adaptor/barcode trimming that are NOT in the 5' or 3' end

[Iria]

Hi,

Is there any tool available that allows to trim adapter (and primers) that are inside of the reads (not in the 5' or 3' ends)?

For whatever reason a portion of my reads have a variable number of random bases before the adapter and primer. I need a tool that is able to trim the adapter+primer and any bases before it (this is between the 5'end and the adapter). I need to do the same for the 3'ends, where I have the same issue.

This is an example of such a read. In red are the adapter and primer sequences. After trimming, only the lower case portion should remain:

>made-up example
ATCGTAGTATTAGGATCGTCGCTGATGACTGAatgatgatagttacctagagcta[...]cgacatcacgaccccgatagaATAGCTAGCTATCGATCGATCTGACATAGACA

A custom script would do it as well. I tried using the bash tool grep, but wasn't able to make it work.

These sequences are amplicons generated using barcoded COI primers in PacBIO (with the Conserved Consensus Sequencing protocol). When I say adapter I mean the barcode that was added to the primers to be able to split the libraries.

Thanks in advance for your answers.

Iria

[dietmar13]

Agrep

agrep, a fuzzy grep version:

http://en.wikipedia.org/wiki/Agrep

best implemeted in TRE:
http://en.wikipedia.org/wiki/TRE_(computing)

[relipmoc]

Skewer

Suppose the content of test.fasta is:

>made-up example
ATCGTAGTATTAGGATCGTCGCTGATGACTGAatgatgatagttacctagagctaGAGACCGCCT
GGGAATACCGGGTGCTGTAGGCTTTGGAATTCcgacatcacgaccccgatagaATAGCTAGCTA
TCGATCGATCTGACATAGACA

and you use the following command:
$ cat test.fasta | skewer -x ATAGCTAGCTATCGATC -e 3 - -1 2>/dev/null | skewer -x ATTAGGATCGTCGCTGATGACTGA -e 5 - -1 2>/dev/null

you will get the following result:

>made-up example
atgatgatagttacctagagctaGAGACCGCCTGGGAATACCGGGTGCTGTAGGCTTTGGAATTCcgacatcacgaccccgataga

where ATAGCTAGCTATCGATC is the 3' end adapter sequence; ATTAGGATCGTCGCTGATGACTGA is the 5' end adapter sequence.

See http://seqanswers.com/forums/showthr...ghlight=skewer for more details.