#!/usr/bin/perl -w

package runmitoprot;

use strict;
use IO::File;

use base 'Exporter';
our @EXPORT = qw/runmitoprot extract_mitoprot_results/;



# Run MITOPROT
sub runmitoprot($$$){
		
	my ($infile,$outpath,$execommand) = @_;
	
	my @args = ();
	
	# get teh names of the parameter files
	my ($inputfilename) = $infile =~ /\/([^\/]+)$/;
	print "$inputfilename\n";

	my ($inputname) = $infile =~ /\/([^\/]+)\..*$/;
	print "$inputname\n";

	my ($command) = $execommand =~ /\/([^\/]+)$/;
	print "$command\n";


	# create the symbolic links for the input file to the working directory
	# otherwise the resulting file cannot be obtained
	@args = ("ln", "-s", $infile, "$outpath/.");
	unless (-e "$outpath/$inputfilename") {system(@args) == 0 or die "system @args failed: $?"};

	# create the symbolic links for the executable to the working directory
	# otherwise the resulting file cannot be obtained
	@args = ("ln", "-s", $execommand, "$outpath/.");
	unless (-e "$outpath/$command") {system(@args) == 0 or die "system @args failed: $?"};

	# RUN the program in the working dir
	# th output file is localized where the command is launched (!?)
	@args = ("$outpath/$command", "$outpath/$inputfilename");
	system(@args) == 0 or die "system @args failed: $?";

	# Placing the output file in the working directory
	@args = ("mv", "$inputname.mitoprot", "$outpath/.");
	system(@args) == 0 or die "system @args failed: $?";

	return $inputname,"$outpath/$inputname.mitoprot";
}




#~ # For each sequence, this function will create a new file : Mitoprot work with only one sequence in one file.
#~ # mkdir the output dir in ergatis component before
#~ sub format_multifasta_to_singlefastas($$){
	#~ 
	#~ my ($inputfile,$outdir) = @_;
	#~ 
	#~ print "runmitoprot::format_multifasta_to_singlefastas: formats a fasta file containing multiple sequences into 1 sequence per file...\n";
	#~ 
	#~ my $listsnps = new IO::File "$inputfile" || die "Can not open file $inputfile : $!";#Data
#~ 
	#~ my $i = 1;
	#~ my $k = 0;
	#~ while(<$listsnps>){
		#~ 
		#~ if($_ =~ /^>(.*)/){
			#~ 
			#~ my $id = $1;
			#~ my $tempmitoprot = new IO::File ">fic/temp-mitoprot_$date" || die "Can not create fic/temp-mitoprot_$date : $! \n";
			#~ my $name = $_;
			#~ print $tempmitoprot $name;#print the name of the SNP
			#~ $_ = <$listsnps>;
			#~ if($_ !~ /^M/){#A seq must begin by a M in Mitoprot II
				#~ print $stderr "$name";
				#~ $k++;
				#~ next;
			#~ }
			#~ print $tempmitoprot $_;#Print the sequence of the SNP
			#~ $tempmitoprot->close();
			#~ system("cp fic/temp-mitoprot_$date $path/Programmes/MitoprotII/mitoprotII-v1.101");#Copy file for run Mitoprot
			#~ chdir "$path/Programmes/MitoprotII/mitoprotII-v1.101";
			#~ `./mitoprot temp-mitoprot_$date`;
			#~ system("rm -f temp-mitoprot_$date");#Clean
			#~ system("rm -f $path/fic/temp-mitoprot_$date");#Clean
			#~ extract_results($id,$path,$date);
			#~ #system("mv tempmitoprot.mitoprot $path/results/MitoprotII/seq$i");#Rename and replace file (optional)
			#~ $i++;
			#~ chdir "$path";
		#~ }
		#~ else{
			#~ print "Error\n";
		#~ }
	#~ }
	#~ $listsnps->close();
	#~ $stderr->close();
	#~ #############################################################
	#~ 
	#~ search_var($path,$delta,$date);
	#~ print "$k errors found. Consult results/error.out.\n";
#~ }

#~ 
#~ sub format_multifasta_to_singlefastas($$){
	#~ 
	#~ my ($delta,$path,$datedate) = @_;
	#~ 
	#~ print "runmitoprot::create: Searching amino acid sequence for each allele...\n";
	#~ 
	#~ #Check the fresults folder.
	#~ print "Running Mitoprot II.\n";
	#~ unless (-e "$path/results/MitoprotII"){
		#~ system("mkdir results/MitoprotII");
	#~ }
	#~ 
	#~ ############################################Batch
	#~ my $listsnps = new IO::File "fic/list_snps_$date" || die "Can not open file list_snps_$date : $!";#Data
	#~ my $stderr = new IO::File ">>results/error_$date" || die "Can not open file results/error_$date : $!";#Errors file
	#~ print $stderr "\nErrors for Mitoprot: A seq must begin by a M. The folowing sequences are untreated.\n";
	#~ my $i = 1;
	#~ my $k = 0;
	#~ while(<$listsnps>){
		#~ if($_ =~ /^>(.*)/){
			#~ my $id = $1;
			#~ my $tempmitoprot = new IO::File ">fic/temp-mitoprot_$date" || die "Can not create fic/temp-mitoprot_$date : $! \n";
			#~ my $name = $_;
			#~ print $tempmitoprot $name;#print the name of the SNP
			#~ $_ = <$listsnps>;
			#~ if($_ !~ /^M/){#A seq must begin by a M in Mitoprot II
				#~ print $stderr "$name";
				#~ $k++;
				#~ next;
			#~ }
			#~ print $tempmitoprot $_;#Print the sequence of the SNP
			#~ $tempmitoprot->close();
			#~ system("cp fic/temp-mitoprot_$date $path/Programmes/MitoprotII/mitoprotII-v1.101");#Copy file for run Mitoprot
			#~ chdir "$path/Programmes/MitoprotII/mitoprotII-v1.101";
			#~ `./mitoprot temp-mitoprot_$date`;
			#~ system("rm -f temp-mitoprot_$date");#Clean
			#~ system("rm -f $path/fic/temp-mitoprot_$date");#Clean
			#~ extract_results($id,$path,$date);
			#~ #system("mv tempmitoprot.mitoprot $path/results/MitoprotII/seq$i");#Rename and replace file (optional)
			#~ $i++;
			#~ chdir "$path";
		#~ }
		#~ else{
			#~ print "Error\n";
		#~ }
	#~ }
	#~ $listsnps->close();
	#~ $stderr->close();
	#~ #############################################################
	#~ 
	#~ search_var($path,$delta,$date);
	#~ print "$k errors found. Consult results/error.out.\n";
#~ }

#Concatenation in /results/MitoprotII/mitoprot
print "WARNING: need to put the correct name instead of ID!!!!!\n";
sub extract_mitoprot_results($$$){	
	
	my ($path, $resfile, $seqname) = @_;
	
	my $score;
	my $found = 0;
	
	my $fh_in = new IO::File $resfile || die "Can not open the output file of mitoprot $resfile : $! \n";
	my $fh_out = new IO::File ">>$path/mitoprot_results.txt" || die "Can not create the result file (mitoprot) $path/mitoprot_results.txt: $! \n";
	
	while(<$fh_in>){
		
		if($_ =~ m/^DFM\s*:\s*.*\s*(\d\.\d*)/){#Search a line with DFM (Discriminant function for mitochondrial proteins)
			$score = $1;
		}
		if($_ =~ m/This protein is probably imported in mitochondria./){#Check if the protein is imported to the mitochondria.
			$found = 1;#$found is true
		}
	}
	if($found == 1){#$found = true => this protein is imported to mitochondria.
		if(defined($score)){
			print $fh_out ">$seqname\timported\n"."$score\n";
		}
	}
	else{
		if(defined($score)){
			print $fh_out ">$seqname\tnot imported\n"."$score\n";
		}
	}
	$fh_in->close();
	$fh_out->close();
}


#Find the transcript who win/lost mitochondrial signal
sub search_var($$$){
	my ($path, $delta,$mitoprotresfile) = @_;
	
	my $fh_in = new IO::File "$mitoprotresfile" || die "Can not create the result file (mitoprot) $mitoprotresfile: $! \n";
	my $results = new IO::File ">$path/mitoprot_results_2.txt" || die "Can not create the file $path/mitoprot_results_2.txt : $! \n";
	
	my $oname;
	my $allele1;
	my $allele2;
	my $found = 0;
	my $vers = 0;
	
	while(<$fh_in>){
		if($_ =~ m/>(.*)_allele.*?(imported|not imported)/){
			my $type = $2;
			my $name = $1;
			if($type eq "imported"){$found++;}
			else{
				if($found == 0){
					$vers++;
				}
			}
			
			$_ = <$fh_in>;#Next line : score.
			$allele2 = $_;
			chomp $allele2;
			if((defined($oname)) && ($oname eq $name)){#If previous name already exist (2 alleles)
				if($found == 2){#If the 2 alleles are imported
					my $score = $allele1 - $allele2;#Difference between the 2 alleles
					if($score >= $delta){
						print $results "$oname : Signal lost - scores: allele1=$allele1 allele2=$allele2\n";#Mean : allele 1 have better signal than allele 2
					}
					if($score <= -$delta){
						print $results "$oname : Signal gain - scores: allele1=$allele1 allele2=$allele2\n";#Mean : allele 2 have better signal
					}
					$found = 0;
				}
				elsif($found == 1){#If 1 allele is non imported
					if($vers == 1){
						print $results "$oname : Signal gain \n";
					}
					else{
						print $results "$oname : Signal lost \n";
					}
					$vers = 0;
					$found = 0;
				}
			}
			else{#new name (first allele)
				$allele1 = $_;
				chomp $allele1;
			}
			$oname = $name;
		}
	}
	$res->close();
	$results->close();
}

1;