How to figure out adapter sequence for Illumina reads?

Question

[mmrcksn]

I have some paired-end sequencing reads, 300 bp long. The adapter content is too high for both runs according to FASTQC. However, the only overrepresented sequences that show up in FASTQC are polyA and polyT. Why would the adapter sequence not show up as an overrepresented sequence? And is there an easy way to figure out the sequence of the adapter that is contaminating my reads?

Thanks!

Answer 1

[GouthamAtla]

Fastqc looks for a set of standard illumina adapters. Given in the configuration directory of source code. If the adapters used for your data are not listed in those files, it would not be shown in the overrepresented sequences.

In short, there could be no adapters, but polyA or polyT in your data. Or check if the adapters used for your samples are listed in the list of adapters of fastqc.

Answer 2

[GenoMax]

If you are using standard illumina adapters then the sequences for all standard adapters are available in "resources" directory of BBMap suite. BBduk.sh is the program you would want to use to scan/trim adapters.

If you are not using standard illumina adapters and you have paired end reads (and there are enough reads with short inserts) you can identify possible adapters by running the BBMerge program from BBMap suite.

$ bbmerge.sh in1=r1.fq in2=r2.fq outa=adapters.fa

[John]

I would be interested in the latter option genomax2 :)

[GenoMax]

See the update above.

[John]

Ah, yeah i tried that recently but my reads didn't overlap enough to call adapters :(
Good tool/idea though.

[GenoMax]

Can you try this (replace the N with expected length of your adapter) to see if it works?

$ commonkmers.sh in=reads.fq out=kmers.txt k=N count=t display=999