Е-Публикации
Технически университет - София

Abstract: Using modern technologies such as machine learning (ML) and artificial intelligence (AI), we can accurately predict the biochemical properties of chemical compounds. The successful application of these approaches depends on both the chosen algorithms and the way of representing the molecular characteristics of the compounds, which can establish important biochemical and biophisics properties. The aim of the current research was to evaluate models based on molecular descriptors for predicting interactions with cannabinoid compounds. The models have demonstrated high accuracy for both regression and classification tasks. It is possible to use the characteristics obtained in this study for future research on cannabinoid compounds by focusing on key properties for compound optimization and the design of new drug molecules. Machine learning and artificial intelligence techniques are used as new approaches to support molecular docking and virtual screening of compounds. This study provides valuable information about the use of machine learning and artificial intelligence in drug design.

References

1. Abyadeh, M., et al.: A proteomic view of cellular and molecular effects of cannabis. Biomolecules 11, 1411 (2021)
2. Paduch, R., Kandefer-Szerszen, M., Trytek, M., Fiedurek, J.: Terpenes: substances useful in human healthcare. Arch. Immunol. Ther. Exp. 55(5), 315–327 (2007)
3. Devane, W., et al.: Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science 258, 1946–1949 (1992)
4. Mechoulam, R., et al.: Identification of an endogenous 2-monoglyceride, present in canine gut, that binds to cannabinoid receptors. Biochem. Pharmacol. 50, 83–90 (1995)
5. Sapundzhi, F., Dzimbova, T.: Computational modeling of compounds that interact with opioid and cannabinoid receptors. J. Chem. Technol. Metall. 59(4), 875–879 (2024)
6. Sapundzhi, F., Dzimbova, T.: Computer modelling of the CB1 receptor by molecular operating environment. Bulg. Chem. Commun. 50(Special Issue B), 15–19 (2018)
7. Sapundzhi, F., Dzimbova, T. Modelling and optimization of ligand binding to CBR2. Bulg. Chem. Commun. 52(A), 197–204 (2020)
8. Sapundzhi, F., Dzimbova, T.: Modelling of the structure-activity relationship of CBR2. J. Chem. Technol. Metall. 55(4), 709–713 (2020)
9. Sapundzhi, F.: Computer modeling and optimization of the structure-activity relationship by using surface fitting methods. Bul. Chem. Commun. 51(4), 569–579 (2019)
10. Sapundzhi, F., Dzimbova, T.: А computational study of cannabinoid receptors and cannabinoid ligands. J. Chem. Technol. Metall. 55(5), 959–964 (2020)
11. Sapundzhi, F., Lazarova, M., Dzimbova, T., Georgiev, S., Ivanova, A.: A structure-activity relationship modelling of opioid compounds by using machine learning. J. Phys. Conf. Ser. 2675(1), 012032 (2023)
12. Sapundzhi, F., Lazarova, M., Dzimbova, T., Georgiev, S.: An application of some machine learning methods for biological data modeling. J. Phys. Conf. Ser. 2675(1), 012021 (2023)
13. Sapundzhi, F., Georgiev, S., Lazarova, M., Popstoilov, M. Application of artificial intelligence and machine learning to study the structurebiological activity relationship. Ser. Biomech. (2025)
14. Muhammad, I., et al.: Cannabinoid and opioid receptor affinity and modulation of cancerrelated signaling pathways of machaeriols and machaeridiols from machaerium pers. Molecules 28(10), 4162 (2023)
15. Grimaldi, C., Bifulco, M.: Anandamide, an endogenous ligand of cannabinoid receptors, inhibits human breast cancer cell proliferation through a lipid raft mediated mechanism. Pharmacologyonline 1, 1–45 (2007)
16. Childers, S., Breivogel, C.: Cannabis and endogenous cannabinoid systems. Drug Alcohol Depend. 51, 173–187 (1998)
17. Hanus, L., et al.: Mechoulam, 2-arachidonyl glyceryl ether, an endogenous agonist of the cannabinoid CB1 receptor. Proc. Natl. Acad. Sci. U.S.A. 98, 3662–3665 (2001)
18. Tarawneh, A.H., et al.: M1, 2, 3-Triazole derivatives as highly selective cannabinoid receptor type 2 (CB2) agonists. Arab. J. Chem. 15(1), 103545 (2022)
19. Verdonk, M., Cole, J., Hartshorn, M., Murray, C., Taylor, R.: Improved protein-ligand docking using GOLD. Proteins 52, 609–623 (2003)
20. Sapundzhi, F., Prodanova, K., Lazarova, M.: AIP Conf. Proc. 2172(100008), 1–6 (2019)
21. Thomsen, R., Christensen, M.: MolDock: a new technique for high-accuracy molecular docking. J. Med. Chem. 49(11), 3315–3321 (2006)
22. Dimov, I., Lirkov, I., Margenov, S., Zlatev, Z. (eds.): NMA 2002. LNCS, vol. 2542. Springer, Heidelberg (2003). https://doi.org/10.1007/3-540-36487-0
23. Russel S., Norvig, P. Artificial Intelligence: A Modern Approach, 4th edn. Pearson, Boston (2020)
24. Kelleher, J.D., Namee, B.M., D’Arcy, A.: Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, Worked Examples, and Case Studies, 2nd edn. The MIT Press, New York (2020)
25. Georgiev, S., Vulkov, L.: Coefficient identification in a SIS fractional-order modelling of economic losses in the propagation of COVID–19. J. Comput. Sci. 69, 102007 (2023)
26. Dimov, I., Georgieva, R., Todorov, V.: Balancing of systematic and stochastic errors in Monte Carlo algorithms for integral equations. Lect. Notes Comput. Sci. 8962, 44–51 (2015)
27. Tsonev, T., Georgiev, S., Georgiev, I., Mihova, V., Pavlov, V.: Portfolio construction using neural networks and multiobjective optimization. In: Slavova, A. (ed.) New Trends in the Applications of Differential Equations in Sciences: NTADES 2023, Saints Constantine and Helena, Bulgaria, July 17–20, pp. 359–370. Springer, Cham (2024). https://doi.org/10.1007/978-3-031-53212-2_32
28. Klimenko, A., Mihova, V., Georgiev, S., Georgiev, I., Pavlov, V.: Comparative analysis on neural networks and ARIMA for forecasting heterogeneous portfolio returns. In: Slavova, A. (ed.) New Trends in the Applications of Differential Equations in Sciences: NTADES 2023, Saints Constantine and Helena, Bulgaria, July 17–20, pp. 335–346. Springer, Cham (2024). https://doi.org/10.1007/978-3-031-53212-2_30

Issue