A comprehensive analysis of pharmacogenomic interactions in cancer reveals how 1,001 human cancer cell lines respond to 265 anti-cancer drugs, based on molecular data from 11,289 tumors. The study integrates diverse molecular data, including somatic mutations, copy number alterations, DNA methylation, and gene expression, to identify oncogenic aberrations that influence drug sensitivity. It highlights the importance of tissue lineage in drug response and demonstrates that cell lines accurately recapitulate oncogenic alterations found in tumors. Logic-based modeling and machine learning approaches are used to uncover combinations of genetic alterations that sensitize cancer cells to drugs, while also identifying the relative importance of different data types in predicting drug response. The study provides a valuable resource for identifying therapeutic options for cancer sub-populations and links genotypes with cellular phenotypes. The results show that a large panel of cell lines can capture clinically relevant genomic alterations, pathway alterations, and global signatures of driver events. The analysis also identifies known and novel gene-drug associations, demonstrating that pharmacogenomic models can predict drug sensitivity with high accuracy. The study validates these models using independent datasets, showing consistent results across different cancer types. The findings suggest that pharmacogenomic models can help in the development of patient stratification strategies for clinical trials and improve the understanding of cancer biology. The study underscores the importance of integrating diverse molecular data to uncover the complex interactions between genetic alterations and drug response in cancer.A comprehensive analysis of pharmacogenomic interactions in cancer reveals how 1,001 human cancer cell lines respond to 265 anti-cancer drugs, based on molecular data from 11,289 tumors. The study integrates diverse molecular data, including somatic mutations, copy number alterations, DNA methylation, and gene expression, to identify oncogenic aberrations that influence drug sensitivity. It highlights the importance of tissue lineage in drug response and demonstrates that cell lines accurately recapitulate oncogenic alterations found in tumors. Logic-based modeling and machine learning approaches are used to uncover combinations of genetic alterations that sensitize cancer cells to drugs, while also identifying the relative importance of different data types in predicting drug response. The study provides a valuable resource for identifying therapeutic options for cancer sub-populations and links genotypes with cellular phenotypes. The results show that a large panel of cell lines can capture clinically relevant genomic alterations, pathway alterations, and global signatures of driver events. The analysis also identifies known and novel gene-drug associations, demonstrating that pharmacogenomic models can predict drug sensitivity with high accuracy. The study validates these models using independent datasets, showing consistent results across different cancer types. The findings suggest that pharmacogenomic models can help in the development of patient stratification strategies for clinical trials and improve the understanding of cancer biology. The study underscores the importance of integrating diverse molecular data to uncover the complex interactions between genetic alterations and drug response in cancer.