Splitting large fasta file according domains in HUMAN

Hi all,

I have a FASTA file which contains protein domain sequences (Pfam-A.fasta) in different species, and need to split the file into multiple FASTAs, one domain in HUMAN per file. What's the best way to go about this? Ideally each file will be named with the name of the domain. I am using Perl.

Example of the sequence in fasta file:

>B4DW62_HUMAN/123-274  PF00198.17;2-oxoacid_dh;
WDGEGPKQLPFIDISVAVATDKGLLTPIIKDAAAKGIQEIADSVKALSKK
ARDGKLLPEEYQGGSFSISNLGMFGIDEFTAVINPPQACILAVGRFRPVL
KLTEDEEGNAKLQQRQLITVTMSSDSRVVDDELATRFLKSFKANLENPIR
LA
>D2HF00_AILME/272-501  PF00198.17;2-oxoacid_dh;
PGTFTEIPASNIRRVIAKRLTESKSTVPHAYATADCDLGAVLKARQSLVR
DDIKVSVNDFIIKAAAVTLKQMPDVNVSWDGEGPKQLPFIDISVAVATDK
GLITPIIKDAAAKGVQEIADSVKALSKKARDGKLLPEEYQGGSFSISNLG
MFGIDEFTAVINPPQACILAVGRFRPVLKLEQDEEGNARLQPHQLITVTM
SSDSRVVDDELATRFLENFKANLENPIRLA

Many Thanks

Hanadi

sequence

• 2,556 views • link

updated 3.7 years ago by Ram

• written 11.1 years ago by hb273 •

Can you post your perl code?

• 0 views • link

updated 3.7 years ago by Ram

• written 11.1 years ago by PoGibas

I am a new in Perl.

I used this script to split the file

#!/usr/bin/perl
use warnings;
use 5.12.4;
use File::Basename;

my $file = "Pfam-A.fasta"; #enter name of your FASTA file here
my $record_per_file = 4;    #Enter how many record you want per file / chunk size
my $file_number = 1;    # a part of your new file names.
my $counter = 0;    #counts number of records

open (FASTA, "<", "$file" ) or die "Cannot open file $file $!";

while (<FASTA>) {
    if (/^>/) {
        if ($counter++ % $record_per_file == 0) {
            my $basename = basename($file);
            my $new_file_name = $basename. $file_number++ . ".fasta";
            close(NEW_FASTA);
            open(NEW_FASTA, ">", $new_file_name) or die "Cannot open file $new_file_name $!";
        }
    }
    print NEW_FASTA $_;
}

• 0 views • link

updated 3.7 years ago by Ram

• written 11.1 years ago by hb273 •

0 answers

No answers yet.