Advancing Identification of DNA-Protein Binding Residues Using Deep Learning Techniques

Accurate identification of DNA-protein binding sites is vital for understanding biological processes and facilitating drug discovery. This study introduces a novel method that inte-grates a Transformer encoder with Bi-directional Long Short-Term Memory BiLSTM to predict DNA-protein binding residues effectively. The method enriches protein representa-tion by combining evolutionary information from the position-specific scoring matrix PSSM with spatial information from predicted secondary structures. Experimental results demonstrate the method's competitiveness, achieving an MCC of 0.349, SP of 96.50 , SN of 44.03 , and ACC of 94.59 on the PDNA-41 dataset.