GA-HiSeq_p-values.txt

Created By Kyle Ellrott kellrott
19 COAD replicates were run on GAII and HiSeq by UNC. To compare the two platforms, we ran a two-tailed paired t-test on all the log-transformed genes. The resulting p-values were adjusted for multiple hypothesis testing by Bonferroni correction. The genes were then sorted by p-value in ascending order. The working hypothesis is that genes with the lowest p-values are contributing most towards platform based batch effects. Removing those genes from your analysis may potentially diminish platform effects, although that remains to be fully tested. We recommend removing those genes with a p-value < 0.05 (the top 1336 genes in the file out of 19666) for your cross platform analysis, if you?re going to combine samples from GAII and HiSeq.