ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning
Abstract Despite the significant potential of generative models, low synthesizability of many generated molecules limits their real-world applications. In response to this issue, we develop ClickGen, a deep learning model that utilizes modular reactions like click chemistry to assemble molecules and...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-54456-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849221103228026880 |
|---|---|
| author | Mingyang Wang Shuai Li Jike Wang Odin Zhang Hongyan Du Dejun Jiang Zhenxing Wu Yafeng Deng Yu Kang Peichen Pan Dan Li Xiaorui Wang Xiaojun Yao Tingjun Hou Chang-Yu Hsieh |
| author_facet | Mingyang Wang Shuai Li Jike Wang Odin Zhang Hongyan Du Dejun Jiang Zhenxing Wu Yafeng Deng Yu Kang Peichen Pan Dan Li Xiaorui Wang Xiaojun Yao Tingjun Hou Chang-Yu Hsieh |
| author_sort | Mingyang Wang |
| collection | DOAJ |
| description | Abstract Despite the significant potential of generative models, low synthesizability of many generated molecules limits their real-world applications. In response to this issue, we develop ClickGen, a deep learning model that utilizes modular reactions like click chemistry to assemble molecules and incorporates reinforcement learning along with inpainting technique to ensure that the proposed molecules display high diversity, novelty and strong binding tendency. ClickGen demonstrates superior performance over the other reaction-based generative models in terms of novelty, synthesizability, and docking conformation similarity for existing binders targeting the three proteins. We then proceeded to conduct wet-lab validation on the ClickGen’s proposed molecules for poly adenosine diphosphate-ribose polymerase 1. Due to the guaranteed high synthesizability and model-generated synthetic routes for reference, we successfully produced and tested the bioactivity of these novel compounds in just 20 days, much faster than typically expected time frame when handling sufficiently novel molecules. In bioactivity assays, two lead compounds demonstrated superior anti-proliferative efficacy against cancer cell lines, low toxicity, and nanomolar-level inhibitory activity to PARP1. We demonstrate that ClickGen and related models may represent a new paradigm in molecular generation, bringing AI-driven, automated experimentation and closed-loop molecular design closer to realization. |
| format | Article |
| id | doaj-art-b7cf1fc20675427e97e37e97f45a536d |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-b7cf1fc20675427e97e37e97f45a536d2024-11-24T12:33:55ZengNature PortfolioNature Communications2041-17232024-11-0115111810.1038/s41467-024-54456-yClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learningMingyang Wang0Shuai Li1Jike Wang2Odin Zhang3Hongyan Du4Dejun Jiang5Zhenxing Wu6Yafeng Deng7Yu Kang8Peichen Pan9Dan Li10Xiaorui Wang11Xiaojun Yao12Tingjun Hou13Chang-Yu Hsieh14College of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityInstitute of Traditional Chinese Medicine, Chengde Medical University, ChengdeCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityInstitute of Traditional Chinese Medicine, Chengde Medical University, ChengdeCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityDr. Neher’s Biophysics Laboratory for Innovative Drug Discovery, State Key Laboratory of Quality Research in Chinese Medicine, Macau Institute for Applied Research in Medicine and Health, Macau University of Science and TechnologyFaculty of Applied Sciences, Macao Polytechnic UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityCollege of Pharmaceutical Sciences and Cancer Center, Zhejiang UniversityAbstract Despite the significant potential of generative models, low synthesizability of many generated molecules limits their real-world applications. In response to this issue, we develop ClickGen, a deep learning model that utilizes modular reactions like click chemistry to assemble molecules and incorporates reinforcement learning along with inpainting technique to ensure that the proposed molecules display high diversity, novelty and strong binding tendency. ClickGen demonstrates superior performance over the other reaction-based generative models in terms of novelty, synthesizability, and docking conformation similarity for existing binders targeting the three proteins. We then proceeded to conduct wet-lab validation on the ClickGen’s proposed molecules for poly adenosine diphosphate-ribose polymerase 1. Due to the guaranteed high synthesizability and model-generated synthetic routes for reference, we successfully produced and tested the bioactivity of these novel compounds in just 20 days, much faster than typically expected time frame when handling sufficiently novel molecules. In bioactivity assays, two lead compounds demonstrated superior anti-proliferative efficacy against cancer cell lines, low toxicity, and nanomolar-level inhibitory activity to PARP1. We demonstrate that ClickGen and related models may represent a new paradigm in molecular generation, bringing AI-driven, automated experimentation and closed-loop molecular design closer to realization.https://doi.org/10.1038/s41467-024-54456-y |
| spellingShingle | Mingyang Wang Shuai Li Jike Wang Odin Zhang Hongyan Du Dejun Jiang Zhenxing Wu Yafeng Deng Yu Kang Peichen Pan Dan Li Xiaorui Wang Xiaojun Yao Tingjun Hou Chang-Yu Hsieh ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning Nature Communications |
| title | ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| title_full | ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| title_fullStr | ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| title_full_unstemmed | ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| title_short | ClickGen: Directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| title_sort | clickgen directed exploration of synthesizable chemical space via modular reactions and reinforcement learning |
| url | https://doi.org/10.1038/s41467-024-54456-y |
| work_keys_str_mv | AT mingyangwang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT shuaili clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT jikewang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT odinzhang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT hongyandu clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT dejunjiang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT zhenxingwu clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT yafengdeng clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT yukang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT peichenpan clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT danli clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT xiaoruiwang clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT xiaojunyao clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT tingjunhou clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning AT changyuhsieh clickgendirectedexplorationofsynthesizablechemicalspaceviamodularreactionsandreinforcementlearning |