Exploring QSAR models for activity-cliff prediction

Internet publication

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, a study to explore the AC-prediction power of modern QSAR methods and its relationship to general QSAR prediction performance is lacking. 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: We observe low AC-sensitivity amongst the tested 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. Our results provide strong support for the hypothesis that indeed QSAR methods frequently fail to predict ACs. We propose twin-network training for deep learning models as a potential future pathway to increase AC-sensitivity and thus overall QSAR performance.

Authors: Dablander, M; Hanser, T; Lambiotte, R; Morris, GM

• Publication status: Accepted
• Peer review status: Reviewed (other)
• Publisher: Cornell University
• Publication date: 2023-01-31
• DOI: 10.48550/arXiv.2301.13644
• Language: English
• Keywords: deep learning; machine learning; molecular representation; physicochemical descriptors; QSAR modelling; activity cliffs; activity cliff prediction; graph isomorphism networks; binding affinity prediction; FFR; extended-connectivity fingerprints