A two-layer mono-objective algorithm based on guided optimization to reduce the computational cost in virtual...

Area of Science:

• Computational Chemistry
• Drug Discovery
• Bioinformatics

Background:

• Virtual screening is crucial for identifying potential drug candidates from large molecule databases.
• Ligand-based virtual screening methods search for molecules similar to a reference compound.
• Existing tools require substantial computational resources for efficient database exploration.

Purpose of the Study:

• To introduce 2L-GO-Pharm, a novel computational tool for ligand-based virtual screening.
• To demonstrate that 2L-GO-Pharm requires less computational effort than existing methods like OptiPharm.
• To maintain or improve the quality of screening results while reducing computational cost.

Main Methods:

• Developed 2L-GO-Pharm with reduced search space dimensions and circular limits for angular variables.
• Implemented a two-layer strategy and guided search with convergence testing for exploration-exploitation balance.
• Evaluated performance using shape similarity and electrostatic potential descriptors.

Main Results:

• 2L-GO-Pharm significantly reduces the number of evaluations needed per query molecule.
• The tool saves up to 87.5 million evaluations compared to previous methods.
• Maintained or improved the quality of identified similar molecules.

Conclusions:

• 2L-GO-Pharm offers a more computationally efficient approach to ligand-based virtual screening.
• The novel strategies employed enhance searchability and reduce computational burden.
• This tool has the potential to accelerate drug discovery by optimizing virtual screening processes.