Shell Programming and Scripting

fragmentID	chromosome	start	end	HEL25E	TRIP1
r5GATC2L00037	chr2L	5301	6026	0.03	0.036
r5GATC2L00038	chr2L	6023	6882	-0.025	-0.041
r5GATC2L00040	chr2R	6921	7695	-0.031	0.005
r5GATC2L00042	chr2R	7715	8554	-0.006	-0.024
r5GATC2L00043	chr3L	8551	8798	0.042	0
r5GATC2L00044	chr3L	8795	9696	0.025	0.042
r5GATC2L00047	chr3R	9801	12443	-0.028	-0.027
r5GATC2L00048	chr4	12440	12695	0.004	0.08
r5GATC2L00050	chrX	12713	13068	0.01	0.1
r5GATC2L00051	chrX	13065	13295	0.137	-0.009

chr2L	5301	6026	0.03
chr2L	6023	6882	-0.025
chr2R	6921	7695	-0.031
chr2R	7715	8554	-0.006
chr3L	8551	8798	0.042
chr3L	8795	9696	0.025
chr3R	9801	12443	-0.028
chr4	12440	12695	0.004
chrX	12713	13068	0.01
chrX	13065	13295	0.137

chr2L	5301	6026	0.036
chr2L	6023	6882	-0.041
chr2R	6921	7695	0.005
chr2R	7715	8554	-0.024
chr3L	8551	8798	0
chr3L	8795	9696	0.042
chr3R	9801	12443	-0.027
chr4	12440	12695	0.08
chrX	12713	13068	0.1
chrX	13065	13295	-0.009

#! /usr/bin/perl -w
use strict;

# Open the input.txt for reading in filehandle FILE
open FILE, "< input.txt";
my $totCols = 0;
my @hdrFields;
my @fields;
my @refs;

 # Read each line of input.txt from the filehandle
while (<FILE>) {
    # Remove the new line character from the line using chomp
    chomp;
    # If it's the first line being read then...
    if ($. == 1) {
        # Split the first line according to whitespaces and store the fields in array @hdrFields
        @hdrFields = split /\s+/;
        # Store the total number of columns in $totCols.
        # $#hdrFields refers to the last index number of attay @hdrFields
        $totCols = $#hdrFields + 1;
        
        # First 4 columns are fixed. So with 5th column as 1, iterate till the last column.
        # In the given example, there are 2 columns after col #4.
        # The below for-loop opens that many files for writing the output. In given example its 2.
        for (1 .. ($totCols - 4)) {
            # Reference to filehandles are stored in array @refs
            open ($refs[$_], "> $hdrFields[4 - 1 + $_].wig");
        }
        
        # After parsing the first line and opening the file handles for writing skip to the next line of file.
        # This will prevent the below code from being executed.
        next;
    }
    
    # For all lines after the header line, do the below:
    # First split the line according to whitespaces and store the fields in array @fields.
    @fields = split /\s+/;
    
    # Again, iterate through that many numbers equivalent to the number of columns after 4 (since first 4 columns are fixed)
    for (1 .. ($totCols - 4)) {
        # Retrieve the reference to file from array @refs
        my $fh = $refs[$_];
        # Print to corresponding file, first 4 fixed columns then the corresponding column. In this case, either col #5 or col #6.
        print $fh join("\t", @fields[1..3]) . "\t$fields[4 - 1 + $_]\n";
    }
}

# Finally close all the files that were opened for writing
for (1 .. ($totCols - 4)) {
    close $refs[$_];
}

# Finally close the input file that was opened for reading
close FILE;

[user@host ~]$ ls -l HEL25E.wig TRIP1.wig
-rw-r--r-- 1 user domain Users 226 Jun  7 16:32 HEL25E.wig
-rw-r--r-- 1 user domain Users 221 Jun  7 16:32 TRIP1.wig
[user@host ~]$

oldIFS=$IFS;IFS=$'\n';for linea in $(cat input.txt);do echo "`echo $linea| awk '{print $1,$2,$3,$4}'`" >> salida.out;\
 echo "`echo $linea| awk '{print $1,$2,$3,$5}'`">>salida2.out ;done ;IFS=$oldIFS

#!/bin/ksh

AllCols=`head -1 $InFile|wc -w`                 # Count all the columns to control the loop

CurCol=5                                        # Set first intersting column
while [ $CurCol -le $AllCols ]                  # Start loop 
do
   OutFile=`head -1 $InFile | \
               cut -f$CurCol -d "tab"`.wig      # Get output filename

   cut -f-4,$CurCol -d "tab" $InFile > $OutFile # Generate output file
   ((CurCol=$CurCol+1))                         # Move counter to next column
done