Samsung Advanced Institute of Technology

Graph neural networks (GNNs) have proven effective in fast and accurate prediction of NMR chemical shifts for a molecule. Despite the effectiveness, the existing methods suffer from high space complexity and thus are limited to relatively small molecules. In this work, we propose a scalable GNN for NMR chemical shift prediction. To reduce the space complexity, We sparsify the graph representation of a molecule by regarding only heavy atoms and their chemical bonds as nodes and edges. To learn from the sparsified representation, we improve the message passing and readout functions in the GNN. For the message passing function, we adapt the attention mechanism and residual connection to better capture local information around each node. For the readout function, we use both node-level and graph-level embeddings as the local and global information to better predict node-level chemical shifts. Through experimental investigation using
13C and 1H NMR datasets, we demonstrate that the proposed method yields higher prediction accuracy and is more scalable to large molecules having many atoms.