Proteins Combined Score Prediction Based on Improved Gene Expression Programming Algorithm and Protein-Protein...

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Protein-protein interaction (PPI) network analysis is crucial for understanding biological systems.
• Existing methods struggle with heterogeneous data integration, nonlinear dependencies, and generalizability.
• Accurate PPI score prediction enhances data reliability and biological insights.

Purpose of the Study:

• To develop an advanced algorithm for predicting combined scores in protein-protein interaction networks.
• To overcome limitations of current intelligent algorithms in data integration and nonlinear dependency capture.
• To improve the accuracy, stability, and generalizability of PPI prediction models.

Main Methods:

• Introduced a Dynamic Factor Gene Expression Programming (DF-GEP) algorithm.
• Integrated Spearman correlation analysis with kernel ridge regression (SC-KRR) for feature weighting.
• Employed dynamic factor adjustment to optimize evolutionary processes (selection, crossover, mutation, fitness evaluation).

Main Results:

• The DF-GEP algorithm demonstrated superior performance compared to baseline models.
• Achieved consistent improvements in both predictive accuracy and model stability.
• Validated the algorithm's effectiveness in predicting PPI combined scores.

Conclusions:

• The proposed DF-GEP algorithm offers a significant advancement in PPI network analysis.
• The method effectively handles complex nonlinear problems and heterogeneous data.
• DF-GEP shows strong potential for broader applications in computational biology and beyond.