Prediction of protein-protein interactions and co-complex models with deep learning

Area of Science:

• Computational Biology
• Bioinformatics
• Artificial Intelligence in Biology

Background:

• Protein-protein interactions (PPIs) are crucial for cellular functions and disease pathogenesis.
• Understanding PPIs requires advanced computational approaches.
• Recent progress in deep learning offers new avenues for studying PPIs.

Purpose of the Study:

• To review recent deep learning-based methods for protein interaction studies.
• To focus on proteome-wide PPI prediction, PPI interface prediction, and PPI co-complex structure prediction.
• To explore the evolution of computational approaches and their applications in biology and medicine.

Main Methods:

• Categorization of recent deep learning approaches by methodological paradigms.
• Summarization of the strengths and limitations of various computational methods.
• Analysis of diverse biological and biomedical applications of PPI prediction techniques.

Main Results:

• Deep learning methods have significantly evolved for PPI prediction, interface analysis, and structure determination.
• Emerging computational concepts are reshaping the field of protein interaction studies.
• A comprehensive overview of current methodologies and their comparative performance is presented.

Conclusions:

• Computational methods, particularly deep learning, are vital for advancing our understanding of complex biological systems through PPI analysis.
• Integrated approaches in PPI prediction, interface, and structure prediction offer powerful tools for biological and biomedical research.
• Continued development in AI is expected to further enhance the study of protein interactions and their roles in health and disease.