ChemTS: an efficient python library for de novo molecular generation

• Published in: Science and technology of advanced materials Vol. 18; no. 1; pp. 972 - 976
• Main Authors: Yang, Xiufeng, Zhang, Jinzhe, Yoshizoe, Kazuki, Terayama, Kei, Tsuda, Koji
• Format: Journal Article
• Language: English
• Published: United States Taylor & Francis 31.12.2017; Taylor & Francis Ltd; Taylor & Francis Group
• Subjects: 404 Materials informatics / Genomics; 60 New topics/Others; Algorithms; Artificial neural networks; Computer simulation; Design optimization; Fragments; Libraries; Molecular design; Monte Carlo tree search; Neural networks; New topics/Others; Organic materials; Python; python library; recurrent neural network; Recurrent neural networks; 60 New topics/Others; Molecular design; Monte Carlo tree search; recurrent neural network; python library; 404 Materials informatics / Genomics
• ISSN: 1468-6996; 1878-5514; 1878-5514
• DOI: 10.1080/14686996.2017.1401424

Automatic design of organic materials requires black-box optimization in a vast chemical space. In conventional molecular design algorithms, a molecule is built as a combination of predetermined fragments. Recently, deep neural network models such as variational autoencoders and recurrent neural networks (RNNs) are shown to be effective in de novo design of molecules without any predetermined fragments. This paper presents a novel Python library ChemTS that explores the chemical space by combining Monte Carlo tree search and an RNN. In a benchmarking problem of optimizing the octanol-water partition coefficient and synthesizability, our algorithm showed superior efficiency in finding high-scoring molecules. ChemTS is available at https://github.com/tsudalab/ChemTS .