Exploring QSAR models for activity-cliff prediction
Abstract Introduction and methodology Pairs of similar compounds that only differ by a small structural modification but exhibit a large difference in their binding affinity for a given target are known as activity cliffs (ACs). It has been hypothesised that QSAR models struggle to predict ACs and t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-04-01
|
| Series: | Journal of Cheminformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13321-023-00708-w |
| _version_ | 1797840859703541760 |
|---|---|
| author | Markus Dablander Thierry Hanser Renaud Lambiotte Garrett M. Morris |
| author_facet | Markus Dablander Thierry Hanser Renaud Lambiotte Garrett M. Morris |
| author_sort | Markus Dablander |
| collection | DOAJ |
| description | Abstract Introduction and methodology Pairs of similar compounds that only differ by a small structural modification but exhibit a large difference in their binding affinity for a given target are known as activity cliffs (ACs). It has been hypothesised that QSAR models struggle to predict ACs and that ACs thus form a major source of prediction error. However, the AC-prediction power of modern QSAR methods and its quantitative relationship to general QSAR-prediction performance is still underexplored. We systematically construct nine distinct QSAR models by combining three molecular representation methods (extended-connectivity fingerprints, physicochemical-descriptor vectors and graph isomorphism networks) with three regression techniques (random forests, k-nearest neighbours and multilayer perceptrons); we then use each resulting model to classify pairs of similar compounds as ACs or non-ACs and to predict the activities of individual molecules in three case studies: dopamine receptor D2, factor Xa, and SARS-CoV-2 main protease. Results and conclusions Our results provide strong support for the hypothesis that indeed QSAR models frequently fail to predict ACs. We observe low AC-sensitivity amongst the evaluated models when the activities of both compounds are unknown, but a substantial increase in AC-sensitivity when the actual activity of one of the compounds is given. Graph isomorphism features are found to be competitive with or superior to classical molecular representations for AC-classification and can thus be employed as baseline AC-prediction models or simple compound-optimisation tools. For general QSAR-prediction, however, extended-connectivity fingerprints still consistently deliver the best performance amongs the tested input representations. A potential future pathway to improve QSAR-modelling performance might be the development of techniques to increase AC-sensitivity. Graphical Abstract |
| first_indexed | 2024-04-09T16:21:37Z |
| format | Article |
| id | doaj.art-76f3ade697194633a3b30a6242ae3645 |
| institution | Directory Open Access Journal |
| issn | 1758-2946 |
| language | English |
| last_indexed | 2024-04-09T16:21:37Z |
| publishDate | 2023-04-01 |
| publisher | BMC |
| record_format | Article |
| series | Journal of Cheminformatics |
| spelling | doaj.art-76f3ade697194633a3b30a6242ae36452023-04-23T11:26:37ZengBMCJournal of Cheminformatics1758-29462023-04-0115111610.1186/s13321-023-00708-wExploring QSAR models for activity-cliff predictionMarkus Dablander0Thierry Hanser1Renaud Lambiotte2Garrett M. Morris3Mathematical Institute, University of OxfordLhasa LimitedMathematical Institute, University of OxfordDepartment of Statistics, University of OxfordAbstract Introduction and methodology Pairs of similar compounds that only differ by a small structural modification but exhibit a large difference in their binding affinity for a given target are known as activity cliffs (ACs). It has been hypothesised that QSAR models struggle to predict ACs and that ACs thus form a major source of prediction error. However, the AC-prediction power of modern QSAR methods and its quantitative relationship to general QSAR-prediction performance is still underexplored. We systematically construct nine distinct QSAR models by combining three molecular representation methods (extended-connectivity fingerprints, physicochemical-descriptor vectors and graph isomorphism networks) with three regression techniques (random forests, k-nearest neighbours and multilayer perceptrons); we then use each resulting model to classify pairs of similar compounds as ACs or non-ACs and to predict the activities of individual molecules in three case studies: dopamine receptor D2, factor Xa, and SARS-CoV-2 main protease. Results and conclusions Our results provide strong support for the hypothesis that indeed QSAR models frequently fail to predict ACs. We observe low AC-sensitivity amongst the evaluated models when the activities of both compounds are unknown, but a substantial increase in AC-sensitivity when the actual activity of one of the compounds is given. Graph isomorphism features are found to be competitive with or superior to classical molecular representations for AC-classification and can thus be employed as baseline AC-prediction models or simple compound-optimisation tools. For general QSAR-prediction, however, extended-connectivity fingerprints still consistently deliver the best performance amongs the tested input representations. A potential future pathway to improve QSAR-modelling performance might be the development of techniques to increase AC-sensitivity. Graphical Abstracthttps://doi.org/10.1186/s13321-023-00708-wQSAR modellingActivity cliffsActivity cliff predictionMachine learningDeep learningMolecular representation |
| spellingShingle | Markus Dablander Thierry Hanser Renaud Lambiotte Garrett M. Morris Exploring QSAR models for activity-cliff prediction Journal of Cheminformatics QSAR modelling Activity cliffs Activity cliff prediction Machine learning Deep learning Molecular representation |
| title | Exploring QSAR models for activity-cliff prediction |
| title_full | Exploring QSAR models for activity-cliff prediction |
| title_fullStr | Exploring QSAR models for activity-cliff prediction |
| title_full_unstemmed | Exploring QSAR models for activity-cliff prediction |
| title_short | Exploring QSAR models for activity-cliff prediction |
| title_sort | exploring qsar models for activity cliff prediction |
| topic | QSAR modelling Activity cliffs Activity cliff prediction Machine learning Deep learning Molecular representation |
| url | https://doi.org/10.1186/s13321-023-00708-w |
| work_keys_str_mv | AT markusdablander exploringqsarmodelsforactivitycliffprediction AT thierryhanser exploringqsarmodelsforactivitycliffprediction AT renaudlambiotte exploringqsarmodelsforactivitycliffprediction AT garrettmmorris exploringqsarmodelsforactivitycliffprediction |