Machine-Learning-Assisted Design of Deep Eutectic Solvents Based on Uncovered Hydrogen Bond Patterns

作     者：Usman L.Abbas Yuxuan Zhang Joseph Tapia Selim Md Jin Chen Jian Shi Qing Shao 

作者机构：Department of Chemical and Materials EngineeringUniversity of KentuckyLexingtonKY 40506USA Department of Biosystems and Agricultural EngineeringUniversity of KentuckyLexingtonKY 40506USA Institute for Biomedical InformaticsDepartment of Computer ScienceUniversity of KentuckyLexingtonKY 40506USA 

基　　金：supported by Ignite Research Collaborations(IRC) Startup funds and the UK Artificial Intelligence(AI)in Medicine Research Alliance Pilot(NCATS UL1TR001998 and NCI P30 CA177558) 

出 版 物：《Engineering》 (工程（英文）)

年 卷 期：2024年第39卷第8期

页      码：74-83页

摘      要：Non-ionic deep eutectic solvents(DESs)are non-ionic designer solvents with various applications in catalysis,extraction,carbon capture,and ***,discovering new DES candidates is challenging due to a lack of efficient tools that accurately predict DES *** search for DES relies heavily on intuition or trial-and-error processes,leading to low success rates or missed *** that hydrogen bonds(HBs)play a central role in DES formation,we aim to identify HB features that distinguish DES from non-DES systems and use them to develop machine learning(ML)models to discover new DES *** first analyze the HB properties of 38 known DES and 111 known non-DES systems using their molecular dynamics(MD)simulation *** analysis reveals that DES systems have two unique features compared to non-DES systems:The DESs have①more imbalance between the numbers of the two intra-component HBs and②more and stronger inter-component *** on these results,we develop 30 ML models using ten algorithms and three types of HB-based *** model performance is first benchmarked using the average and minimal receiver operating characteristic(ROC)-area under the curve(AUC)*** also analyze the importance of individual features in the models,and the results are consistent with the simulation-based statistical ***,we validate the models using the experimental data of 34 *** extra trees forest model outperforms the other models in the validation,with an ROC-AUC of *** work illustrates the importance of HBs in DES formation and shows the potential of ML in discovering new DESs.

主 题 词：Machine learning Deep eutectic solvents Molecular dynamics simulations Hydrogen bond Molecular design 

学科分类：08[工学] 0812[工学-测绘类] 

核心收录：

D　O　I：10.1016/j.eng.2023.10.020

馆 藏 号：203131705...