Aditya Rao,
Graduate Student,
ITI, Khatri Lab, Stanford University

Public data repositories and other data sharing platforms have been a massive boon to researchers in the biomedical sciences. In our lab, we have developed approaches that utilize large amounts of heterogeneous public data to identify robust host-response signatures of disease. However, our current approach to integrating these data requires that case and control samples be present in all datasets, which significantly limits the amount of data that can be utilized. Therefore, we designed a novel statistical framework that uses COCONUT conormalization to integrate data from different cohorts. This framework, which we call MANATEE (Multicohort signature Analysis using AggregaTed gEne Expression), performs comparably to our previous approach, and allows us to conduct analyses that were previously impossible.