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
Benchmarking In Silico Metabolite Prediction Tools against Human Radiolabeled ADME Data for Small-Molecule Drugs

文献类型：期刊论文

作者Gao, Lei1,2; Yan, Shu2; Feng, Kaiyuan2,4; Liu, Haonan2,3; Zhang, Zihao2,4; Diao, Xingxing1,2
刊名JOURNAL OF CHEMICAL INFORMATION AND MODELING
出版日期2026-03-09
卷号66期号:5页码:2918-2928
ISSN号1549-9596
DOI10.1021/acs.jcim.5c03045
通讯作者Diao, Xingxing(xxdiao@simm.ac.cn)
英文摘要Artificial intelligence-based in silico metabolite prediction tools are increasingly used in drug development, but their performance against high-quality human data remains uncertain. We evaluated four open-access models, including SyGMa, GLORYx, BioTransformer 3.0, MetaPredictor, and MetaTrans, using 11 small-molecule drugs with published human radiolabeled ADME data. Predicted metabolites were compared with experimentally identified human metabolites, and model performance was assessed using recall, precision, balanced accuracy, F1, and Jaccard scores. Results indicated a distinct trade-off between coverage and balanced accuracy: SyGMa provided the broadest metabolic coverage but with low precision; GLORYx and BioTransformer achieved a better balance between coverage and relevance, with BioTransformer showing the best overall performance; MetaPredictor produced fewer metabolites, yet yielded competitive scores for several drugs, despite issues with SMILES interpretability. MetaTrans showed moderate overall performance but also generated incorrect metabolite predictions. None of the models captured the mercapturic acid pathway or predicted metabolite abundances. Overall, current artificial intelligence-based tools can partially reproduce human metabolic profiles but remain insufficient to replace experimental studies, serving best as early-stage screening tools. Further advances in molecular representation, pathway completeness, and data quality are needed to improve next-generation metabolite prediction models.
WOS关键词LIQUID-CHROMATOGRAPHY ; MASS-BALANCE ; ENZYMES
资助项目National Natural Science Foundation of China[82204585] ; Special Project for Research and Development in Key areas of Guangdong Province[2023B1111030004] ; National Key Research and Development Program of China[2022YFF1202600]
WOS研究方向Pharmacology & Pharmacy ; Chemistry ; Computer Science
语种英语
WOS记录号WOS:001688886600001
出版者AMER CHEMICAL SOC
源URL[http://119.78.100.183/handle/2S10ELR8/323005]  
专题中国科学院上海药物研究所
通讯作者Diao, Xingxing
作者单位1.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
2.Chinese Acad Sci, Shanghai Inst Mat Med, Shanghai 201210, Peoples R China
3.Tianjin Univ Tradit Chinese Med, Tianjin 301617, Peoples R China
4.Nanjing Univ Chinese Med, Sch Chinese Mat Med, Nanjing 210023, Peoples R China
推荐引用方式
GB/T 7714		 Gao, Lei,Yan, Shu,Feng, Kaiyuan,et al. Benchmarking In Silico Metabolite Prediction Tools against Human Radiolabeled ADME Data for Small-Molecule Drugs[J]. JOURNAL OF CHEMICAL INFORMATION AND MODELING,2026,66(5):2918-2928.
APA Gao, Lei,Yan, Shu,Feng, Kaiyuan,Liu, Haonan,Zhang, Zihao,&Diao, Xingxing.(2026).Benchmarking In Silico Metabolite Prediction Tools against Human Radiolabeled ADME Data for Small-Molecule Drugs.JOURNAL OF CHEMICAL INFORMATION AND MODELING,66(5),2918-2928.
MLA Gao, Lei,et al."Benchmarking In Silico Metabolite Prediction Tools against Human Radiolabeled ADME Data for Small-Molecule Drugs".JOURNAL OF CHEMICAL INFORMATION AND MODELING 66.5(2026):2918-2928.

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。