IDBindT5 is a novel machine learning model that predicts binding residues in intrinsically disordered protein regions (IDPRs) using ProtT5 embeddings and disorder annotation as input. The model outperforms other pLM-based models and shows potential for predicting binding within IDPRs.
