Peculiarities of fully connected neural network design for estimating the lipophilicity of organic compounds
DOI: 10.21293/1818-0442-2024-27-1-86-94
Abstract: The assessment of lipophilicity of small organic compounds plays a crucial role in the development and optimization of new drugs. Unfortunately, experimental methods require significant time and resources, including the use of laboratory equipment and reagents. Additionally, manual verification and data adjustment often increase the process's labor intensity. In contrast, computational methods like machine learning offer faster and less resource-intensive ways to assess lipophilicity, allowing for efficient processing of large data volumes and adaptation to complex relationships between molecular structure and lipophilicity. Developing neural network models for lipophilicity assessment is challenging due to insufficient experimental data and high computational costs with graph neural network models. This work presents an analysis of popular methods for describing chemical structures for building fully connected neural network models, less demanding in training data volume. Based on this analysis, features best describing organic compounds from an open lipophilicity dataset collected from the ChEMBL database are selected. The search for the optimal neural network model architecture for the chosen features is conducted.
Keywords: modeling, neural network, lipophilicity, cheminformatics
For citation:
Pyakillya B. I., Goncharov V. I. Peculiarities of fully connected neural network design for estimating the lipophilicity of organic compounds. Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki, 2024, vol. 27, no. 1, pp. 86–94. DOI: 10.21293/1818-0442-2024-27-1-86-94
Authors and copyright holders:
- Pyakillya B. I. , Tomsk Polytechnic University (Tomsk, Russia)
- Goncharov V. I. , National Research Tomsk Polytechnic University (Tomsk, Russia)
- 1. Wardecki D., Dołowy M., Bober-Majnusz K. Assessment of Lipophilicity Parameters of Antimicrobial and Immunosuppressive Compounds. Molecules, 2023, vol. 28, no. 6, pp. 1–14.
- 2. Miller R., Madeira M., Wood H., Geissler W., Raab C., Martin I. Integrating the Impact of Lipophilicity on Potency and Pharmacokinetic Parameters Enables the Use of Diverse Chemical Space during Small Molecule Drug Optimization. Journal of Medical Chemistry, 2020, vol. 63, no. 21, pp. 12156–12170.
- 3. Piakillia B. I., Goncharov V.I. Ocenivanie lipofil'nosti s pomoshh'ju bajesovskih nejronnyh setej. [Lipophilicity estimation using Bayesian neural networks] Izvestiia TulGu. Tehnicheskie Nauki, 2022, no. 9, pp. 288–292 (in Russ.)
- 4. Tang B., Kramer S., Fang M. A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility. Journal of Cheminformatics, 2020, vol. 12, no. 15, pp. 1–9.
- 5. Song Y., Chen J., Wang W. Double-head transformer neural network for molecular property prediction. Journal of Cheminformatics, 2023, vol. 15, no. 27, pp. 1–16.
- 6. Wang Y., Xiong J., Xiao F. LogD7.4 prediction enhanced by transfer-ring knowledge from chromatographic retention time, mi-croscopic pKa and logP. Journal of Cheminformatics, 2023, vol. 15, no. 76, pp. 1–13.
- 7. Wieder O., Keunemann M., Wieder T., Seidel T., Meyer C., Bryant S., Langer T. Improved Lipophilicity and Aqueous Solubility Prediction with Composite Graph Neural Networks. Molecules, 2021, vol. 26, no. 20, pp. 6185–6223.
- 8. Gafarov F.M, Galimjanov A.F. Iskusstvennye nejronnye seti i prilozhenija: ucheb. posobie [Artificial Neural Networks and Applications: a study guide]. Kazan, Kazan Univ. Publ., 2018, 121 p. (in Russ.)
- 9. Bengio Y., Goodfellow I., Courville A. Deep Learning. MIT Press, 2016. 800 p.
- 10. Zheron A. Prikladnoe mashinnoe obuchenie s pomoshh'ju Scikit-Learn i TensorFlow: koncepcii, instrumenty i tehniki dlja sozdanija intellektual'nyh system [Applied Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques for Creating Intelligent Systems]. Saint-Petersburg. Alfa-kniga Publ., 2018. 688 p. (in Russ.)
- 11. Documentation RDkit: https://www.rdkit.org/docs/ (Accessed: March 26, 2024).
- 12. Documentation Numpy: https://numpy.org/doc/stable/ (Accessed: March 26, 2024).
- 13. Documentation Pandas: https://pandas.pydata.org/docs/reference/index.html (Accessed: March 26, 2024).
- 14. Documentation Matplotlib: https://matplotlib.org/stable/index.html (Accessed: March 26, 2024).
- 15. Documentation TensorFlow: https://www.tensorflow.org/api_docs (Accessed: March 26, 2024).
- 16. Documentation Seaborn: https://seaborn.pydata.org/api.html (Accessed: March 26, 2024).
- 17. Wu Z., Ransundar B., Feinberg E. MoleculeNet: a benchmark for molecular machine. Chemical Science, 2018, vol. 9, no. 2, pp. 513–530.
- 18. Waring M. Lipophilicity in drug discovery. Expert Opinion on Drug Discovery, 2010, vol. 5, no. 3, pp. 235–248.
- 19. Madzhidov T.I., Baskin I.I., Varnek A.A. Vvedenie v hemoinformatiku: uchebnoe posobie. Ch. 1: Komp'juternoe predstavlenie himicheskih struktur [Introduction to Chemoinformatics: A Study Guide. Part 1: Computer Representation of Chemical Structures]. Kazan, Kazan Univ. Publ., 2015, 174 p. (in Russ.).
- 20. Gómez-Bombarelli R., Jennifer W., Duvenaud D. Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules. ACS Central Science, 2018, vol. 4, no. 2, pp. 268–276.
- 21. Madzhidov T.I., Baskin I.I., Varnek A.A. Vvedenie v hemoinformatiku: uchebnoe posobie. Ch. 4: Metody mashinnogo obuchenija. [Introduction to Cheminformatics: Study Guide. Part 4: Machine Learning Methods.] Kazan, Kazan Univ. Publ., 2016, 329 p. (in Russ.).
- 22. Stokes J.M., Swanson K., Yang K. A Deep Learning Approach to Antibiotic Discovery. Cell, 2020, vol. 180, no. 4, pp. 475–483.
- 23. Niazi S., Mariam Z. Advances in Machine-Learning-Based Chemoinformatics: A Comprehensive Review. International Journal of Molecular Sciences, 2023, vol. 24, pp. 11488–11503.
- 24. Rickert C., Henkel M., Lieleg O. Efficiency-driven, correlation-based feature elimination strategy for small datasets. Machine Learning, 2023, vol. 1, no. 1, pp. 1–15.
- 25. Ali Y.A Hyperparameter Search for Machine Learning Algorithms for Optimizing Computational Complexity. Processes, 2023, vol. 11, no. 2, pp. 1–21.
- 26. Gilmer J., Schoenholz S., Riley P., Vinyals O. Neural message passing for quantum chemistry. International Conference on Machine Learning. PMLR, 2017, pp. 1263–1272.
- 27. Kearnes S., McCloskey K., Berndl M., Pande V., Riley P. Molecular graph convolutions: moving beyond fingerprints Journal of Computer-Aided Molecular Design, 2016, vol. 30, pp. 595–608.
- 28. Chen T., Guestrin C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
Last issue
-2026-1_%D0%BA%D0%BE%D0%BF%D0%B8%D1%8F.jpg)
Editorial office address
Executive Secretary of the Editor’s Office
Editor’s Office: 40 Lenina Prospect, Tomsk, 634050, Russia
Phone / Fax: + 7 (3822) 701-582
Viktor N. Maslennikov
Executive Secretary of the Editor’s Office
Editor’s Office: 40 Lenina Prospect, Tomsk, 634050, Russia
Phone / Fax: + 7 (3822) 51-21-21 / 51-43-02
Subscription for updates
