A software tool to mine and analyze multiple large cancer pharmacogenomic datasets.

Thanks to recent advances in robotics and automation, it is now possible to test a large portfolio of anticancer drugs on panels of cell models representative of the molecular diversity of human cancers. The combination of genomic and pharmacological profiling referred to as “pharmacogenomics”, allows researchers to identify vulnerabilities in cancer cells that affect the way they respond to drugs. Although multiple research institutions have made such pharmacogenomic data available to the scientific community, it has been difficult to analyze because they are not standardized and are often stored in different locations.

Researchers in the Haibe-Kains Lab have collected and curated the largest pharmacogenomic datasets using their open-source R/Bioconductor PharmacoGx package and made these standardized pharmacogenomic data available for mining and visualization through their PharmacoDB web-application. These tools can be used to support robust biomarker discovery with the ultimate goal of developing predictors of drug response that will help clinicians in their treatment decision process, a key challenge in precision oncology.