Researchprot t5gnnpeptide protein interactionvirtual screening
PepLM-GNN Integrates ProtT5 for PepPI Prediction
|
8.3
On March 24, 2026, Yan et al. in PLoS Computational Biology present PepLM-GNN, a framework that integrates the ProtT5 pre-trained language model with a hybrid graph neural network to predict peptide–protein interactions. The model builds heterogeneous graphs from ProtT5 semantic embeddings and uses GCN plus GIN layers to improve non‑Euclidean representation and cold‑start generalization, outperforming prior methods and offering public code and an online screening service.
Scoring Rationale
Peer-reviewed methodological advance with usable code and service, but scope is focused on peptide–protein interactions.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
