#!/usr/bin/perl

use strict;
use warnings;

$DNA1 = "GGCT CTGCGCGGNN";

# Remove N from sequence
$DNA1 =~ s/N//ig;

# Remove whitespace Line 5
$DNA1 =~ s/\s//g;

# In a loop, find every 4-base substring & then find its
# GC%, GC-skew & Purine Loading Index (PLI):
$fm = 1.010;    # Line 9
do {
    while ( my $fm = substr( $DNA1, 0, 4 ) ) {
        $A = 0;
        $T = 0;
        $G = 0;
        $C = 0;
        while ( $fm =~ /A/ig ) { $A++ }
        while ( $fm =~ /T/ig ) { $T++ }
        while ( $fm =~ /G/ig ) { $G++ }
        while ( $fm =~ /C/ig ) { $C++ }
        $tot1   = $A + $T + $G + $C;
        $gc1    = $G - $C;
        $gc2    = $G + $C;                      # Line 16
        $cent   = 100;
        $gccon2 = $gc2 / $tot1;
        $gccon3 = $cent * $gccon2;
        $gccon4 = sprintf( "%.2f", $gccon3 );
        $gcskew = $gc1 / $gc2;
        $GCSkew = sprintf( "%.4f", $gcskew );

        # To find Purine Loading Index (PLI):
        $four = 4;
        $at1  = $A - $T;
        $x1   = ( $gc1 + $at1 ) / $tot1;
        $pli  = $thousand * $x1;
        $PLI  = sprintf( "%.0f", $pli );

        # No. of sliding Windows:
        $numberwin = $total1 / $four;
        $NoWindows = sprintf( "%.0f", $numberwin );
        print "  Purine Loading Index of each 1Kb Window=$PLI bases/4-base.\n";
        $output = "GC-SkewResult .txt";
        unless ( open( RESULT, ">$output" ) ) {
            print "Cannot open file\"$output\".\n\n";
            exit;
        }
        print RESULT"\n  RESULTS for substrings:\n
 GC-Skew values of substrings:\n
   $GCSkew\n\n
 Percent GC Content of substrings:\n
   $gccon4\n\n";
        close(RESULT);
    }
} until ( $fm =~ /^\s*$/ );
exit;