MERMAID: an open source automated hit-to-lead method based on deep reinforcement learning

Journal of Cheminformatics

Nobuaki Yasuo¹  Daiki Erikawa²  Masakazu Sekijima³ 

[1] Academy for Convergence of Materials and Informatics (TAC-MI), Tokyo Institute of Technology, S6–23, 2–12–1, Ookayama, Meguro-ku, Tokyo, Japan;Department of Computer Science, Tokyo Institute of Technology, 4259–J3–23, Nagatsuta-cho, Midori-ku, Yokohama, Japan;Department of Computer Science, Tokyo Institute of Technology, 4259–J3–23, Nagatsuta-cho, Midori-ku, Yokohama, Japan;Academy for Convergence of Materials and Informatics (TAC-MI), Tokyo Institute of Technology, S6–23, 2–12–1, Ookayama, Meguro-ku, Tokyo, Japan;

关键词: Molecular generation;    Lead Optimization;    Hit-to-Lead;    Monte Carlo Tree Search;    Drug Discovery;   

DOI  :  10.1186/s13321-021-00572-6

来源: Springer

【 摘 要 】

The hit-to-lead process makes the physicochemical properties of the hit molecules that show the desired type of activity obtained in the screening assay more drug-like. Deep learning-based molecular generative models are expected to contribute to the hit-to-lead process. The simplified molecular input line entry system (SMILES), which is a string of alphanumeric characters representing the chemical structure of a molecule, is one of the most commonly used representations of molecules, and molecular generative models based on SMILES have achieved significant success. However, in contrast to molecular graphs, during the process of generation, SMILES are not considered as valid SMILES. Further, it is quite difficult to generate molecules starting from a certain molecule, thus making it difficult to apply SMILES to the hit-to-lead process. In this study, we have developed a SMILES-based generative model that can be generated starting from a certain molecule. This method generates partial SMILES and inserts it into the original SMILES using Monte Carlo Tree Search and a Recurrent Neural Network. We validated our method using a molecule dataset obtained from the ZINC database and successfully generated molecules that were both well optimized for the objectives of the quantitative estimate of drug-likeness (QED) and penalized octanol-water partition coefficient (PLogP) optimization. The source code is available at https://github.com/sekijima-lab/mermaid.

【 授权许可】

CC BY