining data created separately.
By helping uncover more disease-causing variants, the method could lead to polygenic scores that better predict an individual’s risk of disease, especially for groups poorly represented in genomics research today. “Studying a more broad and diverse set of participants allows us to identify new potential biologies, find novel loci associated with severe mental illnesses, and better understand the roles of variants that we do uncover,” Martin said.
The researchers are grateful for the participants who may not only benefit one day from this work but who make it possible. “We’re incredibly appreciative of their willingness to share their DNA, particularly when many of them have some of these disorders that are pretty stigmatized in different ways around the world,” said Martin.
Funding: This work was supported in part by National Institute of Mental Health (NIMH) under the Populations Underrepresented by Mental illness Association Studies (PUMAS) grant U01MH125047 to the Broad Institute; the National Institute of Mental Health: Powering Genetic Discovery for Severe Mental Illness in Latin American and African Ancestries awarded to: The Broad Institute (U01MH125047); Harvard T.H. Chan School of Public Health (U01MH125045); University of California Los Angeles (U01MH1250452); and Rutgers University (U01MH125049); NIMH (R01115957); NICHD (R01HD081256); NIMH (R01MH113078); the Stanley Family Foundation; NIH (R01MH120642).
Published in journal: Nature Genetics
Authors: Toni A. Boltz, Benjamin B. Chu, Matthew DeFelice, Calwing Liao, Julia M. Sealock, Robert Ye, Jacqueline I. Goldstein, Lerato Majara, Jack M. Fu, Susan K. Service, Lingyu Zhan, Sarah E. Medland, Sinéad B. Chapman, Simone Rubinacci, Jonna L. Grimsby, Tamrat Abebe, Melkam Alemayehu, Fred K. Ashaba, Elizabeth G. Atkinson, Tim B. Bigdeli, Amanda B. Bradway, Harrison Brand, Lori B. Chibnik, Samuel DeLuca, Ana M. Diaz-Zuluaga, Abebaw Fekadu, Michael Gatzen, Bizu Gelaye, Stella Gichuru, Marissa L. Gildea, Toni C. Hill, Hailiang Huang, Kalyn M. Hubbard, Wilfred E. Injera, Roxanne James, Moses Joloba, Christopher Kachulis, Phillip R. Kalmbach, Rogers Kamulegeya, Gabriel Kigen, Soyeon Kim, Nastassja Koen, Edith K. Kwobah, Joseph Kyebuzibwa, Seungmo Lee, Niall J. Lennon, Penelope A. Lind, Esteban A. Lopera-Maya, Johnstone Makale, Serghei Mangul, Justin McMahon, Pierre Mowlem, Henry Musinguzi, Rehema M. Mwema, Noeline Nakasujja, Carter P. Newman, Lethukuthula L. Nkambule, Conor R. O’Neil, Ana Maria Olivares, Catherine M. Olsen, Linnet Ongeri, Sophie J. Parsa, Adele Pretorius, Shengying Qin, Raj Ramesar, Faye L. Reagan, Chiara Sabatti, Jacquelyn A. Schneider, Welelta Shiferaw, Christine Stevens, Anne Stevenson, Erik Stricker, Rocky E. Stroud II, Jessie Tang, Megan Townsend, David Whiteman, Mary T. Yohannes, Mingrui Yu, Kai Yuan, NeuroGAP-Psychosis Study, Dickens Akena, Lukoye Atwoli, Symon M. Kariuki, Karestan C. Koenen, Charles R. J. C. Newton, Dan J. Stein, Solomon Teferra, Zukiswa Zingela, Carlos N. Pato, Michele T. Pato, Carlos Lopez-Jaramillo, Nelson B. Freimer, Roel A. Ophoff, Loes M. Olde Loohuis, Michael E. Talkowski, Benjamin M. Neale, Daniel P. Howrigan, and Alicia R. Martin
Source/Credit: Broad Institute | Leah Eisenstadt
Edited by: Scientific Frontline
Reference Number: geno070826_01
.jpg)