DCMF-PPI: a protein-protein interaction predictor based on dynamic condition and multi-feature fusion

Area of Science:

• Computational Biology
• Bioinformatics
• Systems Biology

Background:

• Protein-protein interactions (PPIs) are vital for biological processes.
• Current PPI prediction methods often neglect protein dynamics and data limitations.
• This impacts model generalizability and predictive accuracy.

Purpose of the Study:

• To develop a novel framework for accurate protein-protein interaction prediction.
• To address limitations of existing methods by incorporating dynamic protein features.
• To improve the generalizability and precision of PPI prediction models.

Main Methods:

• Introduced DCMF-PPI (Dynamic condition and multi-feature fusion framework for PPI).
• Integrated dynamic modeling, multi-scale feature extraction, and probabilistic graph representation learning.
• Utilized PortT5-GAT, MPSWA, and Variational Graph Autoencoder modules for feature extraction and dynamic modeling.

Main Results:

• DCMF-PPI demonstrated superior performance over state-of-the-art methods in PPI prediction.
• Achieved significant improvements in accuracy, precision, and recall.
• Effectively modeled real-world biological complexities through fused dynamic conditions and multi-level features.

Conclusions:

• DCMF-PPI is a robust tool for advancing PPI research.
• The framework's dynamic and multi-feature approach enhances biological modeling.
• Positions DCMF-PPI for downstream applications in systems biology and drug discovery.