Semi-Supervised GCN for learning Molecular Structure-Activity Relationships
This could be a valuable tool for medicinal chemistry problems like activity cliffs and drug design, but it appears incremental as it builds on existing graph neural network methods.
The paper tackles the problem of attributing molecular structure-property relationships by proposing a semi-supervised graph-to-graph neural network, applying it to solubility and molecular acidity with consistency checks against experimental data.
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.