Leveraging Relational Databases for Spatial Transcriptomics

Harvard Center for Computational Biomedicine Computer Science, 2022–23

Liaison(s): Ludwig Geistlinger, Robert Gentleman, Rafael Goncalves, Tyrone Lee, Nathan Palmer, Sunil Poudel, Sam Pullman
Advisor(s): Christopher A. Stone
Students(s): Chris Couto (PM-F), Alicia Lu, Elizabeth Lucas-Foley (PM-S), Mads Mansfield

Harvard’s Center for Computational Biomedicine supports experimental labs to analyze their data computationally. This project focuses on high-resolution microscopy data, particularly images of tissue. Spatial omics aim to identify single molecules within microscopic images. These large datasets are challenging to analyze using R or Python alone. Modern relational databases directly support spatial data types and are designed for working with large data. This project explores whether database queries are a more scalable alternative for processing and analysis of cellular maps.