Hi ,
I have tried to compare two tab delimited file contains different columns to find the match entries between two files but with my script i get nothing as output !
file 1 :
col1 col2 col3 col4 col5 col6
X chrN start end X X
file 2 :
col1 col2 col3
chrN start end
I want to compare the Col2-3-4 in file1 with Col1-2-3 in file2 and printout the matched rows in file1
Thanks in Advance,
this is my script but i got nothing as output!
use strict;
use warnings;
open(FILE1,"<21minus.bed");
open(FILE2,"<enhancers.bed");
open (OUT,">outputfile.txt");
my @arr1=<FILE1>;
my @arr2=<FILE2>;
close FILE1;
close FILE2;
my %chash;
for (@arr2){
chomp;
my($col1,$col2,$rest)=split(/\t/);
my $ckey="$col2$rest";
$chash{$ckey}=1;
}
for (@arr1){
chomp;
my($col1,$col2,$start,$end)=split(/\t/);
$start="" if(!$start);
$end="" if(!$end);
my $ckey="$start$end";
if($chash{$ckey}){
print OUT "$_\n";
}
}
6 answers
Perhaps the command line would be better/quicker?
cut -f2,3,4 file1.txt > ./newfile1.txt
cut -f1,2,3 file2.txt > ./newfile2.txt
diff newfile1.txt newfile2.txt > differences.txt
The exact options for the diff function are here.
Just an idea.
### In Response to Comment ###
If you save the below script as something.py then run as follows:
python something.py file2.txt file1.txt matches.txt
Your matches will be held in matches.txt. I just 'slapped' this together, but it should work fine if your files are not huge. Any issues, just drop me an email.
#! /usr/bin/python
import sys
inp = open(sys.argv[1], 'r')
file2 = [i.split() for i in inp.readlines()]
inp.close()
matches = []
inp = open(sys.argv[2], 'r')
file1 = inp.readline()
while file1:
file1 = file1.split()
if file1[1:4] in file2:
matches.append(file1)
file1 = inp.readline()
inp.close()
out = open(sys.argv[3], 'wt')
out.writelines('\t'.join(i) + '\n' for i in matches)
out.close()
•
written
by
Darren J. Fitzpatrick
Or you try:
cut -f2-4 file1.txt | cat - file2.txt | sort | uniq -c | perl -ane 'print if($F[0]>1);'
But it only works, if your entries in each file are unique.
•
written
by
David Langenberger
would something like this suit your idiom a bit more than the command line options given previously?
while($l1=<FILE1> && $l2=<FILE2>){
chomp $l2;
($f = $l2) =~ s/^(.*?\t.*?\t.*?\t).*$/$1/;
die if ! $f;
print $l1 if $l1 =~ /^$f.*$/;
}
I 'think' it should do what you want
EDIT: I don't think it does what you want
I dont know why none of commands and scripts work with my files ! nor my script!
I guess it's better to show you the head of files !
file 1 :
chr21 9411509 9411529 HWUSI-EAS1809:38:FC:4:46:12139:8182::1:N:0: 1 - 9411509 9411530 0,0,255 1 21 0
chr21 9417056 9417076 HWUSI-EAS1809:38:FC:4:86:7681:16695::1:N:0: 0 - 9417056 9417077 0,0,0 1 21 0
file 2 :
chr1 858256 858648 chr1:858256-858648 5 . 858447 858448 0,0,0 2 1,10 0,382
chr1 918449 918555 chr1:918449-918555 18 . 918504 918505 0,0,0 2 7,1 0,105
In these two files I tried to compare and find the match between columns 1-2-3 in file one and columns 1-2-3 in file two but I couldnt get the match output with my script and yours as well ! any other idea please ? I really stuck here for 3 days!
So its basically column 1,2,3 in file1 and column 1,2,3 in file 2. Is it??
I tried to use your advise Varun but i still have problem , then I made another easier script but I got and error again!
this is my new script!
#!/usr/bin/perl
use strict; use warnings;
open my $fh1, '<', 'file1'; open my $fh2, '<', 'file2';
while ( defined( my $line1 = <$fh1> ) and defined( my $line2 = <$fh2> ) ) {
chomp($line1);
chomp($line2);
my @values1 = split( " ", $line1 );
my @values2 = split( " ", $line2 );
if ( $values1[0] eq $values2[0] and $values1[1] eq $values2[1] and $values1[2] eq $values2[2]) {print "$values2[0] t $values2[1] t $values2[2] t n" ;}}
close $fh1 ; close $fh2 ;
the error is : Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 11.
Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 12.
Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 13.
Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 14.
Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 15.
Use of uninitialized value in string eq at ./match3.pl line 13, <$fh2> line 16.
you have the head of two files in previous posts!
is their away here you can post your 2 files here i mean upload it somewhere and we can download and try to help you.
Varun
Hi, these are sample files i guess, not the complete files????
Moreover your file1 has blank spaces at the end of last lines remove them
I don't think this is the right way to go about your problem. a) you have to catch empty strings before your if; b) [and possibly my fault] this will only return data if line M in file1 matches with line M in file2; c) you're splitting a tab-delim'd file on spaces, not tabs.
Yes Varun I just copy the head of the files as sample file cause they're too big to upload them, I checked the file1 there is no spaces at the end of it! russH I change the script but I dont know how to catch the empty strings before if as you mentioned, I am not really expert in Perl.
Thanks again
I am a perl one-liner. Check my command line:
$ cat 1.txt
1 chr1 100 200 blahblah
2 chr2 50 100 blahblah
$ cat 2.txt
chr2 50 100
chr3 10 1000
$ perl -e 'open(I1,$ARGV[0]);open(I2,$ARGV[1]);while(<I2>){chomp;@_=split;$a{join("\t",$_[0],$_[1],$_[2])}=1;}while(<I1>){chomp;@_=split;print $_,"\n" if exists $a{join("\t",$_[1],$_[2],$_[3])}}' 1.txt 2.txt
2 chr2 50 100 blahblah
Log in to answer this question.
-
Pattern Alignment And Report Identity
written by abhiwash •
-
Rearrange The Order Of Two Sequences According To Ids In 27000 Files
written by biolab
-
Find Nan Or Zero In One Column And Output Content Of The Second Column
written by robjohn7000
-
Ld Mapping Using A ~30G File With Perl Script
written by J.F.Jiang
-
Read 2 Files Perl
written by bioinformatics123 •
-
Find The Max And Min With Awk In Specific Range
written by Ar Es •
-
How To Check Whole Genome With Bigwigsummary ?
written by Ar Es •
-
Find The Match Entries In The Columns And Print Them Out By Perl
written by Ar Es •
-
Help Me To Count! With Perl
written by Ar Es •
-
Print Png Output Using Perl In Html
written by Rajesh •
Perhaps the command line would be better/quicker? cut -f2,3,4 file1.txt > ./newfile1.txt cut -f1,2,3 file2.txt > ./newfile2.txt diff newfile1.txt newfile2.txt > ./differences.txt
The parse differences.txt for your desired output. Just an idea.
Why are you discarding the chromosome id, and collapsing the coords into one contiguous number in generating the hashkey? I'd use foreach(@arr2){@F=split /t/, $; $ckey=join "-", @F[0..2]; $chash{$ckey}++} foreach(@arr1){@F=split /t/, $; $ckey=join "-", @F[0..2]; print $_ if exists $chash{$ckey}}
A plea to everyone. (1) Don't keep posting new questions as answers: edit your original question or use the comments or post links to code elsewhere if required. (2) Format code properly to make it readable (indent lines with 4 spaces).
It was pointed out at your re-post of this question that really, it is a basic Perl programming problem rather than a bioinformatics question. So in fairness, we have to close this as off-topic also. There are an awful lot of very basic errors in the code; it's not really fair for others to work through and fix them. I suggest you spend some time improving your basic skills in Perl - but feel free to keep asking questions with more relevance in bioinformatics.