Commonwealth School student Moe ’23 stands in front of the school's Commonwealth Avenue entrance.
Research with a Purpose: Moe ’23 at the Broad Institute

Moe ’23 knew he wanted to spend the summer before his senior year working on a "big project"—something that would allow him to collaborate with a team toward a meaningful goal. The Summer Scholar Program at the Broad Institute in Cambridge, Massachusetts, fit the bill, marrying his long-standing interests in genetics and computer science, particularly a desire to do more than just write "algorithms for algorithms' sake," Moe says. "I wanted to find a way to take some of what I've learned in computer science and see how it was applied in a lab setting." He was familiar with the Broad Institute, too, mostly because of its "interesting interdisciplinary research with genetics and biology" (also because they processed COVID tests for the brunt of Massachusetts schools, including Commonwealth, over the previous academic year). 

Alongside eighteen other student scholars, Moe spent six weeks in the lab working 9:00 a.m. to 5:00 p.m., enjoying the variety baked into the program. "I might check in on some of the programs that were running the previous day, see if they got expected results," he says, "or I might talk to my mentor." He could attend morning computational analysis meetings, events hosted by the Institute, and sessions just for students in the Summer Scholar Program, designed to teach them skills like how to read scientific journal articles and network. He also spent a whole lab meeting "nitpicking a paper" as a team, preparing it for peer review and, hopefully, publication—not unlike a Commonwealth English class, where time was spent both polishing sentences and interrogating meaning. But Moe spent most of his time in the lab, "working on the project in a pretty self-directed way." 

"The project" involved "evaluating meta-imputation as a method to improve imputation in African individuals." Since whole genome sequencing is prohibitively expensive, researchers use imputation, a method of predicting portions of an individual's genome by comparing a small sample to a large reference panel of fully sequenced genomes. The performance of imputation is heavily reliant on the diversity of the individuals in the reference panel. Last March, a group of researchers proposed meta-imputation: a method of combining the results of imputation against multiple panels. For his project, Moe tested the performance of meta-imputation compared with traditional imputation on a sample of African individuals, a population that has been underrepresented in genetic research. "There was something exciting thinking about how this project affected people, or how the work that my lab did affected people," Moe says. The experience culminated in a scientific poster, which Moe created with help from four researchers in the lab, as well as a better understanding of the "kind of intersections between computer science and biology and other research fields." The altruistic nature of the work "gave a lot of people there this motivation to work harder," Moe says—more motivation, perhaps, than they might have felt working at a garden-variety software company. "I really want to be in a place where people are passionate about what they do, specifically, enthusiasts about the results," he said. 

Open minded about the future, Moe isn't ruling out computational genetics as a career. (A possible next step: exploring how systems programming, a subset of computer science, can be used to make better research tools, including for computational genetics.) "[The program] definitely made me think more intentionally about these sorts of impacts," he says, and "what a good healthy collaborative environment is." 

Other students interested in a similar lab-based internship might benefit from a similar open-mindedness, Moe says, "trying something that doesn't exactly line with what they think they want to do." Moe was "very happy" to be working in computer science, but he still "tried something a little different," he says. This approach can help clarify interest in a subject, he adds, either confirming that it's a good path or revealing that more exploration may be necessary. To get a foot in the door, Moe suggests internship hopefuls start with simply emailing lab directors (respectfully, of course). After arriving, they should get to know their labmates and try to "learn about the other possibilities out there," he said. With any luck, they'll find themselves in a lab populated with researchers like the ones at the Broad Institute.

"People were just incredibly generous and nice with their time," Moe said. One particular researcher, involved in designing automated COVID-testing platforms, was quick to invite interns to stop by his lab sometime—and quick to welcome Moe when he showed up the next day. "'Hey, you made it,'" he said, before giving Moe a tour, robots and all. "That was not the only time something like that had happened. And people were really enthusiastic and very excited about the science and the research." Forgiving, too, when Moe deployed such an enormous program to process genetic data—1.5 terabytes of RAM—that he accidentally crashed the lab's servers, which his labmates patiently pointed out. A "rite of passage for programmers who do computational research," Moe noted with a smile, the incident actually helped him realize how serious his work really was.

Meet More Commonwealth Students