replacing words using perl script, excel as reference

Question: replacing words using perl script, excel as reference

3.4 years ago by

peacezah • 10 wrote:

I have this list, how do I replace sequence id from the old name (left column) to new name (right column). i have been thinking about the condition like this; if found word in column left, will then replace with word in column at the right. any idea by using perl script?

>NC_006351.1_00512 76172 76077 len=96
GTGACGCTGCCCGTCGGCGCCTTGTCGAAAGGCGCGAGCTTCGAAGTCGGCGCGCAGGTC
CAGCGGCCGACCGGCGCGCTGGCGTTGTTCGAGTAA
>NC_006351.1_00969 110672 110583 len=90
GTGTCGGCGAAAAACGACACGTTCTCGCGCCTCGGCAGCCGCGACGCGCACGAAGGCCGA
CAAAACACGCCGGTCGTCTTGACCGCGTAG
>NC_006351.1_01005 116974 117090 len=117
TTGCTCGTCGGGCGGATCATGCCGACGCCCGAAGCCGAATCCGAATCCGAATCCGAATCC
GACGCCGACGCCGAGGCGCAGAAGCGCTTCGCCGGGCTGCGCTACACGGGCACGTAA

NC_006351.1_512	BPSS_001
NC_006351.1_969	BPSS_002
NC_006351.1_005	BPSS_003
NC_006351.1_178	BPSS_004

perl script fasta • 1.4k views

ADD COMMENT • link •

modified 3.4 years ago by biolab • 990 • written 3.4 years ago by peacezah • 10

duplicate of:

replace fasta headers with another name in a text file

Change Fasta File Header

ADD REPLY • link written 3.4 years ago by Pierre Lindenbaum ♦ 106k

3.4 years ago by

biolab • 990

biolab • 990 wrote:

I give you an example, you can try the script. It works on my PC. If not working, please show error message. Hope it ok.

FASTA File:

>N1   110672 110583  len=90
GTGTCGGCGAAAAACGACACGTTCTCGCGCCTCGGCAGCCGCGACGCGCACGAAGGCCGA
>N2   116974 117090  len=117
TTGCTCGTCGGGCGGATCATGCCGACGCCCGAAGCCGAATCCGAATCCGAATCCGAATCC

LIST File:

N1   BPSS1
N2   BPSS2

Script:

#!/usr/bin/perl
use strict;
use warnings;
print "Usage: perl $0 listfile fastafile\n" and exit unless $ARGV[1];

open my $list, '<', $ARGV[0];
my  %h = map { s/\r//; split /\s+/ } <$list>;
close $list;

open my $fasta, '<', $ARGV[1];
while(<$fasta>) {
    next if /^\s+$/;
    s/\r//;
    chomp;
    /^>(\S+)\s+(.+)/ ? print ">$h{$1} $2\n" : print "$_\n";
}
close $fasta;

ADD COMMENT • link modified 3.4 years ago • written 3.4 years ago by biolab • 990

I like your approach, but the use of global variables is quite dangerous and is discouraged. You can type perldoc -f open to see recommendations for opening files (or see the web link), and I recommend putting use strict and use warnings in every script (those are enabled in recent versions of Perl, which will generate warnings from this code).

ADD REPLY • link written 3.4 years ago by SES ♦ 8.1k

Hi, SES, thanks for your suggestions. I have modified the script. I have a question about the disadvantage of bareword filehandle: if I use IN as filehandle, later on I accidentally use IN again, this will generate error. Is the case the only disadvantage of bareword filehandle? I suppose there exist other dangerous cases, could you briefly make an example? Thanks for your comments.

ADD REPLY • link written 3.4 years ago by biolab • 990

The code is much improved, thanks! The main problem with bare file handles is they have global scope. If you use IN at the top of your script you will be able to use that file handle anywhere, and this can happen by accident. The problem with

open IN, $ARGV[0];

is that you didn't specify the mode, so it is possible to say print IN "some text\n"; and now your data is gone with no warnings! It is also a good practice to write or die ... after the open statement to make sure the file can be opened for reading or writing. A nice convenience to avoid writing this is to just put use autodie; at the top of your script and all these tests will be handled for you (here is a link about autodie).

ADD REPLY • link modified 3.4 years ago • written 3.4 years ago by SES ♦ 8.1k

Thanks a lot SES. Your comment is really helpful!

ADD REPLY • link written 3.4 years ago by biolab • 990

Hey!thanks..it works but there`s still problem. the

110672 110583  len=90

behind all ID were disappeared.

I just want to replace the id itself. not including the location and length..

by the way, thank you

ADD REPLY • link written 3.4 years ago by peacezah • 10

Hi peacezah, I have modified the script.

ADD REPLY • link written 3.4 years ago by biolab • 990

3.4 years ago by

PoGibas ♦ 4.7k

Vilnius

PoGibas ♦ 4.7k wrote:

See use python to change the header of a fasta file based on a dictionary in another file by tangming2005.

You only need awk example:

awk -f foo.awk dict.dat user.dat

NR == FNR {
  rep[$1] = $2
  next
} 

{
    for (key in rep) {
      gsub(key, rep[key])
    }
    print
}

ADD COMMENT • link modified 3.4 years ago • written 3.4 years ago by PoGibas ♦ 4.7k

it doesn`t help

ADD REPLY • link written 3.4 years ago by peacezah • 10

why? Can you tell what error it gives?

ADD REPLY • link modified 3.4 years ago • written 3.4 years ago by PoGibas ♦ 4.7k

3.4 years ago by

roy.granit • 640

Israel/LabWorm

roy.granit • 640 wrote:

I would use the left column as hash keys and the right as the hash items. Then parse each row with a '>' ,separate using the spaces and make the replacement.

ADD COMMENT • link written 3.4 years ago by roy.granit • 640

3.4 years ago by

biolab • 990

biolab • 990 wrote:

I give a perl solution. I noticed in fasta file one id is NC_006351.1_00512 76172 76077 len=96, whereas in following list file you change it to NC_006351.1_512. Is it right?

#!/usr/bin/perl
open LIST, $ARGV[0];
%h = map { split /\s+/ } <LIST>;
close LIST;

open FASTA, $ARGV[1];
while(<FASTA>) {
    next if /^\s+$/;
    chomp;
    if (/^>(\S+\_)\S{2}(\S{3})\s+/) {
        $id = $1 . $2;
        print ">$h{$id}\n";
    } else {
        print "$_\n";
    }
}
close FASTA;

Usage: perl changeID.pl listfile fastafile

ADD COMMENT • link modified 3.4 years ago • written 3.4 years ago by biolab • 990

Sorry. I edited the ID before posted it here. But what I wanted to do is still the same.

I did what have you suggest but surprisingly the whole id lost

ADD REPLY • link written 3.4 years ago by peacezah • 10

Please log in to add an answer.

Similar posts • Search »