Page 94 - AJWEP-v22i3

Asian Journal of Water, Environment and Pollution. Vol. 22, No. 3 (2025), pp. 88-103.
doi: 10.36922/AJWEP025070041

ORIGINAL RESEARCH ARTICLE

Environmental applications of molecular graph learning:
Graph neural network based prediction of partition

coefficients

Pravinkumar M. Sonsare * , Roshni Khedgaonkar ,
2
1
Kavita Singh , and Pratik Agrawal 3
2
1 Department of Computer Science and Engineering, Shri Ramdeobaba College of Engineering and Management,
Ramdeobaba University, Nagpur, Maharashtra, India
2 Department of Computer Technology, Yeshvantrao Chavhan College of Engineering, Nagpur, Maharashtra, India
3 Symbiosis Institute of Technology, Nagpur Campus, Symbiosis International (Deemed University),
Pune, Maharashtra, India
(This article belongs to the Special Issue: Renewable Energy Systems and Strategies in Smart Grids and
Smart Cities Development)
*Corresponding author: Pravinkumar M. Sonsare (sonsare@gmail.com)

Received: February 12, 2025; 1st revised: February 23, 2025; 2nd revised: March 24, 2025; 3rd revised: April 15, 2025;
4th revised: April 29, 2025; Accepted: April 29, 2025; Published Online: May 29, 2025

Abstract: In cheminformatics, predicting molecular properties is crucial for enhancing material research,
toxicity assessment, and drug discovery. This research investigates the use of graph neural networks (GNNs)
for predicting molecular properties by examining three different architectures: graph convolutional networks
(GCNs), graph isomorphism networks (GINs), and graph attention networks (GATs). Employing molecular
graph information, these models are evaluated on the MUTEG dataset and measured against key metrics such
as accuracy and area under the receiver operating characteristic curve (AUC). Our experimental findings show
that GIN has the highest accuracy at 89.2%, exceeding GCN (87.5%) and GAT (88.3%). GIN also achieves
the highest AUC of 0.89, whereas the AUCs of GCN and GAT are 0.84 and 0.86, respectively, indicating
GIN’s enhanced ability to effectively model graph isomorphisms. We selected GIN for this study because of
its proven theoretical and empirical strength in capturing graph-level representations, particularly in domains
such as cheminformatics, where molecular structures are naturally modeled as graphs. These results highlight
the efficacy of GNNs in predicting molecular properties and position GIN as a favored framework for tasks
that demand accurate graph feature extraction. This study further plays a pivotal role in understanding the
environmental fate and transport of chemical compounds. We used GIN to identify partition coefficients such as
the octanol-water partition coefficient, air-water partition coefficient, and soil–water partition coefficient from
the MoleculeNet dataset.

Keywords: Graph neural networks; Molecular property prediction; Cheminformatics; Drug discovery; Structure-
property relationships

Volume 22 Issue 3 (2025) 88 doi: 10.36922/AJWEP025070041