Hi,

I have to compare the GC bias about two short-reads sequencing runs (Illumina & Aviti) on the same sample, to check if there is any difference between both technologies.

First, I mapped my paired-end reads on a reference genome using bwa. It goes well for both technologies, with >99% of the reads mapped on it.

Then, I used computeGCBias. I got :

• Aviti

• Illumina

As I understand it, with the "reads per regions of 300bp" graph, more the GC fraction is high, more we get mapped reads, for both techno. Am I right?

With the log2ration graph, I would say when we are below 0 , more we have mapped reads (it is the contrary). And then, more the GC% is high, more we get mapped reads? I do not understand why we are below 0 for the Illumina tech, for a low GC% ? We should be >0 as the "300bp" graph shows it?

Best