TripletGO: Integrating Transcript Expression Profiles with Protein Homology Inferences for Gene Function Prediction

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Gene Ontology (GO) is crucial for annotating gene and protein functions.
• Accurate functional annotation is essential for understanding biological systems.
• Existing methods face challenges in annotating diverse gene types and species.

Purpose of the Study:

• To develop a novel computational method, TripletGO, for improved gene function annotation.
• To integrate diverse biological data sources for enhanced accuracy.
• To provide a robust tool for both protein-coding and non-coding genes.

Main Methods:

• TripletGO integrates four complementary pipelines: transcript expression, genetic sequence alignment, protein sequence alignment, and naïve probability.
• A triplet network-based profiling method coupled with feature space mapping is employed.
• The method was validated on a large dataset across eight species and CAFA3 proteins.

Main Results:

• TripletGO significantly outperforms current state-of-the-art methods in function annotation accuracy.
• The integration of transcript expression and protein-level alignments enhances annotation coverage and precision.
• The triplet network profiling and feature mapping effectively capture functional patterns from expression data.

Conclusions:

• TripletGO offers a powerful and accurate approach for gene function annotation.
• The method's ability to integrate multiple data types provides comprehensive functional insights.
• TripletGO is available as a standalone package and an online server for broader accessibility.