Systematic interrogation of multiple tumor genetic backgrounds with cancer drugs enables the identification of gene mutations that enhance or suppress the activity of those drugs. This identification has a great potential for improving drug efficacy and cancer therapy by providing insights into a drug mechanism-of-action, uncovering novel cancer vulnerabilities, and revealing strategies to tackle drug resistance mechanism. Despite the great advance in the development of targeted therapy, resistance to drugs is common. The advent of clustered regularly interspaced short palindromic repeats (CRISPR) gene editing technology and its adaptation to pooled library screens in mammalian cells greatly benefited this interrogation by enabling high quality chemogenetic screens to be carried directly in human cancer cell line. And with that, allowed us to identify tumor genetic backgrounds in which current drugs are effective; thus, expanding the use of targeted therapies. Other advantage of CRISPR system coupled with the ability to perform saturating genetic perturbation screens across many cell lines is the ability to identify synergistic and suppressor interactions explaining the genetic mechanisms of acquired or innate drug resistance.Drug-gene interaction data requires specific, and appropriate analysis methods. Until this point there have been several methods associated with the analysis of drug-gene interaction experiments. However, most of these methods are built upon the modifications of the methods originally developed for the analysis of RNA-seq differential expression data, which is typically characterized by relatively high read counts across genes. The reason these approaches are not the most sophisticated ones for the analysis of drug-gene CRISPR screens is that the low read counts per gRNA are frequent in CRISPR data and are a principal attribute of genes with fitness defect, which lead to extreme deficiency in sensitivity when applied to CRISPR mediated synthetic chemogenetic interactions. To fill the need for a proper analysis method for such data, we in Hart lab have designed an algorithm1 which does not discriminate low read counts, identifies both synergistic and suppressor interactions and shows significant improvement in sensitivity when compared to the existing methods by identifying larger number of highly enriched hits in specific pathways.
The focus of my research is to employ a functional genomics approach encompassing computational modeling, quantitative analysis and chemogenetic screens to refine and improve genotype specific cancer/drug matching, expand the use of targeted therapies and investigate synergistic drugs to rescue ineffective treatments. This approach would not just attempt to heighten effectiveness of cancer treatments, but would also repurpose existing drugs, which when used in combinations could provide new therapeutic possibilities with reduced cost and time for development, while potentially minimizing side effects.