Po-Ru Loh

The Brigham President’s Scholar Awards are philanthropically supported awards that provide unrestricted funding to early-career investigators. This is the third in an occasional series of stories featuring the 2024 President’s Scholars.

For Po-Ru Loh, PhD, of the Division of Genetics, teasing out complex genetic variants from hundreds of thousands of human genomes is a creative exercise. Genetic variants are differences in DNA sequences across individuals, and they can range from single “letter” changes to longer stretches of DNA that get inserted, deleted, or repeated in a person’s genome. The impacts of this variation are as varied as the forms it can take: a variant may have little to no impact, lead to drastic health consequences, or even be beneficial.  Loh’s statistical genetics group develops new methods of analyzing genetic data to learn more about the complicated contributions of different genetic variants to human health and disease.

“It’s very satisfying to be able to generate new tools and knowledge that will help other researchers and potentially improve the health of people who have these variants,” Loh said.

In recognition of his efforts, Loh was awarded a 2024 Brigham President’s Scholar award to support his work over the next three years.

Solving Genetic Puzzles with Statistics

As the field of genetics has advanced, researchers have figured out how to efficiently detect the most common, single-letter genetic variants in humans. This work has helped inform risk prediction and the development of interventions and treatments, but these variants represent only one class of the variation contained in the human genome. Uncovering further layers of variation holds the possibility of more insights and potential treatments.

Loh seeks to uncover more complicated genetic variants using new computational approaches. “We usually start with some idea to look for a particular type of variant,” Loh said. “Then we ask, ‘What kind of signals of that variant could we see in the sequencing data available from participants in a biobank?’”

One example is a type of variation called repeat expansions, which are short DNA segments that are repeated different numbers of times in the genomes of different people. This number can “expand” throughout life, leading to more repeats of the same segment in a person’s DNA. Too many repeats can lead to conditions such as Huntington’s disease, a rare but fatal neurodegenerative disease. One of Loh’s current projects involves teasing apart the genetic variations that affect how slowly or quickly repeats expand.

While researchers have gotten better at detecting genetic variation, the ever-increasing size of the data sets they use to develop computational methods represents a further challenge. Whereas a biobank ten years ago may have contained the partial genomes of a few hundred or thousand participants, researchers now routinely analyze data sets containing the entire genomes of several hundred thousand people. The scale of this information—which provides the potential for deeper insights into human genetics—means researchers need new methods of analyzing the data in an efficient way. Loh’s group has been working to develop such methods for several years using their statistical genetics expertise.

Another of Loh’s ongoing projects involves developing a software package to efficiently analyze genomic structural variation in data sets such as the UK Biobank, which contains the genetic information of 500,000 participants. The package will help researchers search the data set for complex structural variants and eventually map them onto disease traits and other biomarkers of health.

For Loh, solving these computational challenges does not happen in one eureka moment. Instead, it is the result of sustained statistical problem-solving. “The first 20 percent of the work is designing the methodological approach,” Loh said. “The remaining 80 percent is figuring out what went wrong and how to fix it.”

Making Connections

The support of colleagues and mentors has been critical to Loh’s success since he first began his group at the Brigham in 2017. Soumya Raychaudhuri, MD, PhD, director of the Center for Data Sciences, and Richard Maas, MD, PhD, former chief of the Division of Genetics, were instrumental in helping Loh navigate the process of setting up a lab.

Several of Loh’s recent and ongoing projects are the result of geneticists seeking to apply scalable computational methods to specific research questions. Loh’s work on structural variation and repeat expansions, for example, arose from conversations with Steven McCarroll, PhD, and Bob Handsaker, researchers at Harvard Medical School and the Broad Institute of MIT and Harvard, who saw the potential for new genetic insights from large-scale analysis of genomic repeat variation.

For the past several years, these collaborations have provided a steady stream of research questions for Loh’s team. Now, as some of these projects are approaching their final stages, Loh is interested in exploring new areas, as well as returning to past projects. His group is beginning to look at a topic from earlier in Loh’s research career: genetic variations that occur in a small subset of an individual’s cells, called mosaic variations.

“Before, we were limited by the types of data that were available,” Loh said, describing previous challenges in this area. “We could only get a cursory glimpse at these kinds of mutations, which are hard to detect because they’re only in a small fraction of cells of a person’s body.”

After several more years of progress in the field of genetics and in his own research development, though, Loh is ready to meet the challenge.

Home | President’s Scholars | Profile/Q&A