FGTN: Fragment-based graph transformer network for predicting reproductive toxicity

IF 4.8 2区 医学 Q1 TOXICOLOGY
Jia-Nan Ren, Qiang Chen, Hong-Yu-Xiang Ye, Cheng Cao, Ya-Min Guo, Jin-Rong Yang, Hao Wang, Muhammad Zafar Irshad Khan, Jian-Zhong Chen
{"title":"FGTN: Fragment-based graph transformer network for predicting reproductive toxicity","authors":"Jia-Nan Ren,&nbsp;Qiang Chen,&nbsp;Hong-Yu-Xiang Ye,&nbsp;Cheng Cao,&nbsp;Ya-Min Guo,&nbsp;Jin-Rong Yang,&nbsp;Hao Wang,&nbsp;Muhammad Zafar Irshad Khan,&nbsp;Jian-Zhong Chen","doi":"10.1007/s00204-024-03866-4","DOIUrl":null,"url":null,"abstract":"<div><p>Reproductive toxicity is one of the important issues in chemical safety. Traditional laboratory testing methods are costly and time-consuming with raised ethical issues. Only a few in silico models have been reported to predict human reproductive toxicity, but none of them make full use of the topological information of compounds. In addition, most existing atom-based graph neural network methods focus on attributing model predictions to individual nodes or edges rather than chemically meaningful fragments or substructures. In current studies, we develop a novel fragment-based graph transformer network (FGTN) approach to generate the QSAR model of human reproductive toxicity by considering internal topological structure information of compounds. In the FGTN model, the compound is represented by a graph architecture using fragments to be nodes and bonds linking two fragments to be edges. A super molecule-level node is further proposed to connect all fragment nodes by undirected edges, obtaining global molecular features from fragment embeddings. The FGTN model achieved an accuracy (ACC) of 0.861 and an area under the receiver operating characteristic curve (AUC) value of 0.914 on nonredundant blind tests, outperforming traditional fingerprint-based machine learning models and atom-based GCN model. The FGTN model can attribute toxic predictions to fragments, generating specific structural alerts for the positive compound. Moreover, FGTN may also have the capability to distinguish various chemical isomers. We believe that FGTN can be used as a reliable and effective tool for human reproductive toxicity prediction in contribution to the advancement of chemical safety assessment.</p></div>","PeriodicalId":8329,"journal":{"name":"Archives of Toxicology","volume":null,"pages":null},"PeriodicalIF":4.8000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Archives of Toxicology","FirstCategoryId":"3","ListUrlMain":"https://link.springer.com/article/10.1007/s00204-024-03866-4","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"TOXICOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Reproductive toxicity is one of the important issues in chemical safety. Traditional laboratory testing methods are costly and time-consuming with raised ethical issues. Only a few in silico models have been reported to predict human reproductive toxicity, but none of them make full use of the topological information of compounds. In addition, most existing atom-based graph neural network methods focus on attributing model predictions to individual nodes or edges rather than chemically meaningful fragments or substructures. In current studies, we develop a novel fragment-based graph transformer network (FGTN) approach to generate the QSAR model of human reproductive toxicity by considering internal topological structure information of compounds. In the FGTN model, the compound is represented by a graph architecture using fragments to be nodes and bonds linking two fragments to be edges. A super molecule-level node is further proposed to connect all fragment nodes by undirected edges, obtaining global molecular features from fragment embeddings. The FGTN model achieved an accuracy (ACC) of 0.861 and an area under the receiver operating characteristic curve (AUC) value of 0.914 on nonredundant blind tests, outperforming traditional fingerprint-based machine learning models and atom-based GCN model. The FGTN model can attribute toxic predictions to fragments, generating specific structural alerts for the positive compound. Moreover, FGTN may also have the capability to distinguish various chemical isomers. We believe that FGTN can be used as a reliable and effective tool for human reproductive toxicity prediction in contribution to the advancement of chemical safety assessment.

Abstract Image

FGTN:用于预测生殖毒性的片段图转换器网络
生殖毒性是化学品安全的重要问题之一。传统的实验室测试方法成本高、耗时长,且存在伦理问题。据报道,目前只有少数硅学模型可以预测人类生殖毒性,但这些模型都没有充分利用化合物的拓扑信息。此外,大多数现有的基于原子的图神经网络方法侧重于将模型预测归因于单个节点或边,而不是具有化学意义的片段或子结构。在目前的研究中,我们开发了一种新颖的基于片段的图转换器网络(FGTN)方法,通过考虑化合物的内部拓扑结构信息来生成人类生殖毒性的 QSAR 模型。在 FGTN 模型中,化合物由图结构表示,以片段为节点,连接两个片段的键为边。此外,还提出了一个超级分子级节点,通过无向边连接所有片段节点,从而从片段嵌入中获取全局分子特征。在非冗余盲测中,FGTN 模型的准确度(ACC)达到了 0.861,接收者工作特征曲线下面积(AUC)达到了 0.914,优于传统的基于指纹的机器学习模型和基于原子的 GCN 模型。FGTN 模型可将毒性预测归因于片段,为阳性化合物生成特定的结构警报。此外,FGTN 还具有区分各种化学异构体的能力。我们相信,FGTN 可以作为一种可靠而有效的工具,用于人类生殖毒性预测,为推进化学品安全评估做出贡献。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Archives of Toxicology
Archives of Toxicology 医学-毒理学
CiteScore
11.60
自引率
4.90%
发文量
218
审稿时长
1.5 months
期刊介绍: Archives of Toxicology provides up-to-date information on the latest advances in toxicology. The journal places particular emphasis on studies relating to defined effects of chemicals and mechanisms of toxicity, including toxic activities at the molecular level, in humans and experimental animals. Coverage includes new insights into analysis and toxicokinetics and into forensic toxicology. Review articles of general interest to toxicologists are an additional important feature of the journal.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信