Structure Based Virtual Screen

In order to reduce the synthesis costs in the early stages of drug discovery, commercial purchasable compound libraries are usually screened in the industry. But the traditional CADD virtual screening method has been suffering from slow screening efficiency.Therefore, this module aims to quickly complete the virtual screening of commercially purchasable compound libraries by AI technology.

The core of this module is a deep neural network, trained by using 18 million real bioassay data to predict the affinity between targets and ligands. It surpasses traditional techniques in terms of universal positive rates and enrichment factors. Moreover, it has accomplished the screening of a billion orders of magnitude molecules within a week. The screening of tens of millions of molecules can be completed very quickly on the online platform.

Create a Task

This GIF shows the steps to create a structure based virtual screen task in MolProphet: