Computational methods for designing molecules with desired pharmaceutical properties and binding affinity are essential to facilitate seamless drug development. Despite recent advances of computational approaches in molecular optimization, there are still challenges in developing molecular design algorithms enabling to explore broad chemical space by avoiding being trapped in local minima. In this study, we report GALAPAGOS, a computational tool for automatic fragment-based molecular optimization. A novel feature of GALAPAGOS is simultaneous optimization of drug-likeness and binding affinity, which is achieved by a high-resolution chemical space exploration. A fragment-based approach in genetic operators is used in GALAPAGOS to efficiently explore nascent molecules. The evolutionary trajectory analysis shows the reproducibility of the generated molecules and effective transition of molecules in chemical spaces, supporting the robustness of our algorithm. We also present the ability of GALAPAGOS in designing potential drug candidates through comprehensive benchmark evaluation.

GALAPAGOS: Fragment-based evolutionary algorithm for simultaneous optimization of drug-likeness and affinity

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.