Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions
Nowadays, the problem of the model’s applicability domain (AD) definition is an active research topic in chemoinformatics. Although many various AD definitions for the models predicting properties of molecules (Quantitative Structure-Activity/Property Relationship (QSAR/QSPR) models) were described...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | International Journal of Molecular Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1422-0067/21/15/5542 |
| _version_ | 1797560469794324480 |
|---|---|
| author | Assima Rakhimbekova Timur I. Madzhidov Ramil I. Nugmanov Timur R. Gimadiev Igor I. Baskin Alexandre Varnek |
| author_facet | Assima Rakhimbekova Timur I. Madzhidov Ramil I. Nugmanov Timur R. Gimadiev Igor I. Baskin Alexandre Varnek |
| author_sort | Assima Rakhimbekova |
| collection | DOAJ |
| description | Nowadays, the problem of the model’s applicability domain (AD) definition is an active research topic in chemoinformatics. Although many various AD definitions for the models predicting properties of molecules (Quantitative Structure-Activity/Property Relationship (QSAR/QSPR) models) were described in the literature, no one for chemical reactions (Quantitative Reaction-Property Relationships (QRPR)) has been reported to date. The point is that a chemical reaction is a much more complex object than an individual molecule, and its yield, thermodynamic and kinetic characteristics depend not only on the structures of reactants and products but also on experimental conditions. The QRPR models’ performance largely depends on the way that chemical transformation is encoded. In this study, various AD definition methods extensively used in QSAR/QSPR studies of individual molecules, as well as several novel approaches suggested in this work for reactions, were benchmarked on several reaction datasets. The ability to exclude wrong reaction types, increase coverage, improve the model performance and detect Y-outliers were tested. As a result, several “best” AD definitions for the QRPR models predicting reaction characteristics have been revealed and tested on a previously published external dataset with a clear AD definition problem. |
| first_indexed | 2024-03-10T18:01:00Z |
| format | Article |
| id | doaj.art-680f30657de04cb7bcbec989d52edcd6 |
| institution | Directory Open Access Journal |
| issn | 1661-6596 1422-0067 |
| language | English |
| last_indexed | 2024-03-10T18:01:00Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | International Journal of Molecular Sciences |
| spelling | doaj.art-680f30657de04cb7bcbec989d52edcd62023-11-20T08:53:22ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672020-08-012115554210.3390/ijms21155542Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical ReactionsAssima Rakhimbekova0Timur I. Madzhidov1Ramil I. Nugmanov2Timur R. Gimadiev3Igor I. Baskin4Alexandre Varnek5A.M. Butlerov Institute of Chemistry, Kazan Federal University, 420008 Kazan, RussiaA.M. Butlerov Institute of Chemistry, Kazan Federal University, 420008 Kazan, RussiaA.M. Butlerov Institute of Chemistry, Kazan Federal University, 420008 Kazan, RussiaInstitute for Chemical Reaction Design and Discovery, Hokkaido University, Sapporo 001-0021, JapanA.M. Butlerov Institute of Chemistry, Kazan Federal University, 420008 Kazan, RussiaInstitute for Chemical Reaction Design and Discovery, Hokkaido University, Sapporo 001-0021, JapanNowadays, the problem of the model’s applicability domain (AD) definition is an active research topic in chemoinformatics. Although many various AD definitions for the models predicting properties of molecules (Quantitative Structure-Activity/Property Relationship (QSAR/QSPR) models) were described in the literature, no one for chemical reactions (Quantitative Reaction-Property Relationships (QRPR)) has been reported to date. The point is that a chemical reaction is a much more complex object than an individual molecule, and its yield, thermodynamic and kinetic characteristics depend not only on the structures of reactants and products but also on experimental conditions. The QRPR models’ performance largely depends on the way that chemical transformation is encoded. In this study, various AD definition methods extensively used in QSAR/QSPR studies of individual molecules, as well as several novel approaches suggested in this work for reactions, were benchmarked on several reaction datasets. The ability to exclude wrong reaction types, increase coverage, improve the model performance and detect Y-outliers were tested. As a result, several “best” AD definitions for the QRPR models predicting reaction characteristics have been revealed and tested on a previously published external dataset with a clear AD definition problem.https://www.mdpi.com/1422-0067/21/15/5542applicability domainQuantitative Reaction–Property RelationshipQSAR/QSPRchemical reactionschemoinformaticsmachine learning |
| spellingShingle | Assima Rakhimbekova Timur I. Madzhidov Ramil I. Nugmanov Timur R. Gimadiev Igor I. Baskin Alexandre Varnek Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions International Journal of Molecular Sciences applicability domain Quantitative Reaction–Property Relationship QSAR/QSPR chemical reactions chemoinformatics machine learning |
| title | Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions |
| title_full | Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions |
| title_fullStr | Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions |
| title_full_unstemmed | Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions |
| title_short | Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions |
| title_sort | comprehensive analysis of applicability domains of qspr models for chemical reactions |
| topic | applicability domain Quantitative Reaction–Property Relationship QSAR/QSPR chemical reactions chemoinformatics machine learning |
| url | https://www.mdpi.com/1422-0067/21/15/5542 |
| work_keys_str_mv | AT assimarakhimbekova comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions AT timurimadzhidov comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions AT ramilinugmanov comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions AT timurrgimadiev comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions AT igoribaskin comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions AT alexandrevarnek comprehensiveanalysisofapplicabilitydomainsofqsprmodelsforchemicalreactions |