Using the gene ontology for microarray data mining: a comparison of methods and application to age effects in human...

Area of Science:

• Computational Biology
• Genomics
• Neuroscience

Background:

• Analyzing gene expression data requires contextualizing findings with existing gene relationship information.
• Age-related gene expression changes in the human prefrontal cortex are complex and require robust analytical methods.

Purpose of the Study:

• To compare the effectiveness of two computational methods, overrepresentation analysis (ORA) and functional class scoring (FCS), for analyzing gene expression data.
• To evaluate these methods in the context of age-related gene expression changes in the human prefrontal cortex.

Main Methods:

• Overrepresentation Analysis (ORA): Statistically evaluates the fraction of genes within a specific gene ontology class among those with age-related expression changes.
• Functional Class Scoring (FCS): Assesses the statistical distribution of individual gene scores across all genes in a gene ontology class without a prior gene selection step.

Main Results:

• Functional Class Scoring (FCS) produced more consistent results compared to Overrepresentation Analysis (ORA).
• The outcomes of ORA were highly dependent on the chosen gene selection threshold.
• FCS demonstrated greater reliability by considering all available genomic information.

Conclusions:

• Functional Class Scoring (FCS) is a valuable tool for analyzing complex gene expression datasets.
• Prioritizing comprehensive genomic data analysis over threshold-dependent gene selection enhances the reliability of findings.