OptiPharm: An evolutionary algorithm to compare shape similarity.

Area of Science:

• Computational Chemistry
• Drug Discovery
• Bioinformatics

Background:

• Virtual Screening (VS) accelerates drug discovery by analyzing large compound databases.
• Current Shape Similarity Methods (SSMs) often use local optimization, risking missed compounds or false positives.
• Enormous database sizes necessitate efficient and exhaustive search strategies.

Purpose of the Study:

• To introduce efficient global optimization techniques for improving VS quality.
• To present OptiPharm, a parameterizable metaheuristic designed to enhance prediction accuracy and computational performance.
• To address the limitations of local optimizers in current VS approaches.

Main Methods:

• Development of OptiPharm, a novel parameterizable metaheuristic.
• Implementation of mechanisms for balancing exploration and exploitation in the search space.
• Comparison of OptiPharm's performance against WEGA, a Gaussian-based shape similarity method.

Main Results:

• OptiPharm demonstrates improved prediction accuracy compared to WEGA.
• OptiPharm offers greater computational performance, reducing search time.
• The metaheuristic effectively identifies high-quality solutions by avoiding non-promising areas.

Conclusions:

• OptiPharm represents a significant advancement in virtual screening methodologies.
• Efficient global optimization enhances the reliability and speed of drug discovery pipelines.
• OptiPharm provides a robust solution for navigating large chemical spaces in drug discovery.