Leveraging new approach methodologies: ecotoxicological modelling of endocrine disrupting chemicals to Danio rerio through machine learning and toxicity studies.

IF 3.2 4区医学 Q1 Pharmacology, Toxicology and Pharmaceutics

Toxicology Mechanisms and Methods Pub Date : 2024-09-10 DOI:10.1080/15376516.2024.2400324

Gopal Italiya, Sangeetha Subramanian

{"title":"Leveraging new approach methodologies: ecotoxicological modelling of endocrine disrupting chemicals to Danio rerio through machine learning and toxicity studies.","authors":"Gopal Italiya, Sangeetha Subramanian","doi":"10.1080/15376516.2024.2400324","DOIUrl":null,"url":null,"abstract":"New approach methodologies (NAMs) offer information tailored to the intended application while reducing the use of animals. NAMs aim to develop quantitative structure-activity relationship (QSAR) and quantitive-Read-Across structure-activity relationship (q-RASAR) models to predict and categorize the acute toxicity of known and unknown endocrine-disrupting chemicals (EDCs) against zebrafish. EDCs are a diverse group of toxic substances that disrupt the endocrine system of humans and animals. The q-RASAR model was constructed and verified using validation metrics (R2 = 0.886 and Q2 = 0.814) which found to be more reliable model compare to QSAR model. The substructure fingerprint was well-fitted for the classification model and it was validated using 10-fold average accuracy (Q = 86.88%), specificity (Sp = 88.89%), Matthew's correlation curve (MCC = 0.621) and receiver operating characteristics (ROC = 0.828). The dataset of unknown substances revealed that phenolphthalein (Php) exhibited a significant level of toxicity based on q-RASAR model. The docking and simulation study indicated that the computationally derived important features successfully bound to the target zebrafish sex hormone binding globulin (zfSHBG). The experimental LC50 value of 0.790 mg L-1 was very close to the predicted value of 0.763 mg L-1, which provides high confidence to the developed model.","PeriodicalId":23177,"journal":{"name":"Toxicology Mechanisms and Methods","volume":" ","pages":"1-17"},"PeriodicalIF":3.2000,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Toxicology Mechanisms and Methods","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/15376516.2024.2400324","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Pharmacology, Toxicology and Pharmaceutics","Score":null,"Total":0}

引用次数: 0

Abstract

New approach methodologies (NAMs) offer information tailored to the intended application while reducing the use of animals. NAMs aim to develop quantitative structure-activity relationship (QSAR) and quantitive-Read-Across structure-activity relationship (q-RASAR) models to predict and categorize the acute toxicity of known and unknown endocrine-disrupting chemicals (EDCs) against zebrafish. EDCs are a diverse group of toxic substances that disrupt the endocrine system of humans and animals. The q-RASAR model was constructed and verified using validation metrics (R² = 0.886 and Q² = 0.814) which found to be more reliable model compare to QSAR model. The substructure fingerprint was well-fitted for the classification model and it was validated using 10-fold average accuracy (Q = 86.88%), specificity (Sp = 88.89%), Matthew's correlation curve (MCC = 0.621) and receiver operating characteristics (ROC = 0.828). The dataset of unknown substances revealed that phenolphthalein (Php) exhibited a significant level of toxicity based on q-RASAR model. The docking and simulation study indicated that the computationally derived important features successfully bound to the target zebrafish sex hormone binding globulin (zfSHBG). The experimental LC50 value of 0.790 mg L^-1 was very close to the predicted value of 0.763 mg L^-1, which provides high confidence to the developed model.

查看原文本刊更多论文

利用新方法论：通过机器学习和毒性研究，建立干扰内分泌的化学品对小白鼠的生态毒理学模型。

新出现的干扰内分泌的化学品（EDCs）是一组干扰内分泌系统的多种有毒物质。这些物质只有在获得其毒性的具体证据后才能获得批准。新方法学（NAMs）可提供适合预期应用的信息，同时减少动物的使用。新方法旨在开发定量结构-活性关系（QSAR）和定量-交叉结构-活性关系（q-RASAR）模型，以预测已知和未知 EDC 的急性毒性并对其进行分类。通过验证指标（R2 = 0.886 和 Q2 = 0.814）构建并验证了 q-RASAR 模型。亚结构指纹与分类模型拟合良好，并使用 10 倍平均准确率（Q = 86.88%）、特异性（Sp = 88.89%）、马太相关曲线（MCC = 0.621）和接收器操作特性（ROC = 0.828）对模型进行了验证。未知物质数据集显示，酚酞（Php）具有明显的毒性。对接和模拟研究结果表明，计算得出的重要特征成功地与目标斑马鱼性激素结合球蛋白（zfSHBG）结合。实验LC50值为0.790 mg L-1，与预测值0.763 mg L-1非常接近，这为所建立的模型提供了很高的可信度。

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

Toxicology Mechanisms and Methods 医学-毒理学

CiteScore

6.60

自引率

3.10%

发文量

审稿时长

6-12 weeks

期刊介绍： Toxicology Mechanisms and Methods is a peer-reviewed journal whose aim is twofold. Firstly, the journal contains original research on subjects dealing with the mechanisms by which foreign chemicals cause toxic tissue injury. Chemical substances of interest include industrial compounds, environmental pollutants, hazardous wastes, drugs, pesticides, and chemical warfare agents. The scope of the journal spans from molecular and cellular mechanisms of action to the consideration of mechanistic evidence in establishing regulatory policy. Secondly, the journal addresses aspects of the development, validation, and application of new and existing laboratory methods, techniques, and equipment. A variety of research methods are discussed, including: In vivo studies with standard and alternative species In vitro studies and alternative methodologies Molecular, biochemical, and cellular techniques Pharmacokinetics and pharmacodynamics Mathematical modeling and computer programs Forensic analyses Risk assessment Data collection and analysis.