Meta-analysis of Microarray Data on Muscle Development and Disease
Author Information
Author(s): Jelier Rob, 't Hoen Peter AC, Sterrenburg Ellen, den Dunnen Johan T, van Ommen Gert-Jan B, Kors Jan A, Mons Barend
Primary Institution: Erasmus MC University Medical Center, Rotterdam, The Netherlands
Hypothesis
Automatic association of genes with biological processes through thesaurus-controlled mining of Medline abstracts would be more effective than manual curation.
Conclusion
The literature-based association analysis can uncover hidden biological connections in microarray studies without needing raw data analysis.
Supporting Evidence
- LAMA retrieved many more biologically meaningful links between studies compared to traditional methods.
- The algorithm correctly grouped studies on muscular dystrophy, regeneration, and myositis.
- LAMA identified new associations, such as linking cullin proteins with muscle regeneration.
Takeaway
This study found a better way to connect different research on muscle diseases by looking at related genes in literature instead of just comparing raw data.
Methodology
The study developed an algorithm (LAMA) to compare gene expression studies based on literature-derived gene associations.
Potential Biases
The study may be biased towards studies with similar technical backgrounds or methodologies.
Limitations
The analysis relies on simulations which are computationally intensive and may not capture all associations.
Participant Demographics
The compendium includes studies on human (N = 37), mouse (N = 51), and rat (N = 13).
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website