Chemogenomic approaches are gaining importance currently due to increased interest in the polypharmacology of drugs (i.e. their activity against multiple target proteins) and the ligand‐side relationships between different targets. This is not least due to the availability of increasingly larger amounts of ligand bioactivity data in the public domain; PubChem or the recent announcement to make the former Inpharmatica StARLITe database available through the EBI/EMBL being only the two of the many examples of this development. In this work we firstly summarize recent advances in the chemogenomics area. In the second part of the article we present current primary research origination from our group. This addresses various topics from the chemogenomics field such as substructure mining and chemogenomics analyses of GPCR ligands; novel feature selection methods such as two‐entropies analysis which are now translated from the bioinformatics into the chemogenomics field; as well as proteochemometric studies on subfamilies of G‐protein‐coupled receptors which not only take ligand and sequence information into account, but as a third variable are also able to consider quantitative activity values assigned to a ligand‐receptor pair. Copyright © 2009 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 2: 149–160, 2009