Quick Interpretation of FastQC output

[rnastar]

Dear all,

I have attached a plot of the average per-base quality of reads in my sample.

It appears that surprisingly the 5' ends of my reads have lower quality on average than the 3' end.

Is this indicative of the fact that I need to trim my reads on the 5' end? There does not appear to be any detected adapter contamination. Any thoughts?

Thank you!

[GenoMax]

Nothing to worry about. This looks like a very typical of illumina run. There should be no need to trim from 5'-end.

[rnastar]

Thanks! Any idea on why we see the drop off in quality on the 5' end, or what could be causing it?

[GenoMax]

It would be technically correct to call it a drop-off .. if you compare it to the rest of the plot. But you are still above Q30 for all bases which is excellent quality data.

As RTA ramps up in the first 10-12 cycles it is calculating different metrics (color matrix, phasing etc).

[FWOS]

All Illumina HiSeq reads start out lower. Mostly this is improved after template generation (cycle 4). This is partially due to lower quality bases that result from the prep process, and it is also partially do do with how the per-base sequence quality is assessed. Once template generation is complete and the phasing / pre phasing data are taken into account the overall Q scores should hover above Q30 until the read length becomes a factor on per base sequence quality.