What proportion of 'bad' quality reads are expected using HiSeq 2000 for RNA-Seq

[bob-loblaw]

I am just after processing 24 files, using trimmomatic to quality filter them. On average was 30% of the reads were bad quality (SD = 11).

This seems somewhat high to me, I was just wondering if anyone else had experience with this and could tell me what would be expected? If I should be particularly concerned about this...

Thanks.

[westerman]

30% is indeed high. We -- I work at a sequencing center -- have seen such results but in these cases the problem is tracked down to sequencer problems, sample prep, or just an random problem. We almost always re-run the samples. Does fastQC show any particular place where the problem occurs?

[GenoMax]

Hopefully these samples did not come from an overloaded flowcell. (Assuming that you have taken into account the right Q-score parameters for trimmomatic)