Receptor pharmacophore ensemble (REPHARMBLE): a probabilistic pharmacophore modeling approach using multiple protein-ligand complexes
- 33 Downloads
Ensemble methods are gaining more importance in structure-based approaches as single protein-ligand complexes strongly influence the outcomes of virtual screening. Structure-based pharmacophore modeling based on a single protein-ligand complex with complex feature combinations is often limited to certain chemical classes. The REPHARMBLE (receptor pharmacophore ensemble) approach presented here examines the ability of an ensemble of selected protein-ligand complexes to populate pharmacophore space in the ligand binding site, rigorously assesses the importance of pharmacophore features using Poisson statistic and information theory-based entropy calculations, and generates pharmacophore models with high probabilities. In addition, an ensemble scoring function that combines all the resultant high-scoring pharmacophore models to score molecules is derived. The REPHARMBLE approach was evaluated on ten DUD-E benchmark datasets and afforded good screening performance, as measured by receiver operating characteristic, enrichment factor and Güner-Henry score. Although one of the high-scoring models achieved superior statistical results in each dataset, the ensemble scoring function balanced the shortcomings of each model and passed with close performance measures. This approach offers a reliable way of choosing the best-scoring features to build four-feature pharmacophore queries and customize a target-biased ‘pharmacophore ensemble’ scoring function for subsequent virtual screening.
KeywordsEnsemble Structure-based pharmacophore Probabilistic model Protein-ligand complex Virtual screening Entropy
SPK acknowledges the National Post-Doctoral Fellowship (PDF/2016/000156) funded by Science and Engineering Research Board (SERB), Government of India. SPK is thankful to Prof. N. Srinivasan for his valuable suggestions. The author is very thankful to anonymous reviewers for their valuable suggestions.
Compliance with ethical standards
The author ensured the compliance of ethical standards for publishing the manuscript in this journal.
Disclosure of potential conflicts of interest
The author declared no conflict of interest exists.
- 16.Schneider G, Proschak E, Grabowski K, Schneider P, Tanrikulu Y (2006) More than coloured blobs—“fuzzy” descriptions of potential pharmacophore points. Proceedings of the Beilstein Bozen symposium, molecular interactions—bringing chemistry to life, Bozen, Italy, pp 173–184Google Scholar
- 23.Webb EC (1992) Enzyme nomenclature 1992. Academic, San Diego. http://www.sbcs.qmul.ac.uk/iubmb/enzyme/. Accessed 10 January 2018
- 33.pdbtools, Harms Lab @ University of Oregon, Github repository. https://github.com/JoaoRodrigues/pdb-tools. Accessed 13 February 2018
- 42.Goksuluk D, Korkmaz S, Zararsiz G, Karaagaoglu AE (2016) easyROC: an interactive web-tool for ROC curve analysis using R language environment. R J 8:213–230Google Scholar
- 57.Mason JS, Morize I, Menard PR, Cheney DL, Hulme C, Labaudiniere RF (1999) New 4-point pharmacophore method for molecular similarity and diversity applications: overview of the method and applications, including a novel approach to the design of combinatorial libraries containing privileged substructures. J Med Chem 42:3251–3264CrossRefGoogle Scholar