Illumina Adapter trimming

[Ramet]

Dear all,

when trimming Illumina TrueSeq paired-end reads with fastx_toolkit, is it correct to

- trim the IndexAdapter from the forward reads
GATCGGAAGAGCACACGTCTGAACTCCAGTCAC‐NNNNNN‐ATCTCGTATGCCGTCTTCTGCTTG

- trim the reverse complement UniversalAdapter from the reverse reads
AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT

?

I know this sounds trivial and you may think this has been answered in other threads, but actually I couldn't find a definite conclusion.

So please, just a short affirmation or correction.

Thank you very much in advance!

[Ramet]

Come on, 95 views and no answer?

Or maybe you are playing safe and trim both forward and reverse reads with both adapters, each regular and reverse complement?

Please answer.
Every help is highly appreciated!

[FiReaNG3L]

I can tell you that the FASTX trimmer is very aggressive and has been built with another purpose in mind - personally I use cutadapt. FASTX was trimming after it recognized 5 bases out of the adapter.

[Ramet]

Hmm, I have read this already. However, I specify to only cut if at least 15 bases are matching the adapter string. But I will also try cutadapt!

Certainly my question was "In which direction/orientation is the Universal Adapter present in my reverse reads?".
Any suggestions? I can't believe nobody is aware of it...

[GenoMax]

Certainly my question was "In which direction/orientation is the Universal Adapter present in my reverse reads?".
Any suggestions? I can't believe nobody is aware of it...

See if this helps: http://genomics.med.tufts.edu/docume...q_Adapters.pdf

[fkrueger]

I don't have a schematic drawing handy of the how exactly the adapters look on either side, but in the end both sides of the read will start with the same part of the adapter sequence "AGATCGGAAGAGC" (for this you also need to consider A-tailing on the 3-end of the reads) before they start forking off with a different sequence for the first and second reads. We have written a small wrapper around Cutadapt called Trim Galore, which will in its default setting trim off Illumina (including TruSeq) adapters from both reads of paired end read. The usage would be as simple as writing:

trim_galore file1.fq file2.fq

hth

[Ramet]

Hi all!

I have tried "grep" with a substring of the Universal Adapter sequence and its reverse complement. I got 10 hits for the regular sequence and 10.000 hits for the reverse complement of the published sequence. I guess it is quite clear now...

But thanks for all responses!

[Tobikenobi]

I don't have a schematic drawing handy of the how exactly the adapters look on either side, but in the end both sides of the read will start with the same part of the adapter sequence "AGATCGGAAGAGC" (for this you also need to consider A-tailing on the 3-end of the reads) before they start forking off with a different sequence for the first and second reads. We have written a small wrapper around Cutadapt called Trim Galore, which will in its default setting trim off Illumina (including TruSeq) adapters from both reads of paired end read. The usage would be as simple as writing:

trim_galore file1.fq file2.fq

hth

Does it consider all indexed truseq adapters?
Thank you!

[fkrueger]

Yes, they all start with the same sequence.