Is a compositional bias in Sequence Content of initial bases in WES reads normal?

Question

[gerrit.bostelmann]

Per Base Sequence Content

Any experienced Whole Exome Sequencing analyst/experimentalist:

the Per Base Sequence Content of the initial couple of bases from the 3' end of my WES reads have some type of compositional bias. Is this some "harmless" procedural or technical bias occurring during sequencing or is it something to be worried about for the downstream analysis?

Here is the graphic of the Per Base Sequence Content (FastQC)

https://ibb.co/GJqzm4b

[GenoMax]

If these were created using Nextera type (tagmentation) libraries then yes. See this paper (Figure 1).

[gerrit.bostelmann]

They were created on an Illumina HiSeq 2000 using an Agilent SureSelect Human All Exon v.2 kit. Not sure, but that might be a bit different.

[GenoMax]

What was used for creating the actual libraries for sequencing?

[gerrit.bostelmann]

I am sorry, that is the only information I have. I have no direct contact to the laboratory. I could try to find out though. Anything particular to ask about or consider?

Answer 1

[andres.firrincieli]

I am not a WES expert but that is pretty normal for RNA-seq data

[gerrit.bostelmann]

Yes, that is what I have found in various forums/sources too. Not sure as to what degree WES differs from that though, hence the question here :/