PRINSEQ -derep option

[StephaniePi83]

Hi,
i would like to remove sequences that are identical with the 5' or 3' end of a longer sequence.
Here is an example of what i would like to do :
INPUT :

>pi1
AAAAAAAAAATTAAGGGCCAGCTGA
>pi12
AAAAAAAAAATTAAGGGCCAGCTGAA
>pi13
AAAAAAAAACTTGAACTCTACTGC
>pi14
AAAAAAAAATTAAGGGCCAGCTGAA
>pi15
AAAAAAAAATTTTGGATGATCTTAAT
>pi16
AAAAAAAAATTTTGGATGATCTTAATT
>pi17
AAAAAAAACAAGGTCGGCATAAAG
>pi18
AAAAAAAACGAACATGAGAGGATGGA

OUTPUT :

>pi12
AAAAAAAAAATTAAGGGCCAGCTGAA
>pi13
AAAAAAAAACTTGAACTCTACTGC
>pi16
AAAAAAAAATTTTGGATGATCTTAATT
>pi17
AAAAAAAACAAGGTCGGCATAAAG
>pi18
AAAAAAAACGAACATGAGAGGATGGA

I try to solve my problem with PRINSEQ, with the following comand line, but it did'nt work, it only remove reads that have the exact same sequence

perl prinseq-lite.pl -verbose -fasta tmp1.fa -derep 123 -out_format 1

Someone familiar with this tool can help me ?
Thanks in advance

[SES]

Hi StephaniePi83, the "1" in -derep 123 means to only remove exact sequences. I would try to remove the "1" and go with only 2 (2 means 5' duplicate, 3 means 3' duplicate). I'm not completely certain this will do what you want but it is easy to test with a small data set (like the example you pasted above). If that doesn't do it, then I would email the author. I've emailed him before and he responded.

[StephaniePi83]

Hi SES,
I already try this; i think i'll email to the author as you suggest me

[StephaniePi83]

I email the author. There was an issue with the program that didn't tricker the 5'/3' removal. He gives me the latest version with the fix !