Automatic docking of a small number of ligands into a large number of binding sites

Very fast docking programs [1] enable new applications. In predefined workflows we start with an SDFile, filter the structures by substructure queries, followed by PASS predictions [2]. The remaining few structures are docked into 100 binding sites chosen for predicting adverse effects. The results are good indicators if a lead compound should be considered risky.