Predicting Gene Function and Protein-Protein Interactions with RNA-Seq Co-expression Data

This presentation is by Hyojin Lee, an undergraduate student from Princeton University. Hyojin describes her summer research project with the BD2K-LINCS DCIC in the Ma’ayan Lab at the Icahn School of Medicine at Mount Sinai.

http://lincs-dcic.org/summer-research…
http://icahn.mssm.edu/labs/maayan
http://amp.pharm.mssm.edu/archs4/
https://github.com/MaayanLab/predicti…

Abstract

We have recently processed all of the RNA-seq samples that are currently available on the Gene Expression Omnibus to create the ARCHS4 digital resource. From this massive dataset, we computed gene-gene co-expression correlations across all human and mouse genes. Such adjacency matrix can be utilized to predict gene function and protein-protein interactions by exploiting the fact that genes that strongly co-express have the tendency to also share their function and to physically interact.