Related Experiment Video
Updated: Jun 23, 2026
10:40
Comprehensive Workflow for the Genome-wide Identification and Expression Meta-analysis of the ATL E3 Ubiquitin Ligase Gene Family in Grapevine
Published on: December 22, 2017
A global meta-analysis of microarray expression data to predict unknown gene functions and estimate the
1Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation;, 825 N.E. 13th Street, Oklahoma City, OK 73104-5005, USA. jdwren@gmail.com
Bioinformatics (Oxford, England)
|May 19, 2009
Summary
A global meta-analysis of human gene expression data predicted gene functions with 34% accuracy using co-expression patterns. This approach uncovers hidden genetic knowledge in expression databases for thousands of uncharacterized human genes.
Area of Science:
- Genomics
- Bioinformatics
- Systems Biology
Background:
- A significant portion of human genes lack documented functions.
- Gene function prediction is crucial for understanding mammalian biology.
- The rate of new gene name submissions to literature has decreased.
Purpose of the Study:
- To identify genes with reproducible co-regulation patterns across diverse experimental conditions.
- To develop and compare methods for predicting gene function based on co-expression.
- To leverage expression databases for discovering novel gene functions.
Main Methods:
- Conducted a global meta-analysis (GMA) of 3551 human microarray datasets from GEO.
- Analyzed patterns of parallel and anti-parallel gene co-expression.
- Compared ranking methods for predicting gene function using top co-expressed gene pairs.
Main Results:
- The best prediction method achieved 34% exact match for Gene Ontology (GO) categories compared to 3% for random sets.
- Identified co-expressed gene pairs, with only 2.4% found in MEDLINE.
- 1642 human genes with unknown functions were found to be differentially expressed in at least 30 experiments.
Conclusions:
- Genes co-expressed in parallel are often associated with the same GO categories.
- Analyzing parallel and anti-parallel co-expression separately yields more significant results.
- Expression databases hold substantial, underexplored genetic information, including functions for uncharacterized genes.
