You are here

Moe

Moe

Moe
Commonwealth School
Boston, MA

Mentor:
Ravi Mandla
Medical and Population Genetics
 

Moe has been interested in science for as long as he can remember. His interest narrowed in on computer science, and Moe says “from there I continued to explore the ways that computers and computer programs could be applied to solve other problems.” After learning about BSSP from a list of summer programs that was sent out at his school, Moe decided to apply. As part of the program, Moe and his partner Kien worked on assessing the performance of two computational techniques that can be used to increase the accuracy of genetic studies in diverse ancestries. Genome-wide association studies (GWAS) are a research approach that identifies DNA variations associated with diseases and other conditions within a population. To perform a GWAS, DNA samples need to be collected and variations need to be identified (genotyped). Due to technological constraints, only certain regions of the DNA are genotyped per sample. The remaining regions are identified by using a special computational technique called imputation, where missing genomic regions are identified by using a reference panel. Historically, most reference panels are made with individuals with European ancestry, which decreases the accuracy of traditional imputation techniques for individuals with African ancestry. This summer, Moe worked on a new methodology called meta-imputation, that uses multiple reference panels rather than a single one, to see if it would increase the accuracy of imputation for individuals with African ancestry compared to traditional imputation methods that rely on using popular reference panels such as 1000Gp3 and TOPMed. His results showed that meta-imputation performs just as well as imputing data against TOPMed. This could be due to TOPMed already including a high number of individuals with African ancestry relative to 1000Gp3. Moe’s work will help inform researchers on whether they should consider meta-imputing their genotyping data to improve the power of their analyses. Moe enjoyed his time as a Broadie. “Everyone is very kind and collaborative as well as generous with the attention and feedback.” The societal impacts of Moe’s work inspired him to seek career opportunities with similar impact. “Because of this experience, I have decided that I want to pursue applications of computer science that also have some meaningful impact.”