A computational framework for predicting drug Mode-of-Action

The problem of large scale computational drug screening is considered in this paper. A computational framework is developed for microarray profile of drug treatment from the Connectivity Map (cMap) project. The framework is proposed to address the deficiency of cMap prediction. This scheme includes construction of drug Mode-of-Actions (MoA) and subsequently Mode-of-Action (MoA) network, or MoANet and a MoA prediction algorithm based on MoAs. The scheme is tested on the cMap data and an independent data set. The result shows significant improvement over cMap.