Computational prediction of compound targets is a growing discipline that takes advantage of QSAR approaches applied to large-scale global pharmacology data. For several years, multiple-category Bayesian models have been a mainstay in target prediction due to their ease of computation and application. Bayesian models trained on targets, domain, and other categories are highlighted, including successful deconvolution of targets for compounds discovered in phenotypic screens. Further application of Bayes models as virtual affinity fingerprints in similarity searching is exemplified.

Polypharmacology computational tools: Machine learning with Bayesian classifiers

246th National Meeting (2013)

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