Protein function prediction from dynamic protein interaction network using gene expression data.

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• The increasing volume of protein sequence data outpaces functional annotation capabilities.
• Protein interactions are dynamic and crucial for understanding cellular functions.
• Accurate protein functional annotation is essential for biological research.

Purpose of the Study:

• To develop a computational method for predicting protein functional annotation.
• To leverage dynamic protein-protein interaction networks (PPINs) and time-course gene expression data.
• To address the challenge of annotating uncharacterized active proteins.

Main Methods:

• Utilized dynamic PPINs, time-course gene expression data, and protein sequence similarity (Damerau-Levenshtein edit distance).
• Assessed network edge strength using coefficient variation methods.
• Employed a bottom-up strategy to assign functional annotations from neighboring proteins.

Main Results:

• Achieved an average precision of 0.59, recall of 0.76, and F-Score of 0.61.
• Demonstrated significantly higher performance compared to state-of-the-art methods.
• Successfully annotated uncharacterized active proteins by exploring dynamic network neighbors.

Conclusions:

• The proposed methodology offers a robust approach for protein functional annotation.
• Dynamic network analysis combined with expression data enhances prediction accuracy.
• This work contributes to bridging the gap in protein functional characterization.