Page 107 - AJWEP-v22i3

Advancing molecular property prediction using graph neural networks

Environ Sci. 2016;3:80. et al. Convolutional Networks on Graphs for Learning
doi: 10.3389/fenvs.2015.00080 Molecular Fingerprints. In: NIPS’15: Proceedings of
2. Wu Z, Ramsundar B, Feinberg EN, et al. MoleculeNet: the 29 International Conference on Neural Information
th
A benchmark for molecular machine learning. Chem Sci. Processing Systems; 2015.
2018;9(2):513-530. 16. Scarselli F, Gori M, Tsoi AC, Hagenbuchner M,
doi: 10.1039/c7sc02664a Monfardini G. The graph neural network model. IEEE
3. Debnath AK, Lopez de Compadre RL, Debnath G, Trans Neural Netw. 2009;20(1):61-80.
Shusterman AJ, Hansch C. Structure-activity relationship 17. Shervashidze N, Schweitzer P, Van Leeuwen EJ,
of mutagenic aromatic and heteroaromatic nitro Mehlhorn K, Borgwardt KM. Weisfeiler-lehman graph
compounds. Correlation with molecular orbital energies kernels. J Mach Learn Res. 2011;12:2539-2561.
and hydrophobicity. J Med Chem. 1991;34(2):786-797. 18. Veličković P, Casanova A, Liò P, Cucurull G, Romero A,
doi: 10.1021/jm00106a046 Bengio Y. Graph Attention Networks. In: 6 International
th
4. Kipf TN, Welling M. Semi-Supervised Classification Conference on Learning Representations, ICLR
with Graph Convolutional Networks. In: 5 International 2018 - Conference Track Proceedings; 2018. p. 1-12.
th
Conference on Learning Representations (ICLR); 2017. doi: 10.1007/978-3-031-01587-8_7
5. Xu K, Hu W, Leskovec J, Jegelka S. How Powerful are 19. Schütt KT, Arbabzadah F, Chmiela S, Müller KR,
Graph Neural Networks? In: International Conference Tkatchenko A. Quantum-chemical insights from deep
on Learning Representations; 2018. tensor neural networks. Nat Commun. 2017;8(1):13890.
6. Ramakrishnan R, Dral PO, Rupp M, Von Lilienfeld OA. doi: 10.1038/ncomms13890
Quantum chemistry structures and properties of 134 kilo 20. Kang X, Hu B, Perdana MC, Zhao Y, Chen Z. Extreme
molecules. Sci Data. 2014;1(1):140022. learning machine models for predicting the n-octanol/
doi: 10.1038/sdata.2014.22 water partition coefficient (K ) data of organic
ow
7. Irwin JJ, Shoichet BK. ZINC-A free database of compounds. J Environ Chem Eng. 2022;10(6):108552.
commercially available compounds for virtual screening. doi: 10.1016/j.jece.2022.108552
J Chem Inf Model. 2005;45:177-182. 21. Kenney DH, Paffenroth RC, Timko MT, Teixeira AR.
doi: 10.1021/ci049714+ Dimensionally reduced machine learning model for
8. Zeng XL, Wang HJ, Wang Y. QSPR models of n-octanol/ predicting single component octanol-water partition
water partition coefficients and aqueous solubility of coefficients. J Cheminform. 2023;15(1):9.
halogenated methyl-phenyl ethers by DFT method. doi: 10.1186/s13321-022-00660-1
Chemosphere. 2012;86(6):619-625. 22. Yokogawa D, Suda K. Interpretable Attribution
doi: 10.1016/j.chemosphere.2011.10.051 Assignment for Octanol-Water Partition Coefficient.
9. Wu Z, Ramsundar B, Feinberg EN, et al. MoleculeNet: [Chemrxiv Preprint]; 2023.
A benchmark for molecular machine learning. Chem Sci. doi: 10.26434/chemrxiv-2023-2lwsh-v2
2018;9(2):513-530. 23. Zamora WJ, Viayna A, Pinheiro S, et al. Prediction of
doi: 10.1039/C7SC02664A toluene/water partition coefficients in the SAMPL9 blind
10. Gilmer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE. challenge: Assessment of machine learning and IEF-
Neural Message Passing for Quantum Chemistry. In: PCM/MST continuum solvation models. Phys Chem
Proceedings of the 34 International Conference on Chem Physics. 2023;25(27):17952-17965.
th
Machine Learning; 2017. doi: 10.1039/D3CP01428B
11. Amjath M, Henna S, Rathnayake U. Graph representation 24. Ma W, Wang M, Jiang R, Chen W. A machine learning
federated learning for malware detection in Internet of based approach for estimating site-specific partition
health things. Results Eng. 2025;25:103651. coefficient Kd of organic compounds: Application to
doi: 10.1016/j.rineng.2024.103651 nonionic pesticides. Environ Pollut. 2023;323:121297.
12. Ellis LM, Mobley PE. Machine learning approaches doi: 10.1016/j.envpol.2023.121297
for predicting partition coefficients: A comparison of 25. Zhu Q, Jia Q, Liu Z, et al. Molecular partition coefficient
deep learning and classical methods. J Chem Inf Model. from machine learning with polarization and entropy
2021;61(9):4451-4464. embedded atom-centered symmetry functions. Phys
13. Nevolianis T, Rittig JG, Mitsos A, Leonhard K. Multi- Chem Chem Phys. 2022;24(38):23082-23088.
fidelity Graph Neural Networks for Predicting Toluene/ doi: 10.1039/D2CP02648A
Water Partition Coefficients [Chemrxiv Preprint]; 2024. 26. Ebert RU, Kühne R, Schüürmann G. Octanol/air partition
doi: 10.26434/chemrxiv-2024-3t818 coefficient-a general-purpose fragment model to predict
14. Sudhakar M, Jai A. Explainable AI for molecular log K from molecular structure. Environ Sci Technol.
oa
property prediction: Enhancing interpretability of deep 2023;57(2):976-984.
learning models. ACS Omega. 2022;7(4):2931-2942. doi: 10.1021/acs.est.2c06170
15. Duvenaud D, Maclaurin D, Aguilera-Iparraguirre J, 27. Stojanova M, Barré P, Clivot H, et al. A New

Volume 22 Issue 3 (2025) 101 doi: 10.36922/AJWEP025070041