Pharmacophore modeling is usually considered as a special type of virtual screening without probabilistic nature. Correspondence of at least one conformation of a molecule to pharmacophore is considered as evidence of its bioactivity. We show that pharmacophores can be treated as one-class machine learning models, and the probability the reflecting model's confidence can be assigned to a pharmacophore on the basis of their precision of active compounds identification on a calibration set. Two schemes (Max and Mean) of probability calculation for consensus prediction based on individual pharmacophore models were proposed. Both approaches to some extent correspond to commonly used consensus approaches like the common hit approach or the one based on a logical OR operation uniting hit lists of individual models. Unlike some known approaches, the proposed ones can rank compounds retrieved by multiple models. These approaches were benchmarked on multiple ChEMBL datasets used for ligand-based pharmacophore modeling and externally validated on corresponding DUD-E datasets. The influence of complexity of pharmacophores and their performance on a calibration set on results of virtual screening was analyzed. It was shown that Max and Mean approaches have superior early enrichment to the commonly used approaches. Thus, a well-performing, easy-to-implement, and probabilistic alternative to existing approaches for pharmacophore-based virtual screening was proposed.

A M Butlerov Institute of Chemistry Kazan Federal University 420008 Kazan Russia

Institute of Molecular and Translational Medicine Faculty of Medicine and Dentistry Palacky University and University Hospital in Olomouc 77900 Olomouc Czech Republic

Zobrazit více v PubMed

Wermuth C.G., Ganellin C.R., Lindberg P., Mitscher L.A. Glossary of Terms Used in Medicinal Chemistry (IUPAC Recommendations 1998) Pure Appl. Chem. 1998;70:1129–1143. doi: 10.1351/pac199870051129. DOI

Vuorinen A., Schuster D. Methods for Generating and Applying Pharmacophore Models as Virtual Screening Filters and for Bioactivity Profiling. Methods. 2015;71:113–134. doi: 10.1016/j.ymeth.2014.10.013. PubMed DOI

Sliwoski G., Kothiwale S., Meiler J., Lowe E.W. Computational Methods in Drug Discovery. Pharmacol. Rev. 2014;66:334–395. doi: 10.1124/pr.112.007336. PubMed DOI PMC

Wei D., Jiang X., Zhou L., Chen J., Chen Z., He C., Yang K., Liu Y., Pei J., Lai L. Discovery of Multitarget Inhibitors by Combining Molecular Docking with Common Pharmacophore Matching. J. Med. Chem. 2008;51:7882–7888. doi: 10.1021/jm8010096. PubMed DOI

Bowman A.L., Nikolovska-Coleska Z., Zhong H., Wang S., Carlson H.A. Small Molecule Inhibitors of the MDM2-p53 Interaction Discovered by Ensemble-Based Receptor Models. J. Am. Chem. Soc. 2007;129:12809–12814. doi: 10.1021/ja073687x. PubMed DOI

Hinsberger S., Hüsecken K., Groh M., Negri M., Haupenthal J., Hartmann R.W. Discovery of Novel Bacterial RNA Polymerase Inhibitors: Pharmacophore-Based Virtual Screening and Hit Optimization. J. Med. Chem. 2013;56:8332–8338. doi: 10.1021/jm400485e. PubMed DOI

Temml V., Kaserer T., Kutil Z., Landa P., Vanek T., Schuster D. Pharmacophore Modeling for COX-1 and -2 Inhibitors with LigandScout in Comparison to Discovery Studio. Future Med. Chem. 2014;6:1869–1881. doi: 10.4155/fmc.14.114. PubMed DOI

Kurczab R., Bojarski A.J. New Strategy for Receptor-Based Pharmacophore Query Construction: A Case Study for 5-HT7 Receptor Ligands. J. Chem. Inf. Model. 2013;53:3233–3243. doi: 10.1021/ci4005207. PubMed DOI

Wieder M., Garon A., Perricone U., Boresch S., Seidel T., Maria Almerico A., Langer T. Common Hits Approach: Combining Pharmacophore Modeling and Molecular Dynamics Simulations. J. Chem. Inf. Model. 2017;57:365–385. doi: 10.1021/acs.jcim.6b00674. PubMed DOI

Polishchuk P., Kutlushina A., Bashirova D., Mokshyna O., Madzhidov T. Virtual Screening Using Pharmacophore Models Retrieved from Molecular Dynamic Simulations. Int. J. Mol. Sci. 2019;20:5834. doi: 10.3390/ijms20235834. PubMed DOI PMC

Moya M.M., Hush D.R. Network Constraints and Multi-Objective Optimization for One-Class Classification. Neural Netw. 1996;9:463–474. doi: 10.1016/0893-6080(95)00120-4. DOI

Varnek A., Baskin I. Machine Learning Methods for Property Prediction in Chemoinformatics: Quo Vadis? J. Chem. Inf. Model. 2012;52:1413–1437. doi: 10.1021/ci200409x. PubMed DOI

Kutlushina A., Khakimova A., Madzhidov T., Polishchuk P. Ligand-Based Pharmacophore Modeling Using Novel 3D Pharmacophore Signatures. Molecules. 2018;23:3094. doi: 10.3390/molecules23123094. PubMed DOI PMC

Gaulton A., Hersey A., Nowotka M., Bento A.P., Chambers J., Mendez D., Mutowo P., Atkinson F., Bellis L.J., Cibrián-Uhalte E., et al. The ChEMBL Database in 2017. Nucleic Acids Res. 2017;45:D945–D954. doi: 10.1093/nar/gkw1074. PubMed DOI PMC

ChemAxon Standardizer. Chemaxon; Budapest, Hungary: 2019. version 19.12.

RDKit: Open-Source Cheminformatics. 2019.

Mysinger M.M., Carchia M., Irwin J.J., Shoichet B.K. Directory of Useful Decoys, Enhanced (DUD-E): Better Ligands and Decoys for Better Benchmarking. J. Med. Chem. 2012;55:6582–6594. doi: 10.1021/jm300687e. PubMed DOI PMC

Truchon J.-F., Bayly C.I. Evaluating Virtual Screening Methods: Good and Bad Metrics for the “Early Recognition” Problem. J. Chem. Inf. Model. 2007;47:488–508. doi: 10.1021/ci600426e. PubMed DOI