FastQC per sequence GC content

Question

[jhy]

I have a question when interpreting the FastQC result. I got a failure from "per sequence GC content" result, so I'm not sure if it's okay to use it.

Even when trimming with average quality 20, minimum length 20, and adapter sequence, failure also appeared. (I used Trimmomatic.)

This data is human WES data, and only chromosome 17 and 18 are cut. Before cutting the specific region, I only got a warning in "per sequence GC content". Did the use of only certain chromosomes affect it?

And could you tell me how to solve the failure in "per tile sequence quality" and "per sequence GC content"?

Answer 1

[shelkmike]

The problem with "per tile sequence quality" may mean that the flow cell used in the sequencing was not very good and some of its tiles produced reads with low quality.
Regarding the GC content: did you try to look at the reads? Maybe you see GGGGGGGGGG tails in some reads or other abnormalities?
Also, are you sure that you used correct adapter sequences when you performed trimming? If your reads are paired-end, you can try the program fastp. For paired-end reads it is able to automatically guess and remove adapter sequences.

[jhy]

I'll check the reads and fastp. Thank you!