Accounting of Receptor Flexibility in Ultra-Large Virtual Screens with VirtualFlow Using a Grey Wolf Optimization...

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Structure-based virtual screening accelerates drug candidate discovery.
• High-throughput virtual screening success depends on ligand library scale.
• Protein dynamics pose challenges for traditional static molecular docking.

Purpose of the Study:

• To extend the VirtualFlow platform for enhanced molecular docking.
• To incorporate flexible receptor docking capabilities to account for protein dynamics.
• To improve the efficiency and quality of virtual screening for drug discovery.

Main Methods:

• Developed an extension for the VirtualFlow open-source drug discovery platform.
• Integrated the Grey Wolf Optimization (GWOVina) algorithm for flexible receptor docking.
• Evaluated performance using large-scale computational resources (up to 128,000 CPUs).

Main Results:

• GWOVina demonstrated improved quality and efficiency in flexible receptor docking compared to AutoDock Vina.
• VirtualFlow with GWOVina exhibited linear scaling behavior with increasing computational power.
• The enhanced platform effectively handles protein dynamics in molecular docking.

Conclusions:

• The extended VirtualFlow platform with GWOVina offers a powerful tool for drug discovery.
• This approach is particularly valuable for targets exhibiting significant protein dynamics and flexibility.
• The findings facilitate more accurate and efficient identification of potential drug candidates.