O.V. Nepomnyashchiy1, A.G. Khantimirov2, M.M.I. Al-Saghir3, S.A. Kotov4

1–4 Siberian Federal University (Krasnoyarsk, Russia)

1 ONepomnuashy@sfu-kras.ru, 2 anton-xant@mail.ru, 3 mahamedali51@yahoo.com

The article delves into the complexities surrounding the comprehensive utilization of pharmaceuticals through the lens of computational pharmaceutics. It underscores the significant cost implications associated with diagnosing the outcomes of such interactions using conventional laboratory techniques. As the medication repertoire expands, it becomes increasingly impractical to identify all potential interaction outcomes. The study explores established computational methodologies and highlights the challenge of ensuring the completeness and accuracy of results, stemming from the diverse sets of resulting interactions employed in machine learning.

A hypothesis is posited regarding the potential of deep learning technologies to discern the sought-after function from numerous interaction outcomes, thereby enhancing forecasting accuracy. To address this, a solution is proposed, leveraging a combined approach of a convolutional neural network (CNN) and an evolutionary algorithm. The research outlines the development results of the neural network architecture and elucidates its operational principles. Furthermore, it entails the creation of a computational model and the preparation of datasets for training and testing the neural network.

Training is conducted, and comparative analysis results of various forecasting modes using a range of known algorithms are obtained. The findings demonstrate that employing the particle swarm optimization algorithm leads to enhanced accuracy, sensitivity, selectivity, and reliability compared to conventional methods on identical datasets.

Despite significant advancements in computational pharmaceutics, existing approaches inadequately meet modern requirements for forecast completeness and accuracy, primarily due to the extensive medication repertoire and the multitude of features necessary for training computational models. However, deep learning technologies offer the potential to extract localized features from a plethora of unknowns, thereby augmenting diagnostic accuracy.

The proposed method for predicting drug interactions, based on deep learning technology implemented through a CNN, adjusts key network parameters using the particle swarm optimization evolutionary algorithm. The architecture of the proposed CNN minimizes memory requirements by placing weights in the network's channel layers, potentially improving computational efficiency and expediting computation time.

In the proposed prognostic model, key variables for predicting drug interaction include the size of the convolution layer window, the subsampling layer window size, and the number of neurons in the hidden layer of the fully connected output neural network.

Nepomnyashchiy O.V., Khantimirov A.G., Al-Saghir M.M.I., Kotov S.A. Application of convolutional neural network and evolutionary algorithm to predict the outcome of adverse drug interactions. Neurocomputers. 2024. V. 26. № 4. Р. 45-55. DOI: https://doi.org/10.18127/j19998554-202404-05 (In Russian)

1. Database of medicines withdrawn from circulation. [Electronic resource] – Access mode: http://www.omdrug.ru/pro-info/drug_base/, date of reference 14.03.2024. (In Russian)
2. Gusev A.V., Romanov F.A., Dudanov I.P., Voronin A.V. Medical information systems: Monograph. Petrozavodsk: PetrSU. 2005. 404 p. (In Russian)
3. Alhaj F., Qutishat D., Harahsheh H.A., Obeid N., Hammo B. Detecting DDI Using Ontology: Drug Mechanism of Action. IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology. 2019. P. 179–185. DOI 10.1109/JEEIT.2019. 8717527.
4. Marcath L.A., Coe T.D., Hoylman E.K., Redman B.G., Hertz D.L. Prevalence of drug-drug interactions in oncology patients enrolled on National Clinical Trials Network oncology clinical trials. BMC Cancer. 2018. V. 18. № 1. P. 1–8. DOI 10.1186/s12885-018-5076-0.
5. Zhang C., Zang T. CNN-DDI: A novel deep learning method for predicting drug-drug interactions. IEEE International Conference on Bioinformatics and Biomedicine. 2020. P. 1708–1713. DOI 10.1109/BIBM49941.2020.9313404.
6. Zhang W., Chen Y., Liu F., Luo F., Tian G., Li X. Predicting potential drug-drug interactions by integrating chemical, biological, phenotypic and network data. BMC Bioinformatics. 2017. V. 18. № 1. P. 18. DOI 10.1186/s12859-016-1415-9.
7. Purkayastha S., Mondal I., Sarkar S., Goyal P., Pillai J.K. Drug-Drug Interactions Prediction Based on Drug Embedding and Graph Auto-Encoder. IEEE 19th International Conference on Bioinformatics and Bioengineering. 2019. P. 547–552. DOI 10.1109/BIBE.2019. 00104.
8. Ferdousi R., Safdari R., Omidi Y. Computational prediction of drug-drug interactions based on drugs functional similarities. Journal of Biomedical Informatics. 2017. V. 70. P. 54–64. DOI 10.1016/j.jbi.2017.04.021.
9. Sun X., Ma L., Du X., Feng J., Dong K. Deep Convolution Neural Networks for Drug-Drug Interaction Extraction. IEEE International Conference on Bioinformatics and Biomedicine. 2018. P. 1662–1668. DOI 10.1109/BIBM.2018.8621405.
10. Weng Y.-A., Deng C.-Y., Pu C. Targeting continuity of care and polypharmacy to reduce drug-drug interaction. Scientific reports. 2020. V. 10. № 1. P. 1–9. DOI 10.1038/s41598-020-78236-y.
11. Simon D. Algorithms of evolutionary optimization. Moscow: DMK Press. 2020. 1002 p. (In Russian)
12. Korshunova K.P. Convolutional fuzzy neural networks for solving classification problems. Neurocomputers. 2017. № 3. P. 44-–51. (In Russian)
13. Vilar S., Harpaz R., Uriarte E., Santana L., Rabadan R., Friedman C. Drug-drug interaction through molecular structure similarity analysis. Journal of the American Medical Informatics Association. 2012. V. 19. № 6. P. 1066–1074. DOI 10.1136/amiajnl-2012-000935.
14. Gottlieb A., Stein G.Y., Oron Y., Ruppin E., Sharan R. INDI: a computational framework for inferring drug interactions and their associated recommendations. Molecular systems biology. 2012. V. 8. № 1. P. 592. DOI 10.1038/msb.2012.26.
15. Cheng F., Zhao Z. Machine learning-based prediction of drug-drug interactions by integrating drug phenotypic, therapeutic, chemical, and genomic properties. Journal of the American Medical Informatics Association. 2014. V. 21. № e2. P. e278–e286. DOI 10.1136/ amiajnl-2013-002512.
16. Zhang P., Wang F., Hu J., Sorrentino R. Label Propagation Prediction of Drug-Drug Interactions Based on Clinical Side Effects. Scientific reports. 2015. V. 5. № 1. P. 1–10. DOI 10.1038/srep12339.
17. Salp Swarm Algorithm. [Electronic resource] – Access mode: https://www.baeldung.com/cs/salp-swarm-algorithm, date of reference 14.03.2024.
18. Mirjalili S., Gandomi A.H., Mirjalili S.Z., Saremi S., Faris H., Mirjalili S.M. Salp Swarm Algorithm: A bio-inspired optimizer for engineering design problems. Advances in Engineering Software. 2017. V. 114. P. 163–191. DOI 10.1016/j.advengsoft.2017.07.002.
19. Matsugu M., Mori K., Mitari Y., Kaneda Y. Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Networks. 2003. V. 16. № 5-6. P. 555–559. DOI 10.1016/S0893-6080(03)00115-1.
20. Romanuke V. Appropriate number and allocation of ReLUs in convolutional neural networks. Research Bulletin of the National Technical University of Ukraine Kyiv Politechnic Institute. 2017. V. 1. P. 69–78. DOI 10.20535/1810-0546.2017.1.88156.
21. GitHub –YifanDengWHU/DDIMDL. [Electronic resource] – Access mode: https://github.com/YifanDengWHU/DDIMDL, date of reference 14.03.2024.
22. GitHub – tulga-rdn/re_deepddi2: Rewriting DeepDDI2 using pytorch for better compatibility. [Electronic resource] – Access mode: https://github.com/tulga-rdn/re_deepddi2, date of reference 14.03.2024.
23. GitHub – mims-harvard/decagon: Graph convolutional neural network for multirelational link prediction. [Electronic resource] – Access mode: https://github.com/mims-harvard/decagon, date of reference 14.03.2024.
24. Installation – Locust 2.29 documentation. [Electronic resource] – Access mode: https://docs.locust.io/en/stable/installation.html, date of reference 14.03.2024.