Large-scale integration of heterogeneous pharmacogenomic data for identifying drug mechanism of action

Abstract

A variety of large-scale pharmacogenomic data, such as perturbation experiments and sensitivity profiles, enable the systematical identification of drug mechanism of actions (MoAs), which is a crucial task in the era of precision medicine. However, integrating these complementary pharmacogenomic datasets is inherently challenging due to the wild heterogeneity, high-dimensionality and noisy nature of these datasets. In this work, we develop Mania, a novel method for the scalable integration of large-scale pharmacogenomic data. Mania first constructs a drug-drug similarity network through integrating multiple heterogeneous data sources, including drug sensitivity, drug chemical structure, and perturbation assays. It then learns a compact vector representation for each drug to simultaneously encode its structural and pharmacogenomic properties. Extensive experiments demonstrate that Mania achieves substantially improved performance in both MoAs and targets prediction, compared to predictions based on individual data sources as well as a state-of-the-art integrative method. Moreover, Mania identifies drugs that target frequently mutated cancer genes, which provides novel insights into drug repurposing.

Signatures of cell death and proliferation in perturbation transcriptomics data—from confounding factor to effective prediction

“…While chemical structure based similarities had the superior prediction performance (similar to (31)), it is important to mention that known MoA is mostly based on the on-target activity of drugs (which can be closely associated with chemical structure), but signature based similarity can help to identify off-targets of drugs. While our work focused on using signature, structure and sensitivity profile based similarities independently, recent works (31,49) used the fusion of different similarities to reach optimal MoA predictions. Since removing cell viability correlated genes significantly improved our MoA predictions, using it together with these advanced similarity fusion techniques could further boost MoA prediction in the future.…”

Applications of Genetics, Genomics and Bioinformatics in Drug Discovery

“…In addition, a better knowledge of MoA may allow us to reposition the existing drugs for new indications. In the study by Luo et al 11 , the authors developed a novel method, referred as Mania, for scalable data integration incorporating chemical structure, drug sensitivity and gene expression changes in response to drug treatment. Drug similarity networks were first constructed based on each of these data sources, followed by integration through Mania into a low-dimensional vector representation of each drug.…”