Biological profiling of gene groups utilizing Gene Ontology

Area of Science:

• Genomics
• Bioinformatics
• Statistical Biology

Background:

• High-throughput techniques like microarrays generate large gene datasets requiring biological interpretation.
• Gene Ontology provides functional annotations, but determining statistical significance of biological processes remains challenging.
• Multiple testing is a critical issue in analyzing gene expression data to avoid misleading conclusions.

Purpose of the Study:

• To develop a statistical framework for identifying significantly enriched Gene Ontology terms within gene groups.
• To address the challenge of determining statistical significance while accounting for multiple testing.
• To provide a robust method for automated biological interpretation of high-throughput experimental data.

Main Methods:

• Developed an exact analytical expression for the expected number of false positives.
• Calculated adjusted p-values to control the false discovery rate.
• Implemented the framework in the freely available software package GOSSIP.

Main Results:

• Demonstrated the framework's capabilities using publicly available cell-cycle regulated gene microarray data.
• Analyzed the robustness of the method concerning gene group composition.
• Compared the framework's performance against earlier approaches.

Conclusions:

• The proposed statistical framework effectively identifies significantly enriched biological functions in gene groups.
• The method controls the false discovery rate, offering reliable interpretation of high-throughput data.
• The GOSSIP software package provides a practical tool for researchers to apply this framework.