Hi,
I wonder what the effect is of running a tophat..cuffdiff analysis with single-end read groups of quite different size.
One of our library/sample is 3x larger in read count than the others. Will this bias the analysis or is cuffdiff normalizing for this effect from the total read number in each sample?
The risk being that low-expressed genes appear in the larger lib and are not represented (yet) in the smaller libs.
Thanks for your opinion.
Stephane
I wonder what the effect is of running a tophat..cuffdiff analysis with single-end read groups of quite different size.
One of our library/sample is 3x larger in read count than the others. Will this bias the analysis or is cuffdiff normalizing for this effect from the total read number in each sample?
The risk being that low-expressed genes appear in the larger lib and are not represented (yet) in the smaller libs.
Thanks for your opinion.
Stephane