ISPRED-SEQ: Deep Neural Networks and Embeddings for Predicting Interaction Sites in Protein Sequences

Area of Science:

• Bioinformatics
• Computational Biology
• Structural Biology

Background:

• Protein-protein interactions (PPIs) are fundamental to cellular processes.
• Identifying PPI sites is crucial for protein functional annotation.
• The abundance of protein sequences necessitates computational prediction methods.

Purpose of the Study:

• To develop a novel computational method for predicting protein-protein interaction sites.
• To leverage recent advancements in protein language models and deep neural networks for PPI prediction.
• To provide an accurate and accessible tool for researchers studying protein interactions.

Main Methods:

• Utilized protein language models for feature extraction from protein sequences.
• Employed Deep Neural Networks for the classification of potential PPI sites.
• Developed a predictor named ISPRED-SEQ.

Main Results:

• ISPRED-SEQ demonstrates superior performance compared to existing state-of-the-art predictors.
• The method accurately predicts putative protein-protein interaction sites directly from protein sequences.
• The predictor effectively addresses the challenge posed by the vast number of available protein sequences.

Conclusions:

• ISPRED-SEQ represents a significant advancement in the computational prediction of protein-protein interaction sites.
• The tool enhances protein functional annotation capabilities by providing sequence-based predictions.
• ISPRED-SEQ is available as a free web service for the scientific community.