Drug discovery typically proceeds via a large number of compounds screened in low cost biochemical assays to progressively smaller number of compounds screened in bioassays that are believed to be predictive of the disease process - for example via cell-based assays to tissue based assays, and then in vivo efficacy assays. Alongside these assays to establish efficacy exist ADMET assays, and counter screens. The ChEMBL database http://www.ebi.ac.uk/chembl is a large scale resource of small molecule bioactivity data, constructed in a way that allows rapid data-mining and exploration of target and chemotype data. We have recently developed a series of heuristics to reconstruct assay cascades for a particular disease. We will present an overview of the approach, and some detailed novel insights from this analysis that may have general use in using polypharmacology and target prediction in lead optimisation.

Reconstruction and data-mining of assay cascades from literature bioactivity data

246th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.