Graph-Based Approach for Solubility Prediction of Drugs using SMILES Data
Keywords:
GCN, GNN, Grid Hyperparameter Optimization, RMSE, SMILESAbstract
Graph Neural Networks (GNN) utilization in the case of molecular property prediction is considered a significant advancement in computational chemistry and drug discovery. Initial approaches to molecular property prediction especially solubility prediction depend on empirical rules or physicochemical descriptors, which lack generalization and predictive accuracy. The proposed model Graph Convolutional Network (GCN) which is a variant of GNN learns representations of molecular graphs, enabling accurate prediction of molecular properties directly from raw molecular structures. The molecular graphs are created from the Simplified Molecular Input Line Entry System (SMILES) data which are molecular sequences of drug target compounds. In the proposed work, GCN uses graph pooling, which effectively reduces the node dimensionality. This work shows how the whole graph can be considered as input and how different pooling techniques can be used to handle large and complex graph data and also the effectiveness of GCN for solubility prediction. The proposed GCN model is hyperparameter tuned by using Grid Hyperparameter optimization on ESoL dataset which is a regressive type dataset achieving a low RMSE value of 0.43 outperforming machine learning and many deep learning models.
Downloads
References
Kipf, Thomas N., and Max Welling. "Semi-Supervised Classification with Graph Convolutional Networks." *Proceedings of the International Conference on Learning Representations (ICLR)*, 2017. arXiv:1609.02907.
Hamilton, William L., Rex Ying, and Jure Leskovec. "GraphSAGE: Inductive Representation Learning on Large Graphs." *Advances in Neural Information Processing Systems 30 (NIPS 2017)*, 2017. arXiv:1706.02216.
Veličković, Petar, et al. "Graph Attention Networks." *Proceedings of the International Conference on Learning Representations (ICLR)*, 2018. arXiv:1710.10903.
Xu, Keyulu, et al. "How Powerful Are Graph Neural Networks?" *Proceedings of the International Conference on Learning Representations (ICLR)*, 2019. arXiv:1810.00826.
Ying, Rex, et al. "Hierarchical Graph Representation Learning with Differentiable Pooling." *Proceedings of the 32nd International Conference on Neural Information Processing Systems (NeurIPS 2018)*, 2018. arXiv:1806.08804.
Ying, Rex, et al. "DiffPool: Graph Pooling via Learning Differentiable Graph Structures." *Proceedings of the 32nd International Conference on Neural Information Processing Systems (NeurIPS 2018)*, 2018. arXiv:1806.08804.
Zhou, Jie, et al. "Understanding Graph Convolutional Networks for Node Classification." 2018. arXiv:1806.09978.
Li, Y., Zhang, L., and Liu, Z. "Deep Graph Convolutional Encoder-Decoder Networks for Representation Learning of Chemical Molecules." 2018. arXiv:1711.07553.
Selsam, Stefan W. R., Nikolaj Bjørner, and William Cook. "Mean Field Graph Convolutional Neural Networks." 2019. arXiv:1810.05997.
Gao, Hongyang, and Shuiwang Ji. "Graph U-Nets." *Proceedings of the 36th International Conference on Machine Learning (ICML)*, 2019. arXiv:1905.05178.
Rieck, Bastian, Christian Bock, and Heiko Strathmann. "Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks." *Proceedings of the 34th AAAI Conference on Artificial Intelligence (AAAI 2020)*, 2020. arXiv:1906.01277.
Karpov, Pavel, Grigory Godin, and Igor V. Tetko. "TRANSFORMER-CNN: Fast and Reliable Tool for QSAR." 2020. arXiv:2002.06038.
Varikuti, H., Vatsavayi, V. (2024). malC: A Novel Deep Learning Architecture for Malware Classification. In: Roy, N.R., Tanwar, S., Batra, U. (eds) Cyber Security and Digital Forensics. REDCYSEC 2023. Lecture Notes in Networks and Systems, vol 896. Springer, Singapore, pp 435-448 .
Downloads
Published
How to Cite
Issue
Section
License
![Creative Commons License](http://i.creativecommons.org/l/by-sa/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.