vSDC: a method to improve early recognition in virtual screening when limited experimental resources are available

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Drug design often faces challenges in identifying inhibitors for novel targets with limited experimental data.
• Virtual screening methods like molecular docking can accelerate hit discovery but require high efficiency.

Purpose of the Study:

• To develop a method to overcome the divergence in results from multiple virtual screening programs.
• To enhance the yield and reliability of virtual screening for drug discovery.

Main Methods:

• Developed and applied Standard Deviation Consensus (SDC) and variable SDC (vSDC) methods.
• Integrated results from multiple docking programs (Gold, Surflex, FlexX, Glide).
• Utilized the DUD-E benchmarking database for validation.

Main Results:

• SDC successfully identified hits for two new protein targets by testing minimal compound sets.
• vSDC outperformed individual docking programs in hit identification across 102 targets in the DUD-E database.
• vSDC demonstrated a 6-24% improvement in hit discovery compared to individual programs when testing limited molecule sets.

Conclusions:

• The vSDC method effectively addresses the unpredictability of virtual screening results.
• vSDC significantly improves hit identification rates, making it a valuable tool for academic drug discovery.
• This consensus approach enhances experimental guidance and reduces the number of compounds needing testing.