Learning with uncertainty to accelerate the discovery of histone lysine-specific demethylase 1A (KDM1A/LSD1) inhibitors
文献类型:期刊论文
作者 | Wang, Dong1; Wu, Zhenxing1; Shen, Chao1; Bao, Lingjie1; Luo, Hao1; Wang, Zhe1; Yao, Hucheng2; Kong, De-Xin2; Luo, Cheng2,3; Hou, Tingjun1 |
刊名 | BRIEFINGS IN BIOINFORMATICS |
出版日期 | 2022-12-27 |
页码 | 14 |
ISSN号 | 1467-5463 |
关键词 | uncertainty estimation graph neural network KDM1A LSD1 inhibitors upper confidence bound method virtual screening |
DOI | 10.1093/bib/bbac592 |
通讯作者 | Kong, De-Xin(dxkong@mail.hzau.edu.cn) ; Luo, Cheng(cluo@simm.ac.cn) ; Hou, Tingjun(tingjunhou@zju.edu.cn) |
英文摘要 | Machine learning including modern deep learning models has been extensively used in drug design and screening. However, reliable prediction of molecular properties is still challenging when exploring out-of-domain regimes, even for deep neural networks. Therefore, it is important to understand the uncertainty of model predictions, especially when the predictions are used to guide further experiments. In this study, we explored the utility and effectiveness of evidential uncertainty in compound screening. The evidential Graphormer model was proposed for uncertainty-guided discovery of KDM1A/LSD1 inhibitors. The benchmarking results illustrated that (i) Graphormer exhibited comparative predictive power to state-of-the-art models, and (ii) evidential regression enabled well-ranked uncertainty estimates and calibrated predictions. Subsequently, we leveraged time-splitting on the curated KDM1A/LSD1 dataset to simulate out-of-distribution predictions. The retrospective virtual screening showed that the evidential uncertainties helped reduce false positives among the top-acquired compounds and thus enabled higher experimental validation rates. The trained model was then used to virtually screen an independent in-house compound set. The top 50 compounds ranked by two different ranking strategies were experimentally validated, respectively. In general, our study highlighted the importance to understand the uncertainty in prediction, which can be recognized as an interpretable dimension to model predictions. |
WOS关键词 | REVERSIBLE INHIBITORS ; APPLICABILITY DOMAIN ; DRUG DISCOVERY ; PREDICTION ; DESIGN ; QUANTIFICATION ; TUTORIAL |
资助项目 | National Key R&D Program of China[2021YFE0206400] ; National Natural Science Foundation of China[22220102001] ; Natural Science Foundation of China of Zhejiang Province[LD22H300001] ; Natural Science Foundation of China of Zhejiang Province[LZ19H300001] |
WOS研究方向 | Biochemistry & Molecular Biology ; Mathematical & Computational Biology |
语种 | 英语 |
出版者 | OXFORD UNIV PRESS |
WOS记录号 | WOS:000904614600001 |
源URL | [http://119.78.100.183/handle/2S10ELR8/304174] |
专题 | 中国科学院上海药物研究所 |
通讯作者 | Kong, De-Xin; Luo, Cheng; Hou, Tingjun |
作者单位 | 1.Zhejiang Univ, Coll Pharmaceut Sci, Hangzhou, Peoples R China 2.Huazhong Agr Univ, Coll Informat, Wuhan, Peoples R China 3.Shanghai Inst Mat Med, Shanghai, Peoples R China |
推荐引用方式 GB/T 7714 | Wang, Dong,Wu, Zhenxing,Shen, Chao,et al. Learning with uncertainty to accelerate the discovery of histone lysine-specific demethylase 1A (KDM1A/LSD1) inhibitors[J]. BRIEFINGS IN BIOINFORMATICS,2022:14. |
APA | Wang, Dong.,Wu, Zhenxing.,Shen, Chao.,Bao, Lingjie.,Luo, Hao.,...&Hou, Tingjun.(2022).Learning with uncertainty to accelerate the discovery of histone lysine-specific demethylase 1A (KDM1A/LSD1) inhibitors.BRIEFINGS IN BIOINFORMATICS,14. |
MLA | Wang, Dong,et al."Learning with uncertainty to accelerate the discovery of histone lysine-specific demethylase 1A (KDM1A/LSD1) inhibitors".BRIEFINGS IN BIOINFORMATICS (2022):14. |
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。