These methods often showed extremely high agreement in top fold change and (especially Li-Wong) p value lists.
There was excellent agreement in the top fold change lists of genes differentially expressed for all three platforms, again greater than for the p values.
We analyzed the top fold-change I/U-DEGs, 100 Upr and 100 Dnr that were significant (adjusted p-value < 0.05), in order to compare nearly equal numbers of I/U-DEGs, since I/U-DEGs with >2 fold-change were so few.
Second, a study [ 14] comparing early gastric cancer (EGC) with advanced gastric cancer (AGC) - roughly separating stages I and II - also identified a similar differentially expressed gene list of which again COL11A1 (probe 37892_at) was at the top (log fold change: 4.26) followed by COL10A1 and FAP.
For each cluster group, transcripts that were differentially expressed more than twofold relative to C early are listed with fold changes, top BLASTX and tBLASTX hits (E <1e-5), Gene Ontology identifiers and significant (E <1e-5) proteinotein domain matches in additional file 1.
As shown in Table 2, both MMP11 and HPSE2 were found on the top 28 genes list with fold changes ≥ 10.
Not like the top obviously expressed genes with fold changes close to 3.0, the SMAD group showed the relatively lower differential expression most of them had slight positive changes less than 0.1.
Among 273 upregulated genes in ACs vs TCs, the top 10 scoring the highest fold changes (ranging from 38.766 to 534.063, ACs vs TCs) were selected and the absolute expression values for each sample were considered.
Successively inclusive sets of top-DEGs (based on absolute log fold changes of mean expression) or top-DCPs (based on absolute dC values) were chosen at three increasing levels: 0.1%, 0.5%, or 1%.
Since a t-test was used for microarray data sets on top of a fold change cutoff to define differentially expressed genes, fewer genes with lower expression levels and therefore higher variability will pass the t-test cutoffs.
