STRUCTURE-ACTIVITY RELATIONSHIPS OF HIV-1 REVERSE TRANSCRIPTASE INHIBITORS: HOW TO INCREASE THE ACCURACY AND PREDICTABILITY OF MODELS?

Study objective was to assess the effect of data inconsistency on the quality of structural systemic analysis (SSA) models of reverse transcriptase inhibitors and to develop approaches to increasing the accuracy and predictive power of such models. Materials and methods: Structure-activity relationships in HIV-1 reverse transcriptase inhibitors were modeled using biologically active compound databases. Computer-assisted analysis and modeling of relationships between the structures and biological activities of chemical compounds allows predicting the activity of substances that were not studied experimentally, including even those only intended to be synthesized. Databases of biologically active compounds are extremely valuable sources for the selection of samples used to train SSA models. The quantitative characteristics of activity (IC50 and Ki) of a particular compound may be highly variable, especially if they have been determined at different laboratories, and this may cause considerable discrepancies in databases. Results: The adequacy of freely accessed and commercial databases to developing accurate and predictive SSA models for substances having an antiretroviral activity (HIV-1 reverse transcriptase inhibition) was assessed. The accuracy of SSA models was found to depend on the procedures of construction of a training sample. Certain limitation were found in databases used to construct training samples when sample entries have been testes under almost identical conditions. Therefore, it is expedient to develop a method for selecting low molecular weight compounds featuring low variability of their quantitative characteristics for being used in training of SSA models. Such method would increase the accuracy of SSA models employed in the design of novel antiretroviral substances.

ВИЧ-1, базы данных биологически активных веществ, компьютерное моделирование, взаимосвязь между структурой и активностью, вариабельность данных, HIV-1, databases of biologically active compounds, computer-assisted structure-activity relationships analysis, data inconsistency

1. Guasch L., Zakharov A., Tarasova O., Poroikov V.V, Liao C., Nicklaus M.C. Novel HIV-1 integrase inhibitor development by virtual screening based on QSAR models // Curr. Top. Med. Chem.- 2015.- Vol. 16, № 4.- Р. 441-448.

2. Kramer C., Kalliokoski T., Gedeck P., Vulpetti A. The experimental uncertainty of heterogeneous public K(i) data // J. Med. Chem.- 2012.- Vol. 55, № 11.- Р. 5165-5173.

3. Tarasova O., Urusova A., Filimonov D., Nicklaus M.C., Zakharov A.V., Poroikov V.V. QSAR Modeling Using Large-Scale Databases: Case Study for HIV-1 Reverse Transcriptase Inhibitors // J. Chem. Inf. Model.- 2015.- Vol. 55, № 7.- Р. 1388-1399.

4. Filimonov D., Zakharov A., Lagunin A., Poroikov V.V.QNA based «Star Track» QSAR approach // SAR and QSAR Environ. Res.- 2009.- Vol. 20, № 7-8.- Р. 679-709.

5. Zakharov A., Peach M., Sitzmann M., Nicklaus M.C. A new approach to radial basis function approximation and its application to QSAR // J. Chem. Inf. Model.- 2014.- Vol. 54, № 3.- Р. 713-719.

6. Berthold M., Cebron N., Dill F., Gabriel T., Kötter T., Meinl T., Ohl P., Sieb C., Thiel K., Wieswedel K. The Konstanz Information Miner. In: Studies in Classification, Data Analysis, and Knowledge Organization. Springer, Heidelberg, 2007.

7. Kelly T., Proudfoot J., McNeil D., Patel U., David E., Hargrave K., Grob P., Cardozo M., Agarwal A., Adams J. Novel non-nucleoside inhibitors of human immunodeficiency virus type 1 reverse transcriptase. 6.2-Indol-3-yl- and 2-azaindol-3-yl-dipyridodiazepinones // J. Med. Chem.- 1997.- Vol. 40, № 15.- Р. 2430-2433.

8. Mai A., Sbardella G., Artico M., Massa S., Novellino E., Greco G., Lavecchia А. Structure-based design, synthesis, and biological evaluation of conformationally restricted novel 2-alkylthio-6-[1-(2,6-difluorophenyl)alkyl]-3,4-dihydro-5-alkylpyrimidin-4(3H)-ones as non-nucleoside inhibitors of HIV-1 reverse transcriptase // J. Med. Chem.- 2001.- Vol. 44, № 16.- Р. 2544-2554.

9. Wyatt P., Bethell R., Cammack N., Charon D., Dodic N., Dumaitre B., Evans D., Green D., Hopewell P., Humber D., Lamont R., Orr D., Plested S., Ryan M., Sollis S., Storer R., Weingarten G. Benzophenone derivatives: a novel series of potent and selective inhibitors of human immunodeficiency virus type 1 reverse transcriptase // J. Med. Chem.- 1995.- Vol. 38, № 10.- Р. 1657-1665.

10. O’Meara J., Yoakim C., Bonneau P., Bos M., Cordingley M., Deziel R., Doyon L. Novel 8-substituted dipyridodiazepinone inhibitors with a broad-spectrum of activity against HIV-1 strains resistant to non-nucleoside reverse transcriptase inhibitors // J. Med. Chem.- 2005.- Vol. 48, № 17.- Р. 5580-5588.

11. Orchard S., Al-Lazikani B., Bryant S., Clark D., Calder E. Minimum information about a bioactive entity (MIABE) // Nat. Rev. Drug. Discov.- 2011.- Vol. 10, № 9.- Р. 661-669