In this dissertation, we describe the cheminformatics analysis of diverse chemogenomics data sources as well as the application of these data to several drug discovery efforts. In Chapter 1, we describe the discovery and characterization of novel Ebola virus inhibitors through QSAR-based virtual screening. In Chapter 2, we report the discovery and analysis of a series of potent and selective doublecortin-like kinase 1 (DCLK1) inhibitors using QSAR modeling, virtual screening, Matched Molecular Pair Analysis (MMPA), and molecular docking. In Chapter 3, we performed a large-scale analysis of publicly available data in PubChem to probe the reliability and applicability of Pan-Assay INterference compoundS (PAINS) alerts, a popular computational drug screening tool. In Chapter 4, we explore the PubMed database as a novel source of biomedical data and describe the development of Chemotext, a publicly available web server capable of text-mining the published literature.