From multi-omics data to the cancer druggable genediscovery: a novel machine learning-based approach

Abstract

The development of targeted drugs allows precision medicine in cancer treatment and achieving optimal targeted therapies. Accurate identification of cancer drug genes is helpful to strengthen the understanding of targeted cancer therapy and promote precise cancer treatment. However, rare cancer-druggable genes have been found due to the multi-omics datas diversity and complexity. This study proposes DF-CAGE, a novel machine learning-based method for cancer-druggable gene discovery. DF-CAGE integrated the somatic mutations, copy number variants, DNA methylation, and RNA-Seq data across ~10000 TCGA profiles to identify the landscape of the cancer-druggable genes. We found that DF-CAGE discovers the commonalities of currently known cancer-druggable genes from the perspective of multi-omics data and achieved excellent performance on OncoKB, Target, and Drugbank data sets. Among the ~20,000 protein-coding genes, DF-CAGE pinpointed 465 potential cancer-druggable genes. We found that the candidate cancer druggable genes (CDG-genes) are clinically meaningful and can be divided into highly reliable, reliable, and potential gene sets. Finally, we analyzed the contribution of the omics data to the identification of druggable genes. We found that DF-CAGE reports druggable genes mainly based on the CNAs data, the gene rearrangements, and the mutation rates in the population. These findings may enlighten the study and development of new drugs in the future.