Chapter 9: Pharmacological and Genetic Screening of Molecularly Characterized Cell Lines

Cell lines are often used to measure drug efficacy in cancer. Recently, advances in genome-wide molecular profiling and high throughput drug screening technologies have highlighted their applications in cancer treatment, resulting in the development of multiple large pharmacogenomic datasets. These datasets contain detailed molecular profiles of large panels of cell lines and their sensitivity measures for cytotoxic and targeted therapies. The application of machine learning techniques in large scale pharmacogenomic datasets offers an opportunity to improve the prediction of response to anticancer drugs, and understand their effects on the molecular features of cells. These predictions may lead to the detection of personalized biomarkers predictive of the individualized response of patients to drugs, the discovery of new drugs, and repurposing the existing therapies or finding synergistic combinations of drugs. Computational analysis of these important datasets, however, faces adverse challenges given the variety and complexity of experimental protocols.