InterPred: A pipeline to identify and model protein-protein interactions

Area of Science:

• Computational biology
• Structural bioinformatics
• Molecular modeling

Background:

• Protein-protein interactions (PPIs) are essential for cellular functions.
• Experimental methods for PPI identification are time-consuming and cannot characterize all interactions.
• Computational approaches are needed to predict and model PPIs at a molecular level.

Purpose of the Study:

• To present InterPred, an automated pipeline for predicting and modeling PPIs from protein sequences.
• To improve the accuracy and efficiency of PPI detection and modeling.

Main Methods:

• Utilizes structural modeling, massive structural comparisons, and molecular docking.
• Employs a novel random forest classifier integrating structural features for accurate PPI model discrimination.

Main Results:

• InterPred demonstrates performance comparable to or better than experimental high-throughput techniques for PPI detection.
• The protein-protein complex modeling pipeline outperforms state-of-the-art protein docking methods.
• InterPred achieved top predictor status in the CAPRI37 experiment.

Conclusions:

• InterPred offers a significant advancement in computational prediction and modeling of protein-protein interactions.
• The pipeline provides a powerful tool for understanding molecular interactions and protein function.