Primera part (Programa que llegeix el fitxer de matrius, matriu a matriu, enregistrant-les en la memoria com un hash de vectors)


#!/usr/bin/perl -w

use strict;

my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

open(FITXER,"<$ARGV[0]");

my $FT;
my $p = 0; #recorre dins les files del firxer#
my $c = 0; #conta les files del fitxer# 

while (<FITXER>) {
    
    chomp($_);

    if ($_ =~m/FA/){
	
	$FT = $_;
	print "$FT\n";
    }
    
    
    if ($_=~m/[\d]+[\t](\d+)[\s]+(\d+)[\s]+(\d+)[\s]+(\d+)/){

	    $hash{"A"}[$p] = $1;
	    $hash{"C"}[$p] = $2;
	    $hash{"G"}[$p] = $3;
	    $hash{"T"}[$p] = $4;
	    
	    $p = $p + 1;
	    
	    $c = $c +1;
    } 
   
    if (($_=~m/(\/\/)\Z/) && $c>1){

		my @k = ("A","C","G","T");
	    
		my $i = 0;
		
		while ($i < scalar(@k)){
		
		    print "$k[$i]: ";

		    my $j = 0;

		    while ($j < $p){
			print "\t$hash{$k[$i]}[$j]";
			$j = $j + 1;
		    }
		    print"\n";
		    $i = $i + 1;
		}

	
    my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	        "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	        "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	        "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

    $p = 0;
    $c = 0;


    }
}


close(FITXER);







Segona part (Programa que transforma cada matriu de cadascun dels FTs en una matriu de pesos)

#!/usr/bin/perl -w

use strict;

open (SEQPROMOTORA,"<$ARGV[0]");

my $i = 0;

my $sumaA = 0;

my $sumaC = 0;

my $sumaG = 0;

my $sumaT = 0;

my $propA = 0;

my $propC = 0;

my $propG = 0;

my $propT = 0;

my $seqfinal = "";

my $linia;
 
while (<SEQPROMOTORA>){

   chomp($_);

   $linia = $_;
   
   $seqfinal = $seqfinal.$linia;

}

my @v = split (//,$seqfinal);
    
my $longitutvector = scalar (@v);

while ($i < $longitutvector){
	
	if ($v[$i] eq "A" || $v[$i] eq "a"){

	    $sumaA = $sumaA + 1;
	}
	
	if ($v[$i] eq "C" || $v[$i] eq "c"){

	    $sumaC = $sumaC + 1;
	}

	if ($v[$i] eq "G" || $v[$i] eq "g"){

	    $sumaG = $sumaG + 1;
	}

	if ($v[$i] eq "T" || $v[$i] eq "t"){

	    $sumaT = $sumaT + 1;
	}

	$i = $i + 1;
	
}

$propA = ($sumaA / $longitutvector);

$propC = ($sumaC / $longitutvector);

$propG = ($sumaG / $longitutvector);

$propT = ($sumaT / $longitutvector);


close (SEQPROMOTORA);


my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

open(FITXER,"<$ARGV[1]");

my $FT;
my $p = 0; #recorre dins les files de la matriu#
my $c = 0; #conta les files del fitxer# 

while (<FITXER>) {
    
    chomp($_);

    if ($_ =~m/FA/){
	
	$FT = $_;
	print "$FT\n";
    }
    
    
    if ($_=~m/[\d]+[\t](\d+)[\s]+(\d+)[\s]+(\d+)[\s]+(\d+)/){

	    $hash{"A"}[$p] = $1;
	    $hash{"C"}[$p] = $2;
	    $hash{"G"}[$p] = $3;
	    $hash{"T"}[$p] = $4;
	    
	    $p = $p + 1;
	    
	    $c = $c +1;
    }

   if (($_=~m/(\/\/)\Z/) && $c>1){
       
       my $sm = $p; #$sm es la maxima posicio de la matriu#

      
       $p = 0;
       
       while ($p < $sm){
	   
	   my $total = $hash{"A"}[$p] + $hash{"C"}[$p] + $hash{"G"}[$p]+ $hash{"T"}[$p]; 

             
            if ($hash{"A"}[$p] == 0){

		$hash{"A"}[$p] = -999;
	    }
	    else {
			
		$hash{"A"}[$p] = log($hash{"A"}[$p]/$total) - log($propA);
	       
	    }
	    
	    if ($hash{"C"}[$p] == 0){

		$hash{"C"}[$p] = -999;
	    }
	    else {

		$hash{"C"}[$p] = log($hash{"C"}[$p]/$total) - log($propC);
	    }  
			
	    if ($hash{"G"}[$p] == 0){

		$hash{"G"}[$p] = -999;
	    }
	    else {
			
		$hash{"G"}[$p] = log($hash{"G"}[$p]/$total) - log($propG);
	    }
			 
	    if ($hash{"T"}[$p] == 0){

		$hash{"T"}[$p] = -999;
	    }
	    else {
		$hash{"T"}[$p] = log($hash{"T"}[$p]/$total) - log($propT);
	    }
       $p = $p + 1;

	}
       my @k = ("A","C","G","T");
	    
       my $i = 0;
		
       while ($i < scalar(@k)){
		
	      print "$k[$i]: ";

	      my $j = 0;

	      while ($j < $p){
		  my $tmp = sprintf("%.3f",$hash{$k[$i]}[$j]);#fem la variable tmp per poder printar els valors nomes amb 3 decimals#
		  if ($tmp == -999.000){
		      print "\t-999";
		  }
		  else{
		  print "\t$tmp";
	          }
		    $j = $j + 1;
	      }
	      print"\n";
	      $i = $i + 1;
       }
       
       my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	           "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	           "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	           "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

       $p = 0;
       $c = 0;


    }
}


close(FITXER);




Tercera part (Programa que calcula la puntuacio maxima que s'obte al llarg de la sequencia, aixi com la posicio del lloc d'unio que la proporciona) 

#!/usr/bin/perl -w

use strict;

open (SEQPROMOTORA,"<$ARGV[0]");

my $i = 0;

my $sumaA = 0;

my $sumaC = 0;

my $sumaG = 0;

my $sumaT = 0;

my $propA = 0;

my $propC = 0;

my $propG = 0;

my $propT = 0;

my $seqfinal = "";

my $linia;
 
while (<SEQPROMOTORA>){

   chomp($_);

   $linia = $_;
   
   $seqfinal = $seqfinal.$linia;

}
close (SEQPROMOTORA);

my @v = split (//,$seqfinal);
    
my $longitutvector = scalar (@v);

while ($i < $longitutvector){
	
	if ($v[$i] eq "A" || $v[$i] eq "a"){

	    $sumaA = $sumaA + 1;
	}
	
	if ($v[$i] eq "C" || $v[$i] eq "c"){

	    $sumaC = $sumaC + 1;
	}

	if ($v[$i] eq "G" || $v[$i] eq "g"){

	    $sumaG = $sumaG + 1;
	}

	if ($v[$i] eq "T" || $v[$i] eq "t"){

	    $sumaT = $sumaT + 1;
	}

	$i = $i + 1;
	
}

$propA = ($sumaA / $longitutvector);

$propC = ($sumaC / $longitutvector);

$propG = ($sumaG / $longitutvector);

$propT = ($sumaT / $longitutvector);





my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

open(FITXER,"<$ARGV[1]");

my $FT;
my $p = 0; #recorre les columnes de la matriu#
my $c = 0; #conta les files del fitxer# 
my $sm;
while (<FITXER>) {
    
    chomp($_);

    if ($_ =~m/FA/){
	
	$FT = $_;
	print "$FT\n";
    }
    
    
    if ($_=~m/[\d]+[\t](\d+)[\s]+(\d+)[\s]+(\d+)[\s]+(\d+)/){

	    $hash{"A"}[$p] = $1;
	    $hash{"C"}[$p] = $2;
	    $hash{"G"}[$p] = $3;
	    $hash{"T"}[$p] = $4;
	    
	    $p = $p + 1;
	    
	    $c = $c +1;
    }

   if (($_=~m/(\/\/)\Z/) && $c>1){
       
       $sm = $p; #$sm es la maxima posicio de la matriu#
       
       $p = 0;
       
       while ($p < $sm){
	   
	   my $total = $hash{"A"}[$p] + $hash{"C"}[$p] + $hash{"G"}[$p]+ $hash{"T"}[$p]; 
	   
             
	   if ($hash{"A"}[$p] == 0){

		$hash{"A"}[$p] = -999;
	    }
	   else {
	       
	       $hash{"A"}[$p] = log($hash{"A"}[$p]/$total) - log($propA);
	       
	   }
	    
	   if ($hash{"C"}[$p] == 0){
	       
	       $hash{"C"}[$p] = -999;
	   }      
	   
	   else {
	       
	       $hash{"C"}[$p] = log($hash{"C"}[$p]/$total) - log($propC);
	   }  
	   
	   if ($hash{"G"}[$p] == 0){
	       
	       $hash{"G"}[$p] = -999;
	   }
	   else {
	       
	       $hash{"G"}[$p] = log($hash{"G"}[$p]/$total) - log($propG);
	   }
			 
	   if ($hash{"T"}[$p] == 0){
	       
	       $hash{"T"}[$p] = -999;
	   }
	   else {
	       $hash{"T"}[$p] = log($hash{"T"}[$p]/$total) - log($propT);
	   }
	   
	   $p = $p + 1;
       }
       
       my $z = 0;  #posició del nucleòtid dins la seqüència#
       my $n = 0; #posició del nucleòtid dins del motiu# 
       my $score = 0; #variable singular que emmagatzema el valor del pes del nucleòtid en la posició que correspon en el motiu#
	   
       my $scoremax = -9999;
       
       my $posiciomax = 0;
       
       while ($z < scalar(@v) - $sm + 1){
	   
	   $n = 0;
	   $score = 0;
	   
	   while ($n < $sm){
	       
	       if ($v[$z + $n] eq "A" || $v[$z + $n] eq "a"){
		   
		   $score = $score + $hash{"A"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "C" || $v[$z + $n] eq "c"){
		   
		   $score = $score + $hash{"C"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "G" || $v[$z + $n] eq "g"){
		   
		   $score = $score + $hash{"G"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "T" || $v[$z + $n] eq "t"){
		   
		   $score = $score + $hash{"T"}[$n];
	       }
	       $n = $n + 1;
	       
	   }
	   
	   
	   if ($score > $scoremax){
	       
	       $scoremax = $score;
	       
	       $posiciomax = $z;
	       
	   }
	   
	   $z = $z + 1;
	   
       }
       print "$scoremax\n";
       print "$posiciomax\n";
  
	

       
    %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);
    $p = 0;
    $c = 0;
 
   }
}

close(FITXER);





Quarta part (Programa que calcula el p-value de cada FT respecte la sequencia promotora)


#!/usr/bin/perl -w

use strict;

open (SEQPROMOTORA,"<$ARGV[0]");

my $i = 0;

my $sumaA = 0;

my $sumaC = 0;

my $sumaG = 0;

my $sumaT = 0;

my $propA = 0;

my $propC = 0;

my $propG = 0;

my $propT = 0;

my $seqfinal = "";

my $linia;
 
while (<SEQPROMOTORA>){

   chomp($_);

   $linia = $_;
   
   $seqfinal = $seqfinal.$linia;

}
close (SEQPROMOTORA);

my @v = split (//,$seqfinal);
    
my $longitutvector = scalar (@v);

while ($i < $longitutvector){
	
	if ($v[$i] eq "A" || $v[$i] eq "a"){

	    $sumaA = $sumaA + 1;
	}
	
	if ($v[$i] eq "C" || $v[$i] eq "c"){

	    $sumaC = $sumaC + 1;
	}

	if ($v[$i] eq "G" || $v[$i] eq "g"){

	    $sumaG = $sumaG + 1;
	}

	if ($v[$i] eq "T" || $v[$i] eq "t"){

	    $sumaT = $sumaT + 1;
	}

	$i = $i + 1;
	
}

$propA = ($sumaA / $longitutvector);

$propC = ($sumaC / $longitutvector);

$propG = ($sumaG / $longitutvector);

$propT = ($sumaT / $longitutvector);





my %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	    "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);

open(FITXER,"<$ARGV[1]");

my $FT;
my $p = 0; #recorre les columnes de la matriu#
my $c = 0; #conta les files del fitxer# 
my $sm;
while (<FITXER>) {
    
    chomp($_);

    if ($_ =~m/FA/){
	
	$FT = $_;
	print "$FT\n";
    }
    
    
    if ($_=~m/[\d]+[\t](\d+)[\s]+(\d+)[\s]+(\d+)[\s]+(\d+)/){

	    $hash{"A"}[$p] = $1;
	    $hash{"C"}[$p] = $2;
	    $hash{"G"}[$p] = $3;
	    $hash{"T"}[$p] = $4;
	    
	    $p = $p + 1;
	    
	    $c = $c +1;
    }

   if (($_=~m/(\/\/)\Z/) && $c>1){
       
       $sm = $p; #$sm es la maxima posicio de la matriu#
       
       $p = 0;
       
       while ($p < $sm){
	   
	   my $total = $hash{"A"}[$p] + $hash{"C"}[$p] + $hash{"G"}[$p]+ $hash{"T"}[$p]; 
	   
             
	   if ($hash{"A"}[$p] == 0){

		$hash{"A"}[$p] = -999;
	    }
	   else {
	       
	       $hash{"A"}[$p] = log($hash{"A"}[$p]/$total) - log($propA);
	       
	   }
	    
	   if ($hash{"C"}[$p] == 0){
	       
	       $hash{"C"}[$p] = -999;
	   }      
	   
	   else {
	       
	       $hash{"C"}[$p] = log($hash{"C"}[$p]/$total) - log($propC);
	   }  
	   
	   if ($hash{"G"}[$p] == 0){
	       
	       $hash{"G"}[$p] = -999;
	   }
	   else {
	       
	       $hash{"G"}[$p] = log($hash{"G"}[$p]/$total) - log($propG);
	   }
			 
	   if ($hash{"T"}[$p] == 0){
	       
	       $hash{"T"}[$p] = -999;
	   }
	   else {
	       $hash{"T"}[$p] = log($hash{"T"}[$p]/$total) - log($propT);
	   }
	   
	   $p = $p + 1;
       }

       my $z = 0;  #posició del nucleòtid dins la seqüència#
       my $n = 0; #posició del nucleòtid dins del motiu# 
       my $score = 0; #variable singular que emmagatzema el valor del pes del nucleòtid en la posició que correspon en el motiu#
	   
       my $scoremax = -9999;
       
       my $posiciomax = 0;
       
       while ($z < scalar(@v) - $sm + 1){
	   
	   $n = 0;
	   $score = 0;
	   
	   while ($n < $sm){
	       
	       if ($v[$z + $n] eq "A" || $v[$z + $n] eq "a"){
		   
		   $score = $score + $hash{"A"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "C" || $v[$z + $n] eq "c"){
		   
		   $score = $score + $hash{"C"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "G" || $v[$z + $n] eq "g"){
		   
		   $score = $score + $hash{"G"}[$n];
	       }
	       
	       if ($v[$z + $n] eq "T" || $v[$z + $n] eq "t"){
		   
		   $score = $score + $hash{"T"}[$n];
	       }
	       $n = $n + 1;
	       
	   }
	   
	   
	   if ($score > $scoremax){
	       
	       $scoremax = $score;
	       
	       $posiciomax = $z;
	       
	   }
	   
	   $z = $z + 1;
	   
       }
       print "L'score màxim és: $scoremax\n";
       print "La posició d'aquest score és: $posiciomax\n";
       
       my $bucle = 0;
       my $pvalue;
       my $cops = 0;

       while ($bucle < 100){

	   my @permutat = @v;
       
	   my $x = $longitutvector - 1;
       
	   while ($x >= 0) {
	       my $y = int(rand($x+1));
	       if ($x != $y) {
		   my $lloc = $permutat[$x];

		   $permutat[$x] = $permutat[$y];
		   $permutat[$y] = $lloc;
	       }
	       $x = $x - 1;
	   }
       	   
	   my $scorefals = -9999;
       
	   my $posiciofals = 0;

	   $z = 0;
	   
	   while ($z < scalar(@permutat) - $sm + 1){
	   
	       $n = 0;
	       $score = 0;
	   
	       while ($n < $sm){
	       
		   if ($permutat[$z + $n] eq "A" || $permutat[$z + $n] eq "a"){
		   
		       $score = $score + $hash{"A"}[$n];
		   }
	       
		   if ($permutat[$z + $n] eq "C" || $permutat[$z + $n] eq "c"){
		   
		       $score = $score + $hash{"C"}[$n];
		   }
	       
		   if ($permutat[$z + $n] eq "G" || $permutat[$z + $n] eq "g"){
		   
		       $score = $score + $hash{"G"}[$n];
		   }
	       
		   if ($permutat[$z + $n] eq "T" || $permutat[$z + $n] eq "t"){
		   
		       $score = $score + $hash{"T"}[$n];
		   }
		   $n = $n + 1;
	       
	       }
	   
	       
	       if ($score > $scorefals){
	       
		   $scorefals = $score;
	       
		   $posiciofals = $z;
	       
	       }
	   
	       $z = $z + 1;
	   
	   }
	   if ($scorefals >= $scoremax){
	       $cops = $cops + 1;
	   } 
	   $bucle = $bucle + 1;
       }
       $pvalue = $cops/100;
       print"El p-value és: $pvalue\n";
       

 %hash = ("A" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "C" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "G" =>[0, 0, 0, 0, 0, 0, 0, 0, 0],
	     "T" =>[0, 0, 0, 0, 0, 0, 0, 0, 0]);
    $p = 0;
    $c = 0;
 
   }
}

close(FITXER);