RSGPT: a generative transformer model for retrosynthesis planning pre-trained on ten billion datapoints

Yafeng Deng, Xinda Zhao, Hanyu Sun, Yu Chen, Xiaorui Wang, Xi Xue, Liangning Li, Jianfei Song, Chang-Yu Hsieh, Tingjun Hou, Xiandao Pan, Taghrid Saad Alomar, Xiangyang Ji () and Xiaojian Wang ()
Additional contact information
Yafeng Deng: Tsinghua University
Xinda Zhao: Hangzhou Carbonsilicon AI Technology Co., Ltd
Hanyu Sun: Peking Union Medical College and Chinese Academy of Medical Sciences
Yu Chen: Hangzhou Carbonsilicon AI Technology Co., Ltd
Xiaorui Wang: Zhejiang University
Xi Xue: Peking Union Medical College and Chinese Academy of Medical Sciences
Liangning Li: Peking Union Medical College and Chinese Academy of Medical Sciences
Jianfei Song: Hangzhou Carbonsilicon AI Technology Co., Ltd
Chang-Yu Hsieh: Hangzhou Carbonsilicon AI Technology Co., Ltd
Tingjun Hou: Hangzhou Carbonsilicon AI Technology Co., Ltd
Xiandao Pan: Peking Union Medical College and Chinese Academy of Medical Sciences
Taghrid Saad Alomar: Princess Nourah bint Abdulrahman University
Xiangyang Ji: Tsinghua University
Xiaojian Wang: Hangzhou Carbonsilicon AI Technology Co., Ltd

Nature Communications, 2025, vol. 16, issue 1, 1-14

Abstract: Abstract Retrosynthesis planning is a crucial task in organic synthesis, and deep-learning methods have enhanced and accelerated this process. With the advancement of the emergence of large language models, the demand for data is rapidly increasing. However, available retrosynthesis data are limited to only millions. Therefore, we pioneer the utilization of the template-based algorithm to generate chemical reaction data, resulting in the production of over 10 billion reaction datapoints. A generative pretrained transformer model is subsequently developed for template-free retrosynthesis planning by pre-training on 10 billion generated data. Inspired by the strategies of large language models, we introduce reinforcement learning to capture the relationships among products, reactants, and templates more accurately. Experiments demonstrate that our model achieves state-of-the-art performance on the benchmark, with a Top-1 accuracy of 63.4%, substantially outperforming previous models.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-025-62308-6 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-62308-6

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-025-62308-6

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().