VSDMIP 1.5: an automated structure- and ligand-based virtual screening platform with a PyMOL graphical user interface

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• The VSDMIP platform previously offered structure-based virtual screening (SBVS).
• Existing tools often require known receptor 3D structures, limiting applicability.
• A need exists for flexible virtual screening approaches.

Purpose of the Study:

• To introduce a graphical user interface (GUI) plugin for the VSDMIP platform.
• To implement a ligand-based virtual screening (LBVS) module.
• To compare the performance of LBVS and SBVS and evaluate combined approaches.

Main Methods:

• Developed a PyMOL plugin for VSDMIP.
• Implemented an LBVS module to complement the existing SBVS module.
• Tested LBVS and SBVS on the Directory of Useful Decoys database using a cluster of ~100 processors.

Main Results:

• LBVS demonstrated superior performance compared to SBVS, confirming previous findings.
• Combining LBVS and SBVS enabled complete virtual screening of millions of molecules in under a month.
• VSDMIP processes are highly scalable, allowing for reduced screening times with more computational power.

Conclusions:

• The updated VSDMIP platform with its GUI and LBVS module enhances virtual screening capabilities.
• LBVS is a powerful tool, especially when receptor structures are unknown or for pre-filtering.
• The VSDMIP platform offers an efficient and scalable solution for large-scale virtual screening in drug discovery.