SwissTargetPrediction: a web server for target prediction of bioactive small molecules

TLDR

Small molecules modulate phenotypes by binding proteins, and mapping these targets is essential for understanding bioactivity and predicting side effects, especially now that large protein–small molecule interaction datasets enable knowledge‑based target identification. SwissTargetPrediction was developed to accurately predict targets of bioactive molecules using combined 2D and 3D similarity to known ligands. The server supports predictions in five organisms and maps results across species via homology for close paralogs and orthologs. SwissTargetPrediction is freely available without login at http://www.swisstargetprediction.ch.

Abstract

Bioactive small molecules, such as drugs or metabolites, bind to proteins or other macro-molecular targets to modulate their activity, which in turn results in the observed phenotypic effects. For this reason, mapping the targets of bioactive small molecules is a key step toward unraveling the molecular mechanisms underlying their bioactivity and predicting potential side effects or cross-reactivity. Recently, large datasets of protein–small molecule interactions have become available, providing a unique source of information for the development of knowledge-based approaches to computationally identify new targets for uncharacterized molecules or secondary targets for known molecules. Here, we introduce SwissTargetPrediction, a web server to accurately predict the targets of bioactive molecules based on a combination of 2D and 3D similarity measures with known ligands. Predictions can be carried out in five different organisms, and mapping predictions by homology within and between different species is enabled for close paralogs and orthologs. SwissTargetPrediction is accessible free of charge and without login requirement at http://www.swisstargetprediction.ch.