ZJU-CPS Scientists propose a new algorithm: ChemistGA (and its variant R-ChemistGA)

2022-11-01   |   药学院英文网

In September 2022, Professor Tingjun Hou and Professor Changzhu Xie's team from CPS-ZJU, Professor Xi Chen's team from Wuhan University, and Professor Dongsheng Cao's team from Central South University jointly published the paper entitled ChemistGA: A Chemical Synthesizable Accessible Molecular Generation Algorithm for Real-World Drug Discovery,  in the Journal of Medicinal Chemistry. A Synthesizable Accessible Molecular Generation Algorithm for Real-World Drug Discovery, proposing a new method for generating property-specific synthesizable molecules. In this study, the authors combine a genetic algorithm (GA) with DL to propose a new algorithm, ChemistGA (and its variant R-ChemistGA), in which DL combined with a genetic algorithm redefines hybridization in traditional genetic algorithms and employs an innovative backcrossing operation to improve the efficiency of molecule generation.

Many deep learning (DL)-based molecular generative models have been proposed to design novel molecules. These models may perform well on benchmarks, but they usually do not take real-world constraints into account, such as available training data set, synthetic accessibility, and scaffold diversity in drug discovery. In this study, a new algorithm, ChemistGA, was proposed by combining the traditional heuristic algorithm with DL, in which the crossover of the traditional genetic algorithm (GA) was redefined by DL in conjunction with GA, and an innovative backcrossing operation was implemented to generate desired molecules. Our results clearly show that ChemistGA not only retains the strength of the traditional GA but also greatly enhances the synthetic accessibility and success rate of the generated molecules with desired properties. Calculations on the two benchmarks illustrate that ChemistGA achieves impressive performance among the state-of-the-art baselines, and it opens a new avenue for the application of generative models to real-world drug discovery scenarios.

CPS-ZJU is the first author of this paper; Jike Wang, a PhD student jointly trained by CPS-ZJU and the School of Computer Science, Wuhan University, and Xiao Rui Wang, a PhD student at the School of Traditional Chinese Medicine, Macau University of Science and Technology, are co-first authors; Professors Tingjun Hou and Chang Xie of Zhejiang University, Xi Chen of Wuhan University, and Dongsheng Cao of Central South University are co-corresponding authors.

link: https://pubs.acs.org/doi/10.1021/acs.jmedchem.2c01179

Translator: Yichen Zhu

Editor: Xiao Xu