Prediction of the potency of mammalian cyclooxygenase inhibitors with ensemble proteochemometric modeling

Abstract

Cyclooxygenases (COX) are present in the body in two isoforms, namely: COX-1, constitutively expressed, and COX-2, induced in physiopathological conditions such as cancer or chronic inflammation. The inhibition of COX with non-steroideal anti-inflammatory drugs (NSAIDs) is the most widely used treatment for chronic inflammation despite the adverse effects associated to prolonged NSAIDs intake. Although selective COX-2 inhibition has been shown not to palliate all adverse effects (e.g. cardiotoxicity), there are still niche populations which can benefit from selective COX-2 inhibition. Thus, capitalizing on bioactivity data from both isoforms simultaneously would contribute to develop COX inhibitors with better safety profiles. We applied ensemble proteochemometric modeling (PCM) for the prediction of the potency of 3,228 distinct COX inhibitors on 11 mammalian cyclooxygenases. Ensemble PCM models (, and RMSEtest = 0.71) outperformed models exclusively trained on compound (, and RMSEtest = 1.09) or protein descriptors ( and RMSEtest = 1.10) on the test set. Moreover, PCM predicted COX potency for 1,086 selective and non-selective COX inhibitors with and RMSEtest = 0.76. These values are in agreement with the maximum and minimum achievable and RMSEtest values of approximately 0.68 for both metrics. Confidence intervals for individual predictions were calculated from the standard deviation of the predictions from the individual models composing the ensembles. Finally, two substructure analysis pipelines singled out chemical substructures implicated in both potency and selectivity in agreement with the literature.Graphical AbstractPrediction of uncorrelated bioactivity profiles for mammalian COX inhibitors with Ensemble Proteochemometric Modeling.Electronic supplementary materialThe online version of this article (doi:10.1186/s13321-014-0049-z) contains supplementary material, which is available to authorized users.

Active Learning for Computational Chemogenomics

“…However, the cardinality of the theoretically available space generated through the product of all possible organic compounds and all biomacromolecular targets is computationally intractable, with estimates for the number of possible pairs reaching no less than 10 65 [39]. This might serve in part as an explanation why most computational chemogenomic applications have focused on specific subfamilies of proteins and ligands [40].…”

Small Random Forest Models for Effective Chemogenomic Active Learning

“…Potential application areas of chemogenomic approaches therefore also include assessment of target selectivities, receptor deorphanising, and drug repurposing. The slow but steady increase in retro-and prospective studies using chemogenomic methodologies hints at its utility and benefit for different applications in medicinal chemistry and chemical biology [14,[19][20][21][22][23]. Nevertheless, the sheer data volume, and sparseness and complexity of the compound-protein matrix often necessitate complex chemogenomic machine learning approaches.…”

Bioalerts: a python library for the derivation of structural alerts from bioactivity and toxicity data sets

“…Additionally, the library bioalerts permits the computation of unhashed (keyed) Morgan fingerprints. The performance of unhashed and hashed Morgan fingerprints has been shown to be comparable on continuous bioactivity data sets [16] (Additional file 2). Nonetheless, building predictive models with unhashed fingeprints enables the deconvolution of the models in a chemically meaningful way [16, 25, 26], thus increasing model interpretability.…”

“… or values. The workflow proposed by Ahlberg et al [13] is followed for the derivation of structural alerts from categorical data—with the exception that to generate compound substructures circular Morgan fingerprints are used instead of a signature descriptor [14, 15]—whereas the pipeline published by Cortes-Ciriano et al [16] is followed when using continuous bioactivity data. Ahlberg and coworkers showed that their pipeline leads to comparable results to both manual derivation of structural alerts and a clique-based method, namely PAFI [17].…”

“…Thus, they appear to efficiently account for chemical aspects related to bioactivity. In addition to the derivation of structural alerts, bioalerts provides functionalities for the calculation of unhashed (keyed) Morgan fingerprints, which can be used in predictive bioactivity modelling with the advantage of allowing for a chemically meaningful deconvolution of the chemical space [16]. Finally, bioalerts provides functionalities for the easy visualization of the derived structural alerts.…”