Linking datasets together to understand disease

Schmidt Fellow Hilary Finucane uses her mathematical gifts to understand disease genetics.

Casey Atkins Photography
Credit: Casey Atkins Photography

Hilary Finucane has been captivated by math since her freshman year of college. When asked why, she lights up with enthusiasm: “Math just fits. There’s a rigor and a clarity to it that’s fun and satisfying.”

Today, she applies math as a toolkit to solve complex genetic problems as a Schmidt Fellow at the Broad Institute of MIT and Harvard. Her lab sits in the Stanley Center for Psychiatric Research, where she runs collaborative research projects with other labs throughout the institute. Much of her focus is on developing new computational methods to identify and map genetic changes associated with disease, and then connect that data with different types of cells and biological processes.

Finucane’s career trajectory has taken many turns over time, veering through human rights, theoretical math, and computational biology.

A theory for everything

Finucane grew up in Maryland, in a science- and music-oriented family. Through high school, she considered majoring in international relations — she started an Amnesty International student chapter as a teenager, driven by a desire to create real-world solutions to the problems created by inequality. Then as an undergrad at Harvard College, she discovered theoretical math and was instantly dazzled. By the end of her first year, she had already declared it as her major, and she spent the next few years zeroed in on math and chamber music.

“Theoretical math requires a lot of creativity, to hold these abstractions in your head and to manipulate them towards what you’re trying to prove,” she says. “And the way in which everything fits together and makes sense can be deeply rewarding.”

After earning her degree in 2009, she entered a master’s program in theoretical computer science at the Weizmann Institute of Science in Israel. The move was, for Finucane, an additional lucky confluence of professional and personal goals. “An important character in this story is my now-husband, Yakir Reshef, who I met in middle school,” she notes. Reshef had family in Israel, and the couple had wanted to move there for years.

At the Weizmann Institute, Finucane completed her thesis work on several projects, focused on coding theory and geometric group theory. After earning her degree, she spent a third year in Israel continuing to do research and applying to PhD programs back in the United States.

But something had shifted. Her love of theory began competing with her drive to see real-world impact. She wanted to bring math to bear on problems that mattered to her.

“Up to that point, theory had been so enjoyable for me — and it’s also important, because understanding math more fundamentally leads to advances that are often really hard to predict,” she says. “But I wanted to see the impact of my work a little more directly.”

In this short video, Finucane describes her love for both collaborative music and collaborative science.

From math to genetics

Finucane and Reshef returned to the United States in 2012. She turned her energy to applied math with a PhD from MIT, while he pursued an MD/PhD program at Harvard University.

After a few rotations, Reshef connected Finucane to Alkes Price — a professor in statistical genetics at the Harvard T.H. Chan School of Public Health and an associate member at the Broad Institute — who became her PhD advisor. Price’s lab was developing statistical methods for uncovering the genetic basis of human disease, and Finucane was fascinated by the problems his team was tackling.

“Statistical genetics seemed like an area where there was a real bottleneck — a need for better computational methods — where folks with my skillset could be very impactful,” she says. “Being able to think about mathematical ideas in the service of understanding disease? That was amazing.”

Changing gears from math to genetics wasn’t particularly straightforward, however. Finucane filled in some of her knowledge gaps with university courses, lab meetings, and shadowing postdocs, and still more with her own reading and tutorials, to pick up new skills in biology, statistics, machine learning, and coding. Computational biology was speedy and energizing; she never regretted leaving theoretical math.

“Hilary adapted to an applied biological setting very quickly and successfully,” remembers Price. “In many instances, she would raise points regarding mathematical derivations or explanations for certain phenomena in group meetings, and over time, I came to see that she was almost invariably correct.”

During her PhD, Finucane explored how different parts of the genome relate to activity in different cell types and disease phenotypes, and identified new ways to model those associations using data from genome-wide association studies

Ben Neale, an institute member in Broad’s Stanley Center for Psychiatric Research and associate professor in the Analytic and Translational Genetics Unit at Massachusetts General Hospital, recalls his first memory of Finucane: she was laying out a series of mathematical functions on a whiteboard during her PhD work. “The work and her explanation were so crisp and clean, and I instantly knew that she was going to be an impactful scientist,” he says. “It’s such a rare trait for someone to be able to think so clearly, and communicate sophisticated ideas with such transparency.”

Toward the end of her graduate work, Finucane was accepted into the Schmidt Fellows Program to build her own lab at the Broad Institute, which she began in 2017. The Fellows program, funded by Eric and Wendy Schmidt, provides full support for early-career scientists to establish independent research programs at the intersection of biomedical and quantitative science.

She felt strongly drawn to the Broad’s collaborative nature. “Experts in all sorts of specialties could engage with each other here, and there was an important role for me to play as a computational person in developing better approaches to data,” she says.

Turning data into insight

“Starting here at Broad, with my own lab right off the bat, was crazy — but in a really fun way,” Finucane emphasizes. “It was another learning curve. I remember thinking at the very beginning, ‘I need to have a to-do list,’ and I didn’t even know what goes on the to-do list, let alone how to do the things on the list.”

But with thoughtful selection of projects and team members, and a solid support network at Broad, including Neale and Price, Finucane has met with amazing success. She leapt into the Medical and Population Genetics program at Broad, and soon became the program’s associate director. She’s also heavily involved in an initiative at Broad called Variant-to-Function, a collective effort to systematically determine how disease-associated gene variants impact molecular, cellular, and physiological functions in organisms.

“Hilary has an uncanny sense for the field,” says Neale. “She’s one of the key developers of methods that we’ll use to advance our understanding with genetic datasets, and there is a real clarity to her thought that is meaningful and powerful. Her approach is always, ‘How do we get to the questions that really matter?’”

Now, Finucane has a seemingly endless list of research opportunities that she’s excited to pursue, and she discusses each with characteristic enthusiasm. Her team is helping experimental biologists identify and prioritize which genes and variants to study, adapting analytical methods to better interpret genomic datasets gathered from multiple populations, and much more.

“When we’re able to take all the data that we’ve been collecting for the last decade, and turn it into a deeper understanding of what causes disease — that’s success,” she says. “I want to be on the team that does that.”