Overrepresented sequences poly(C) followed by poly(T)

Question

[Jakesaunders]

I'm working on an RNA-seq project and fastqc keeps identifying overrepresented sequences consisting of poly(C) followed by poly(T). I see a range from

CCCCCCTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT to CCCCCCCCCCCCTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT

I know the poly(T) is probably from how the RNA was enriched bu where is the poly(C) coming from? Has anybody else seen this before?

Answer 1

[Kevin Blighe]

There have been a few threads on this topic already:

• How to remove poly T in RNA-sequencing data
• Statistics About Poly-A Tails In Rnaseq Reads
• https://www.researchgate.net/post/Is_it_necessary_to_remove_over_represented_sequences_in_the_NGS_data
• Fastqc Over-Represented Sequences Are Adapters ?

In conclusion, I would just remove the standard adapters that are known to CutAdapt (or whatever program that you are using) from the sequences, and also filter / trim reads based on length and quality, and then proceed to alignment. My feeling is that the main thing that is affected by trimming and filtering reads is the quality metrics like percent alignment. Most 'junk' reads, including poly A and T, will not align anyway.

Kevin