Supercomputer-Based Ensemble Docking Drug Discovery Pipeline with Application to Covid-19

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular dynamics simulations

Background:

• Drug discovery requires efficient methods to identify potential therapeutic compounds.
• Understanding protein dynamics is crucial for accurate drug-target interactions.
• SARS-CoV-2 proteome presents multiple targets for antiviral drug development.

Purpose of the Study:

• To develop and validate a supercomputer-driven pipeline for in silico drug discovery.
• To integrate enhanced sampling molecular dynamics (MD) with ensemble docking for improved accuracy.
• To assess the pipeline's efficacy against SARS-CoV-2 targets.

Main Methods:

• Utilized temperature replica exchange enhanced sampling molecular dynamics (MD) on the Summit supercomputer.
• Employed ensemble docking, using MD-derived protein conformations for compound screening.
• Performed large-scale virtual screening of repurposing databases against 24 SARS-CoV-2 systems.

Main Results:

• Achieved over 1 ms of enhanced sampling MD per day.
• Demonstrated high hit rates for predicted compounds when compared to experimental data.
• Successfully docked one billion compounds in under 24 hours using Autodock-GPU.

Conclusions:

• The presented pipeline effectively leverages supercomputing for rapid and accurate in silico drug discovery.
• Ensemble docking, informed by MD, significantly enhances the identification of potential drug candidates.
• Future improvements include integrating quantum mechanics, machine learning, and AI for further pipeline optimization.