Path-Augmented Graph Transformer Network
This addresses the problem of improving molecular property prediction accuracy for researchers in chemistry and drug discovery, representing an incremental advancement over existing graph-based methods.
The paper tackled the limitation of Graph Convolution Networks (GCNs) in capturing higher-order graph properties for molecular representation learning by proposing Path-Augmented Graph Transformer Networks (PAGTN), which consistently outperformed GCNs on molecular property prediction datasets including quantum chemistry (QM7, QM8, QM9), physical chemistry (ESOL, Lipophilicity), and biochemistry (BACE, BBBP).
Much of the recent work on learning molecular representations has been based on Graph Convolution Networks (GCN). These models rely on local aggregation operations and can therefore miss higher-order graph properties. To remedy this, we propose Path-Augmented Graph Transformer Networks (PAGTN) that are explicitly built on longer-range dependencies in graph-structured data. Specifically, we use path features in molecular graphs to create global attention layers. We compare our PAGTN model against the GCN model and show that our model consistently outperforms GCNs on molecular property prediction datasets including quantum chemistry (QM7, QM8, QM9), physical chemistry (ESOL, Lipophilictiy) and biochemistry (BACE, BBBP).