I'm analyzing some Affy data and not sure how to interpret what appears to be a weird MA plot. In essence, more highly expressed genes tend to have positive foldChange and vice versa. I'm not super familiar with array analysis (just using ComBat + limma). Has anyone run across something like this before? Am I doing something wrong..?
EDIT: in response to @Devon and @ddiez comments below:
To explain a bit more, I have data from 3 timepoints (0D, 1D, 7D). The problematic comparison is 1D-0D, which corresponds to the first MA plot in the thumbnails along the left side of the attached image. The second and third thumbnails are 7D-0D and 7D-1D, respectively, and these seem to confirm that there is an enrichment of high expression values in the 1D samples. Looking at a density plot of the expression values, I see that indeed 1D is depleted for values ~ 6 (green peak is lower than others in main image). Is that really the root cause of this? My off-the-cuff guess would have been that this difference is negligible!