Encoding protein dynamic information in graph representation for functional residue identification
This work addresses the challenge of identifying functional residues in proteins for biotechnology and pharmaceutical applications, though it appears incremental by augmenting existing graph representations with dynamic information.
The authors tackled the problem of protein function prediction by incorporating protein dynamics into graph-based deep learning, resulting in a method (ProDAR) that achieved a remarkable performance gain in multilabel function classification tasks.
Recent advances in protein function prediction exploit graph-based deep learning approaches to correlate the structural and topological features of proteins with their molecular functions. However, proteins in vivo are not static but dynamic molecules that alter conformation for functional purposes. Here we apply normal mode analysis to native protein conformations and augment protein graphs by connecting edges between dynamically correlated residue pairs. In the multilabel function classification task, our method demonstrates a remarkable performance gain based on this dynamics-informed representation. The proposed graph neural network, ProDAR, increases the interpretability and generalizability of residue-level annotations and robustly reflects structural nuance in proteins. We elucidate the importance of dynamic information in graph representation by comparing class activation maps for hMTH1, nitrophorin, and SARS-CoV-2 receptor binding domain. Our model successfully learns the dynamic fingerprints of proteins and pinpoints the residues of functional impacts, with vast untapped potential for broad biotechnology and pharmaceutical applications.