Hi folks,
I am doing QC on a batch of ribo-depleted RNA-seq samples (human) and am seeing something strange in the GC content histogram. There appears to be spikes interspersed in the GC% histogram. I thought this could be due to un-depleted rRNA sequences, but I have tried to remove reads mapping to human rRNA sequences, and this artefact remains. Has anyone else seen something like this?
•
written
by
wengkhong •
Given the regular spacing of these spikes, it looks to me like it is an algorithmic artifact caused by the read length and number of bins. Try changing the number of bins to exactly the read length.
You're right. It was some issue with the binning. I plotted the underlying data, which was binned at exactly the read length and it looked perfectly normal. Thanks!
Log in to answer this question.
It's clear from the histogram that a duplication was occurred, but the significance of this observation is correlate with data.Can you add details, length of sequence an the method used: as a stream or with fixed windows.