The drop in correlation coefficients of expression measurements across platforms from genes impervious to alternative splicing based differences to those genes most susceptible to alternative splicing based differences seems to indicate alternative splicing plays a role in platform discordance.
Unfortunately none of these studies are directly applicable to one of the most commonly used human transcriptomic microarray platforms, namely Affymetrix Gene 1.0 ST arrays, either because they do not use a random priming approach or because the design of the microarray platform differs substantially from Gene 1.0 ST arrays.
There were 194 genes out of the 228 represented in both platforms, from which four genes were common to both analyses: AQP1, DUSP6, FOS and TRIM24.
To minimize the variability of different platforms and inconsistency that arise from gene ID mapping, we extract expression profile data collected from only two Affymatrix platforms (GPL339 and GPL340).
It is now possible to address significant evolutionary genomics questions by applying high-throughput sequencing to discover the majority of genes for ecologically tractable species, and by subsequently developing microarray platforms from which to investigate gene regulatory networks that function in natural systems.
In the companion article, we emphasize effective experimental design based on the in vivo physiology of the rhesus macaque, whereas this article emphasizes considerations for gene annotation and data interpretation using gene microarray platforms from Affymetrix®.
Data mining of 4 to 15 microarray datasets from the Oncomine platform for genes overexpressed in six human cancers compared with their expression in normal tissues led to the identification of a list of 3,064 to 19,645 upregulated gene expression profiles per cancer type.
For gene sets developed from different platforms, gene identifiers in the gene set were mapped to the cDNA clones using the following linkers: UniGene, gene symbol and gene alias, where UniGene had the highest priority while gene alias had the lowest priority.
Our study shows that although there is variability between the platforms, the gene expression profiles emerging from using all three technologies are highly correlated to the biological variation in the data and the same tumor subtype pattern was identified with all three methods.
We used the predictor to reclassify the 76 MFH and the 10 NOS samples into the 6 subtypes after mapping the predictor genes across different platforms (from cDNA to Affymetrix U133) as above.
