Using Principal Components in Gene Expression Linkage Analysis
Author Information
Author(s): Elizabeth J Atkinson, Brooke L Fridley, Ellen L Goode, Shannon K McDonnell, Wen Liu-Mares, Kari G Rabe, Zhifu Sun, Susan L Slager, Mariza de Andrade
Primary Institution: Mayo Clinic College of Medicine
Hypothesis
Can principal components analysis improve the identification of linkage signals in gene expression data?
Conclusion
Principal components analysis did not improve the identification of linkage signals in this study.
Supporting Evidence
- PC analyses under-performed the single-probe analysis for known signals.
- The PC that best reproduced the single-probe analysis was primarily composed of that probe.
- No new signals were detected in the PC analysis.
- No new pleiotropic effects were detected in the PC analysis.
Takeaway
The researchers wanted to see if combining gene expression data into groups would help find genetic links, but it didn't work as well as looking at each gene individually.
Methodology
The study used principal components analysis on gene expression data from 196 participants to perform a genome-wide linkage scan.
Limitations
The principal components analysis under-performed compared to single-probe analysis and did not detect new signals.
Participant Demographics
196 participants from 14 three-generation pedigrees.
Statistical Information
P-Value
<10-9
Statistical Significance
p<10-9
Want to read the original?
Access the complete publication on the publisher's website