BindingDB

BindingDB is a public, web-accessible database of experimentally measured binding affinities between proteins and small molecules. It contains over 2.8 million binding data points for more than 8,000 protein targets and over 1.1 million small molecules. For researchers in drug discovery, computational chemistry, and structural biology, BindingDB is an invaluable resource because it provides the actual experimental data needed to develop and validate computational models for predicting drug-target interactions. You can search the database by protein target, chemical compound, or binding affinity range, and the results include detailed information about the experimental conditions, assay type, and source publication. The platform also offers tools for structure-activity relationship analysis and links to related resources like the Protein Data Bank and PubChem. What makes BindingDB particularly useful for the research community is that all its data is freely downloadable, making it one of the largest openly available datasets for training machine learning models in drug discovery.