中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 上海药物研究所
  3. 中国科学院上海药物研究所
  4. 新药研究国家重点实验室
Transformer-based multitask learning for reaction prediction under low-resource circumstances

文献类型：期刊论文

作者Qiao, Haoran1; Wu, Yejian2; Zhang, Yun2; Zhang, Chengyun2; Wu, Xinyi2; Wu, Zhipeng2; Zhao, Qingjie3; Wang, Xinqiao2; Li, Huiyu1; Duan, Hongliang2,4
刊名RSC ADVANCES
出版日期2022-11-03
卷号12期号:49页码:32020-32026
DOI10.1039/d2ra05349g
通讯作者Li, Huiyu(huiyuli@shiep.edu.cn) ; Duan, Hongliang(hduan@zjut.edu.cn)
英文摘要Recently, effective and rapid deep-learning methods for predicting chemical reactions have significantly aided the research and development of organic chemistry and drug discovery. Owing to the insufficiency of related chemical reaction data, computer-assisted predictions based on low-resource chemical datasets generally have low accuracy despite the exceptional ability of deep learning in retrosynthesis and synthesis. To address this issue, we introduce two types of multitask models: retro-forward reaction prediction transformer (RFRPT) and multiforward reaction prediction transformer (MFRPT). These models integrate multitask learning with the transformer model to predict low-resource reactions in forward reaction prediction and retrosynthesis. Our results demonstrate that introducing multitask learning significantly improves the average top-1 accuracy, and the RFRPT (76.9%) and MFRPT (79.8%) outperform the transformer baseline model (69.9%). These results also demonstrate that a multitask framework can capture sufficient chemical knowledge and effectively mitigate the impact of the deficiency of low-resource data in processing reaction prediction tasks. Both RFRPT and MFRPT methods significantly improve the predictive performance of transformer models, which are powerful methods for eliminating the restriction of limited training data.
WOS关键词NEURAL-NETWORK ; MODEL ; IDENTIFICATION
资助项目National Natural Science Foundation of China ; Natural Science Foundation of Zhejiang Province ; [81903438] ; [LD22H300004]
WOS研究方向Chemistry
语种英语
WOS记录号WOS:000881825000001
出版者ROYAL SOC CHEMISTRY
源URL[http://119.78.100.183/handle/2S10ELR8/304692]  
专题新药研究国家重点实验室
通讯作者Li, Huiyu; Duan, Hongliang
作者单位1.Shanghai Univ Elect Power, Coll Math & Phys, Shanghai 200090, Peoples R China
2.Zhejiang Univ Technol, Artificial Intelligence Aided Drug Discovery Inst, Coll Pharmaceut Sci, Hangzhou 310014, Peoples R China
3.Shanghai Univ Tradit Chinese Med, Innovat Res Inst Tradit Chinese Med, Shanghai 201203, Peoples R China
4.Chinese Acad Sci, Shanghai Inst Mat Med SIMM, State Key Lab Drug Res, Shanghai 201203, Peoples R China
推荐引用方式
GB/T 7714		 Qiao, Haoran,Wu, Yejian,Zhang, Yun,et al. Transformer-based multitask learning for reaction prediction under low-resource circumstances[J]. RSC ADVANCES,2022,12(49):32020-32026.
APA Qiao, Haoran.,Wu, Yejian.,Zhang, Yun.,Zhang, Chengyun.,Wu, Xinyi.,...&Duan, Hongliang.(2022).Transformer-based multitask learning for reaction prediction under low-resource circumstances.RSC ADVANCES,12(49),32020-32026.
MLA Qiao, Haoran,et al."Transformer-based multitask learning for reaction prediction under low-resource circumstances".RSC ADVANCES 12.49(2022):32020-32026.

入库方式： OAI收割

来源：上海药物研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。