Drug mechanism enrichment analysis improves prioritization of therapeutics for repurposing

Abstract

Motivation: Targeted therapeutics have the potential for efficacy against tumors with minimal effects on normal tissues. However, predicting effective drugs from molecular signatures remains a challenge. Here, we present Drug Mechanism Enrichment Analysis (DMEA), a method that uses a transcriptomic signature to predict drug mechanism(s) of action to which a tumor cell may be sensitive or resistant. The method derives its power by aggregating data from many drugs with a shared mechanism of action. Results: We first tested the sensitivity of DMEA using synthetic data. We next validated that DMEA recapitulated known sensitivities to HMGCR, EGFR, and RAF inhibitors while also identifying drug mechanisms for resistant cancers. Finally, we predicted tissue-dependent drug sensitivity for tumors with high and low expression of the cystine/glutamate antiporter xCT. Collectively, DMEA is a novel bioinformatic tool that uses molecular signatures to predict targeted therapeutics sharing a common mechanism of action. Availability and implementation: DMEA is freely available to download as an R package at: https://github.com/BelindaBGarana/DMEA.