Impact of Gene Expression Data Pre-Processing on eQTL Mapping
Author Information
Author(s): Aurelie Labbe, Marie-Paule Roth, Pierre-Hugues Carmichael, Maria Martinez
Primary Institution: Université Laval, Canada
Hypothesis
How do different pre-processing methods for microarray data affect eQTL mapping results?
Conclusion
Different pre-processing methods for microarray data lead to varying results in eQTL mapping, with significant discrepancies observed based on the selection criteria of expression traits.
Supporting Evidence
- More linkage signals were found for the most heritable traits than for those with the highest variance.
- GCRMA detected more linkage signals than the other methods, especially in the set of non-expressed traits.
- Concordance rates between methods were higher for cis-regulating loci compared to trans-regulating loci.
Takeaway
This study looked at how different ways of preparing gene data can change the results when trying to find which genes affect traits. It found that the method used can make a big difference.
Methodology
The study analyzed gene expression levels in lymphoblastoid cells from 194 individuals using three pre-processing methods (MAS5, RMA, GCRMA) and performed linkage analysis on selected traits.
Potential Biases
Potential biases may arise from the choice of pre-processing methods and the selection of traits for analysis.
Limitations
The true genetic determinants of the studied traits are unknown, which limits definitive conclusions about the best pre-processing method.
Participant Demographics
The study involved 194 individuals from 14 three-generation CEPH Utah families.
Want to read the original?
Access the complete publication on the publisher's website