Maksim Yu. Sidorov , Mikhail E. Gasanov , Artur A. Dzeranov , Lyubov S. Bondarenko , Anastasiya P. Kiryushina , Vera A. Terekhova , Gulzhian I. Dzhardimalieva , Kamila A. Kydralieva
{"title":"Machine learning-enabled prediction of ecotoxicity (EC50) of diverse organic compounds via infrared spectroscopy","authors":"Maksim Yu. Sidorov , Mikhail E. Gasanov , Artur A. Dzeranov , Lyubov S. Bondarenko , Anastasiya P. Kiryushina , Vera A. Terekhova , Gulzhian I. Dzhardimalieva , Kamila A. Kydralieva","doi":"10.1016/j.mencom.2024.10.004","DOIUrl":null,"url":null,"abstract":"<div><div>A new, less time-consuming and resource-intensive approach to predicting the EC<sub>50</sub> ecotoxicity index, which is crucial for assessing the impact of compounds on ecosystems, is proposed. Efficient EC<sub>50</sub> prediction based on infrared spectroscopy data and EC<sub>50</sub> values from the EcoTOX database is achieved using machine learning. The best results with an F1-score of 0.83 were obtained with the SVC and XGBoost models.</div></div>","PeriodicalId":18542,"journal":{"name":"Mendeleev Communications","volume":"34 6","pages":"Pages 780-782"},"PeriodicalIF":1.8000,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Mendeleev Communications","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S095994362400302X","RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
A new, less time-consuming and resource-intensive approach to predicting the EC50 ecotoxicity index, which is crucial for assessing the impact of compounds on ecosystems, is proposed. Efficient EC50 prediction based on infrared spectroscopy data and EC50 values from the EcoTOX database is achieved using machine learning. The best results with an F1-score of 0.83 were obtained with the SVC and XGBoost models.
期刊介绍:
Mendeleev Communications is the journal of the Russian Academy of Sciences, launched jointly by the Academy of Sciences of the USSR and the Royal Society of Chemistry (United Kingdom) in 1991. Starting from 1st January 2007, Elsevier is the new publishing partner of Mendeleev Communications.
Mendeleev Communications publishes short communications in chemistry. The journal primarily features papers from the Russian Federation and the other states of the former USSR. However, it also includes papers by authors from other parts of the world. Mendeleev Communications is not a translated journal, but instead is published directly in English. The International Editorial Board is composed of eminent scientists who provide advice on refereeing policy.