Why do I have sequential 100 G (GGGGGGGGGG...) in R2 FastQC report?

Question

[Giulia.cosenza]

We did a shotgun paired-ended sequencing (200 cycles, 100 forward and 100 reverse) on a sample and than we took off the adapters and checked the fastQC report. R1 is quite good and doesn't show consistent problems, but R2 seems to still have adapters and there are a lot of overrepresented sequence of sequential 100 G (GGGGGGGGGG...), since its length the repetition is not just at the 3' but it characterized the entire fragment. How is this possible if besides R1 doesn't show sequential G stretches ? Any idea?

[Arup Ghosh]

Please include details related to the sequencing instrument/chemistry and how the adapter was trimmed.

Answer 1

[rpolicastro]

G is absense of color in illumina sequencing, so it's likely the machine lost track of the clusters/spots when switching from R1 to R2 read sequencing.

[GenoMax]

In addition to what rpolicastro said, if every read is poly-G in read 2 part of your run then there is a hadrware/software/reagent problem with the run. Your sequencing provider should have investigated this with Illumina instead of releasing the data.

[Giulia.cosenza]

No, not every R2, you can see the % in the picture I posted

[rpolicastro]

That's still a large percentage, so it may be worth talking to the sequencing provider.