Comparing gene annotation enrichment tools for functional modeling of agricultural microarray data.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Microarray technology enables functional genomics studies, requiring robust gene annotation.
• Existing gene annotation enrichment tools often lack support for agricultural species.
• AgBase serves as a key functional annotation reference for agricultural species.

Purpose of the Study:

• To evaluate and compare four public computational tools (Onto-Express, EasyGO, GOstat, DAVID) for gene expression analysis in agricultural species.
• To assess tools based on features, statistical methods, and annotation/enrichment performance.
• To analyze chicken gene annotation datasets using AgBase as a reference.

Main Methods:

• Comparative analysis of four publicly available gene annotation tools.
• Evaluation criteria included features, accessibility, statistical methods, and performance.
• Annotation and enrichment analysis performed on chicken gene sets (random and experimental).

Main Results:

• Tools produced varying annotations and functional groups for chicken datasets.
• DAVID provided the most annotations per gene, but with redundancy and high-level terms.
• GOstat, Onto-Express, and EasyGO showed similar GO distributions to AgBase, unlike DAVID.
• EasyGO failed to identify significantly enriched terms; other tools showed limited agreement.

Conclusions:

• Significant discrepancies exist in functional annotation and GO category enrichment results across different tools.
• The study highlights the need for caution when interpreting functional genomics analysis results.
• Lack of standardization in gene annotation approaches necessitates careful tool selection and result validation.