FastQC Results -- Per Sequence GC Content and Kmer Contents

Question

[haiying.kong]

I have 32 fastq data from whole exome sequencing. I ran FastQC on them, and the following 3 items have poor results.

Per Sequence GC Contents: Failed on all samples.

Kmer contents: 18 out of 32 samples failed, and the rest 14 samples have warnings.

Sequence_Duplication_Levels: 10 out of 32 samples failed, and the rest 22 samples have warnings.

What could be the problem? What is disadvantage of having these qualities failed if we use these day to identify somatic mutations?

A company performed library building and sequencing. Should we stop using this company?

I do not know what is wrong with the website. I cannot upload figures any more.

Answer 1

[Atu]

I think you should assess carefully the quality of your data, and if all the other modules are OK a think data is still valuable. It's a pity you can't show graphs. By post-mapping duplicate removal you could deal with the issue of duplications, and about the other two... How was DNA fragmented, by means of enzymes or abiotic methods? I think enzymatic digestion by transposases may introduce some bias regarding the other two modules that failed.