Studies conducted at the US National Cancer Institute (NCI) and in our laboratory show that databases including the drug sensitivities of panels of many human cancer cell lines provide valuable information on the molecular pharmacology of anticancer drugs. We established a panel of 39 cell lines of various human cancers and developed a database of their chemosensitivities. Drugs were profiled in terms of their "fingerprints", patterns of differential activity against the cell lines. There was a significant correlation between a drug's fingerprint and its mode of action, as observed in the NCI panel of 60 cell lines. Therefore our cell-line panel is a powerful tool to predict the modes of action of new compounds. We have been using this system for drug discovery, coupled with various target-based drug screenings. We used the system to identify a novel DNA minor-groove binder, MS-247, which has inhibitory activity against topoisomerases I and II, and potent in vivo antitumor activity against various human cancer xenografts. We also discovered a potent novel telomerase inhibitor, FJ5002, by mining our database with the COMPARE algorithm, followed by experimental validation. We investigated the gene expression profiles of the cell lines by using DNA microarrays to find profiles determining cellular chemosensitivity and new targets for anticancer drugs. Our integrated database, including the chemosensitivities and gene expression profiles of the cell-line panel, could provide a basis for drug discovery and personalized therapy.