Semi-Supervised GCN for learning Molecular Structure-Activity Relationships

TL;DR

This paper introduces a semi-supervised graph neural network approach to analyze how molecular structures influence properties like solubility and acidity, aiding drug discovery and chemical analysis.

Contribution

It presents a novel semi-supervised GCN method for learning structure-property relationships in molecules, applicable to medicinal chemistry tasks.

Findings

Consistent with experimental chemical data

Effective in predicting solubility and acidity

Potential to assist drug design processes

Abstract

Since the introduction of artificial intelligence in medicinal chemistry, the necessity has emerged to analyse how molecular property variation is modulated by either single atoms or chemical groups. In this paper, we propose to train graph-to-graph neural network using semi-supervised learning for attributing structure-property relationships. As initial case studies we apply the method to solubility and molecular acidity while checking its consistency in comparison with known experimental chemical data. As final goal, our approach could represent a valuable tool to deal with problems such as activity cliffs, lead optimization and de-novo drug design.