FragmentNet: Adaptive Graph Fragmentation for Graph-to-Sequence Molecular Representation Learning
This provides a scalable and chemically optimal tool for molecular property prediction, aiding in molecular design and optimization.
The authors tackled the problem of generating chemically interpretable molecular representations by introducing FragmentNet, a graph-to-sequence model with an adaptive tokenizer that decomposes molecular graphs into valid fragments, which outperformed similarly scaled models and rivaled larger state-of-the-art ones on MoleculeNet tasks.
Molecular property prediction uses molecular structure to infer chemical properties. Chemically interpretable representations that capture meaningful intramolecular interactions enhance the usability and effectiveness of these predictions. However, existing methods often rely on atom-based or rule-based fragment tokenization, which can be chemically suboptimal and lack scalability. We introduce FragmentNet, a graph-to-sequence foundation model with an adaptive, learned tokenizer that decomposes molecular graphs into chemically valid fragments while preserving structural connectivity. FragmentNet integrates VQVAE-GCN for hierarchical fragment embeddings, spatial positional encodings for graph serialization, global molecular descriptors, and a transformer. Pre-trained with Masked Fragment Modeling and fine-tuned on MoleculeNet tasks, FragmentNet outperforms models with similarly scaled architectures and datasets while rivaling larger state-of-the-art models requiring significantly more resources. This novel framework enables adaptive decomposition, serialization, and reconstruction of molecular graphs, facilitating fragment-based editing and visualization of property trends in learned embeddings - a powerful tool for molecular design and optimization.