#!/usr/bin/perl -w

# === === === === === === START === === === === === Documents === === === === === START === === === === === === #
#														#
# American 7, contigs  	  -> time perl UNIVERSAL_ART.pl A7 C all(or 001 or 002 ...)   //3 parameters:		#
# American 8, comtigs  	  -> time perl UNIVERSAL_ART.pl A8 C all(or 001 or 002 ...)   //ARGV[0] = sample name	#
# whalefall1, contigs	  -> time perl UNIVERSAL_ART.pl W1 C all(or 001 or 002 ...)   //ARGV[1] = data type	#
# whalefall2, contigs	  -> time perl UNIVERSAL_ART.pl W2 C all(or 001 or 002 ...)   //ARGV[2] = file count	#
# whalefall3, contigs	  -> time perl UNIVERSAL_ART.pl W3 C all(or 001 or 002 ...)   //(all indicates all reads#
# American 7, reads(raw)  -> time perl UNIVERSAL_ART.pl A7 R all(or 001 or 002 ...)   //in one file; 001, 002,  #
# American 8, reads(raw)  -> time perl UNIVERSAL_ART.pl A8 R all(or 001 or 002 ...)   //... denotes separative	#
# American 7, reads(qual) -> time perl UNIVERSAL_ART.pl A7 Q all(or 001 or 002 ...)   //files in one sample)	#
# American 8, reads(qual) -> time perl UNIVERSAL_ART.pl A8 Q all(or 001 or 002 ...)   //ARGV[3] = set E-value	#
# In-A	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-A R all(or 001 or 002 ...) //ARGV[4] = set # of hits	#
# In-B	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-B R all(or 001 or 002 ...)				#
# F1-T	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F1-T R all(or 001 or 002 ...)				#
# F1-U	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F1-U R all(or 001 or 002 ...)				#
# F2-V	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F2-V R all(or 001 or 002 ...)				#
# In-D	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-D R all(or 001 or 002 ...)				#
# In-E	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-E R all(or 001 or 002 ...)				#
# In-M	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-M R all(or 001 or 002 ...)				#
# In-R	    , reads(raw)  -> time perl UNIVERSAL_ART.pl In-R R all(or 001 or 002 ...)				#
# F1-S	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F1-S R all(or 001 or 002 ...)				#
# F2-W	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F2-W R all(or 001 or 002 ...)				#
# F2-X	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F2-X R all(or 001 or 002 ...)				#
# F2-Y	    , reads(raw)  -> time perl UNIVERSAL_ART.pl F2-Y R all(or 001 or 002 ...)				#
#														#
# MC	    , reads(raw)  -> time perl UNIVERSAL_ART.pl MC S all 1e-2 250 (or 001 or 002 ...)			#
#														#
# pseudo-   , short seq.  -> time perl UNIVERSAL_ART.pl PD_S P all(or 001 or 002 ...)				#
# pseudo-   , long seq.   -> time perl UNIVERSAL_ART.pl PD_L P all(or 001 or 002 ...)				#
#														#
# === === === === === ==== END ==== === === === === Documents === === === === ==== END ==== === === === === === #

use Bio::SeqIO;
use Bio::SearchIO;

# === === === === === START === === === === === sample chosen === === === === === START === === === === ===

if( $ARGV[0] =~ "A7" )  { $sample_name = "American_7"; }
elsif( $ARGV[0] =~ "A8" )  { $sample_name = "American_8"; }
elsif( $ARGV[0] =~ "W1" )  { $sample_name = "whalefall_1"; }
elsif( $ARGV[0] =~ "W2" )  { $sample_name = "whalefall_2"; }
elsif( $ARGV[0] =~ "W3" )  { $sample_name = "whalefall_3"; }
elsif( $ARGV[0] =~ "In-A" )  { $sample_name = "In-A"; }
elsif( $ARGV[0] =~ "In-B" )  { $sample_name = "In-B"; }
elsif( $ARGV[0] =~ "F1-T" )  { $sample_name = "F1-T"; }
elsif( $ARGV[0] =~ "F1-U" )  { $sample_name = "F1-U"; }
elsif( $ARGV[0] =~ "F2-V" )  { $sample_name = "F2-V"; }
elsif( $ARGV[0] =~ "In-D" )  { $sample_name = "In-D"; }
elsif( $ARGV[0] =~ "In-E" )  { $sample_name = "In-E"; }
elsif( $ARGV[0] =~ "In-M" )  { $sample_name = "In-M"; }
elsif( $ARGV[0] =~ "In-R" )  { $sample_name = "In-R"; }
elsif( $ARGV[0] =~ "F1-S" )  { $sample_name = "F1-S"; }
elsif( $ARGV[0] =~ "F2-W" )  { $sample_name = "F2-W"; }
elsif( $ARGV[0] =~ "F2-X" )  { $sample_name = "F2-X"; }
elsif( $ARGV[0] =~ "F2-Y" )  { $sample_name = "F2-Y"; }
elsif( $ARGV[0] =~ "MC" )  { $sample_name = "MC"; }
elsif( $ARGV[0] =~ "PD_S" )  { $sample_name = "short"; }
elsif( $ARGV[0] =~ "PD_L" )  { $sample_name = "long"; }
if( $ARGV[1] =~ "C" )  { $data_type = "contigs"; }
elsif( $ARGV[1] =~ "R" )  { $data_type = "raw_reads"; }
elsif( $ARGV[1] =~ "Q" )  { $data_type = "qual_reads"; }
elsif( $ARGV[1] =~ "S" )  { $data_type = "simulation"; }
elsif( $ARGV[1] =~ "P" )  { $data_type = "pseudo_reads"; }

# === === === === ==== END ==== === === === === sample chosen === === === === ==== END ==== === === === ===

# === === === === === START === === === === === from Chien-Hao === === === === === START === === === === ===
open TMP1, " < merged.dmp";
open TMP2, " < names.dmp";
open TMP3, " < nodes.dmp";

while($line =<TMP1>){

	chomp $line;
	@array = split/\t/, $line;
	
	if (exists $merged{$array[0]}){
		print "Exists the same key of the merged hash table !! \n";
	}else{
		$merged{$array[0]} = $array[2];
	}
}
close TMP1;

while($line =<TMP2>){

	chomp $line;
	@array = split/\t/, $line;
	
	if ( $array[6] eq 'scientific name' ){

		if (exists $names{$array[0]}){
			print "Exists the same key of the name hash table !! \n";
		}else{
			$names{$array[0]} = $array[2];
		}
	}
}
close TMP2;

while($line =<TMP3>){

	chomp $line;
	@array = split/\t/, $line;
	
	if ( (exists $parent{$array[0]}) or (exists $rank{$array[0]}) ){
		print "Exists the same key of the parent or rank hash table !! \n";
	}else{
		$parent{$array[0]} = $array[2];
		$rank{$array[0]} = $array[4];
	}
}
close TMP3;
# === === === === === END === === === === ===== from Chien-Hao ===== === === === === END === === === === ===

=cut
#results
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/Results_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_result, ">$path" ) || die "$!\n";
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/NoResults_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_no_result, ">$path" ) || die "$!\n";
#log
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/debug/LOG_input_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_input_reads_log, ">$path" ) || die "$!\n";
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/debug/NO_speciesID_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_ID_log, ">$path" ) || die "$!\n";
=cut

#results
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/Results_".$ARGV[3]."_".$ARGV[4]."_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_result, ">$path" ) || die "$!\n";
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/NoResults_".$ARGV[3]."_".$ARGV[4]."_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_no_result, ">$path" ) || die "$!\n";
#log
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/debug/LOG_input_".$ARGV[3]."_".$ARGV[4]."_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_input_reads_log, ">$path" ) || die "$!\n";
$path = "/bioinfo_cluster/home/mineyours/results/" . $data_type . "/" . $sample_name . "/debug/NO_speciesID_".$ARGV[3]."_".$ARGV[4]."_".$ARGV[0]."_".$ARGV[1]."_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
open( FHD_ID_log, ">$path" ) || die "$!\n";

#hash table ( protein_name to accession_number )
open( $fp, "/bioinfo_cluster/home/mineyours/metaDATA/phylogeny/ProteinName_to_AccessionNumber.txt" );  #+-+-+-+-+-+-+-+-+-+#
while( $line=<$fp> )
{
    @element = split( '\t', $line );
    $element[0] =~ s/^\s+//g;  #cut damn head (protein name)
    $element[0] =~ s/\s+$//g;  #cut damn tail
    $element[1] =~ s/^\s+//g;  #cut damn head (accession number)
    $element[1] =~ s/\s+$//g;  #cut damn tail
    $protein_to_id{ $element[0] } = $element[1];
}
close($fp);

# Accession number to Taxon id
open( $fp, "/bioinfo_cluster/home/mineyours/metaDATA/phylogeny/TAXONID_ACCESSION_GI.txt" );  #+-+-+-+-+-+-+-+-+-+#
while( $line=<$fp> )
{
    @element = split( '\t', $line );
    $element[0] =~ s/^\s+//g;  #cut damn head  ( Taxon id )
    $element[0] =~ s/\s+$//g;  #cut damn tail
    $element[1] =~ s/^\s+//g;  #cut damn head  ( accession number )
    $element[1] =~ s/\s+$//g;  #cut damn tail
    $element[2] =~ s/^\s+//g;  #cut damn head  ( GI )
    $element[2] =~ s/\s+$//g;  #cut damn tail
    $Taxon_ID{ $element[1] } = $element[0];
}
close($fp);

#open file
if( $ARGV[2] =~ "all" )  { $open_path = "/bioinfo_cluster/home/mineyours/metaDATA/sample/Dumped_sequences/" . $sample_name."_".$data_type; }  #+-+-+-+-+-+-+-+-+-+#
else  			   { $open_path = "/bioinfo_cluster/home/mineyours/metaDATA/sample/Dumped_sequences/subQ/" . $sample_name."_".$data_type."_".$ARGV[2]; }  #+-+-+-+-+-+-+-+-+-+#
$in_fasta = Bio::SeqIO->new( -file => $open_path, -format => 'fasta' );

$count_reads = 0;
while( $seq_Unit = $in_fasta->next_seq() )
{
    if( $sample_name =~ "American_7" || $sample_name =~ "American_8" || $sample_name =~ "whalefall_1" || $sample_name =~ "whalefall_2" || $sample_name =~ "whalefall_3" || $sample_name =~ "short" || $sample_name =~ "long" )  
    { $read_ID = $seq_Unit->desc(); }
    else  # Japan: singletons are generated from .scf file, so no gi number. id = read name
    { $read_ID = $seq_Unit->id(); $read_des = $seq_Unit->desc(); }
    $sequence = $seq_Unit->seq();
    #$read_length = $seq_Unit->length;
    
    #log
    print FHD_input_reads_log $read_ID . "\n";
    
    #save to input.txt while blastall
    #$path="/bioinfo_cluster/home/mineyours/BLAST/blast/" . $sample_name . "/" . $data_type . "/all_input_".$ARGV[2].".txt";  #+-+-+-+-+-+-+-+-+-+#
    $path="/bioinfo_cluster/home/mineyours/BLAST/blast/" . $sample_name . "/" . $data_type . "/all_input_".$ARGV[2]."_".$ARGV[3]."_".$ARGV[4].".txt";  #+-+-+-+-+-+-+-+-+-+#
    open( FHD, ">$path" ) || die "$!\n";
    print FHD $sequence;
    close( FHD );
    
    #run 1st BLAST
    $cmd = "cd ~/BLAST/blast\n blastall -e ".$ARGV[3]." -b ".$ARGV[4]." -v ".$ARGV[4]." -p blastx -i ./".$sample_name."/".$data_type."/all_input_".$ARGV[2]."_".$ARGV[3]."_".$ARGV[4].".txt -d aaDB -o ./".$sample_name."/".$data_type."/all_output_".$ARGV[2]."_".$ARGV[3]."_".$ARGV[4].".fsa";  #+-+-+-+-+-+-+-+-+-+#  (root)
    system( $cmd );
    
    # construct @Tables, which contains total number of queries of species(non-redundant)
    #$in_blast = new Bio::SearchIO( -format => 'blast', -file => '/bioinfo_cluster/home/mineyours/BLAST/blast/'.$sample_name.'/'.$data_type.'/all_output_'.$ARGV[2].'.fsa' );  #+-+-+-+-+-+-+-+-+-+#
    $in_blast = new Bio::SearchIO( -format => 'blast', -file => '/bioinfo_cluster/home/mineyours/BLAST/blast/'.$sample_name.'/'.$data_type.'/all_output_'.$ARGV[2].'_'.$ARGV[3]."_".$ARGV[4].'.fsa' );  #+-+-+-+-+-+-+-+-+-+#
    while( $result = $in_blast->next_result() )
    {
	while( $hit = $result->next_hit() )
	{
	    while( $hsp = $hit->next_hsp() )
	    {
		$temp_protein_name = $hit->name();
		if( $temp_protein_name =~ /ref\|(\w+)\..+/ )
		{
		    $protein_name = $1;
		}
		# there might be two files corresponding to one species (chromosome & plasmid), which means we have different accession number for single species. To solve this problem, we use proteinID to find corresponding accession number of species, cuz that's one-to-one relationship.
		if( $species_ID = $protein_to_id{ $protein_name } )  # there's not a problem to find corresponding speciesID by using proteinID (one-to-one)
		{
		    if( !$Tables{ $species_ID } )
		    {
			$Tables{ $species_ID } = 1;  # construct Tables¡Gthe number of hash table = the number of different species (non-redundant) from all queries in output file
		    }
		    else
		    {
			$Tables{ $species_ID }++;  # construct Tables¡Gthe number of hash table = the number of different species (non-redundant) from all queries in output file
		    }
		}
		else
		{
		    print FHD_ID_log "Couldn't find the corresponding *.faa file from this protein: [" . $protein_name . "]\n";
		}
	    }
	}
    }
    
    #  parse informations (we only have interest to the blast results, that have at least 2 queries corresponding to the same species)
    #$in_blast = new Bio::SearchIO( -format => 'blast', -file => '/bioinfo_cluster/home/mineyours/BLAST/blast/'.$sample_name.'/'.$data_type.'/all_output_'.$ARGV[2].'.fsa' );  #+-+-+-+-+-+-+-+-+-+#
    $in_blast = new Bio::SearchIO( -format => 'blast', -file => '/bioinfo_cluster/home/mineyours/BLAST/blast/'.$sample_name.'/'.$data_type.'/all_output_'.$ARGV[2].'_'.$ARGV[3].'_'.$ARGV[4].'.fsa' );  #+-+-+-+-+-+-+-+-+-+#
    while( $result = $in_blast->next_result() )
    {
	while( $hit = $result->next_hit() )
	{
	    while( $hsp = $hit->next_hsp() )
	    {
		$temp_species_name = $hit->description();
		if( $temp_species_name =~ /.+\[(.+)\]/ )
		{
		    #$species_name = $1;
		}
		$temp_protein_ID = $hit->name();
		if( $temp_protein_ID =~ /ref\|(\w+)\..+/ )
		{
		    $protein_ID = $1;
		}
		$Score = $hsp->bits();
		$E_value = $hsp->evalue();
		$blast_frame = ($hsp->query->frame + 1) * $hsp->query->strand;
		$Identities = $hsp->percent_identity(2);
		$identical_hits = $hsp->num_identical() . "/" .$hsp->length('total');
		$start_query = $hsp->start('query');
		$stop_query = $hsp->end('query');
		$start_subject = $hsp->start('hit');
		$stop_subject = $hsp->end('hit');
		
		$species_ID = $protein_to_id{ $protein_ID };
		if( $Tables{ $species_ID } >= 2 )  # parse the results, that at least 2 queries corresponding to same species
		{
		    @Informations = ( @Informations, [ $species_ID, $protein_ID, $Score, $E_value, $blast_frame, $Identities, $identical_hits, $start_query, $stop_query, $start_subject, $stop_subject ] );
		}
	    }
	}
    }
    @key = keys %Tables;  # clear Tables
    for( $i=0;$i<=$#key;$i++ )
    {
	delete $Tables{ $key[$i] };
    }
    
    $min_E = 0.01;
    $got_it = FALSE;
    $considerLCA = FALSE;
    $number_of_LCA = 0;
    for( $k=0;$k<21;$k++ )  { push(@Min_Informations, 0); }
    $threshold = $#Informations;
    for( $i=0;$i<=$threshold-1;$i++ )
    {
	# same species & adjacent proteins
	$speciesID = $Informations[$i]->[0];  # Note that there might be two files corresponding to one species (chromosome & plasmid), which means we have different accession number for single species. To solve this problem, we use proteinID to find corresponding accession number of species, cuz that's one-to-one relationship.
	$file_protein_pre = $file_protein_post = $file_strain_pre = $file_strain_middle = $file_strain_post = undef;
	$path = "/bioinfo_cluster/home/mineyours/metaDATA/phylogeny/ffn/" . $speciesID . ".ffn";
	$in_ffn = Bio::SeqIO->new( -file => $path, -format => 'fasta' );
	$path = "/bioinfo_cluster/home/mineyours/metaDATA/phylogeny/faa/" . $speciesID . ".faa";
	$in_faa = Bio::SeqIO->new( -file => $path, -format => 'fasta' );
# === === === === === START === === === === === search for proteins and gene direction from .faa file === === === === === START === === === === ===
	while( $faa_Unit = $in_faa->next_seq() )
	{
	    if( $ffn_Unit = $in_ffn->next_seq() ) {} else { last; }  # weird! the total number of gene is not always consistent with total number of protein even if they have the same accession number. ex,NC_003047
	    if( $ffn_Unit->id() =~ /.+\:(c*).+/ ) { push( @strain_order, $1 ); }  #ffn, find gene direction
	    if( $faa_Unit->id() =~ /.+\|.+\|.+\|(\w+)\..+/ ) { push( @protein_order, $1 ); }  #faa, find protein name
	    if( $#protein_order > 1 ) { shift( @protein_order ); shift( @strain_order ); }  
	    
	    if( $Informations[$i]->[1] =~ $protein_order[0] )  # match in the beginning: no pre_protein
	    {
		if( $faa_Unit = $in_faa->next_seq() )  # a way to fix the weird problem above
		{
		    if( $ffn_Unit = $in_ffn->next_seq() )  # a way to fix the weird problem above
		    {
			if( $ffn_Unit->id() =~ /.+\:(c*).+/ )  # ffn
			{
			    push( @strain_order, $1 );
			    if( !$strain_order[0] ){ $file_strain_middle = 1; }else{ $file_strain_middle = -1; }  # positive strain: without char c; negative strain: with char c
			    if( !$strain_order[1] ){ $file_strain_post = 1; }else{ $file_strain_post = -1; }
			}
			if( $faa_Unit->id() =~ /.+\|.+\|.+\|(\w+)\..+/ )  # faa
			{
			    push( @protein_order, $1 );
			    $file_protein_post = $protein_order[1];
			    last;
			}
		    }
		}
	    }
	    if( $Informations[$i]->[1] =~ $protein_order[1] )  # match in the middle¡Aexist pre_protein & post_protein
	    {
		if( $faa_Unit = $in_faa->next_seq() )  # a way to fix the weird problem above
		{
		    if( $ffn_Unit = $in_ffn->next_seq() )  # a way to fix the weird problem above
		    {
			if( $ffn_Unit->id() =~ /.+\:(c*).+/ )  # ffn
			{
			    push( @strain_order, $1 );
			    if( !$strain_order[0] ){ $file_strain_pre = 1; }else{ $file_strain_pre = -1; }  # positive strain: without char c; negative strain: with char c
			    if( !$strain_order[1] ){ $file_strain_middle = 1; }else{ $file_strain_middle = -1; }
			    if( !$strain_order[2] ){ $file_strain_post = 1; }else{ $file_strain_post = -1; }
			}
			if( $faa_Unit->id() =~ /.+\|.+\|.+\|(\w+)\..+/ )  # faa
			{
			    push( @protein_order, $1 );
			    $file_protein_pre = $protein_order[0];
			    $file_protein_post = $protein_order[2];
			    last;
			}
		    }
		}
		else  # match in the end¡Ano post_protein
		{
		    if( $ffn_Unit = $in_ffn->next_seq() )  {}  # a way to fix the weird problem above
		    else
		    {
			if( !$strain_order[0] ){ $file_strain_pre = 1; }else{ $file_strain_pre = -1; }  # positive strain: without char c; negative strain: with char c
			if( !$strain_order[1] ){ $file_strain_middle = 1; }else{ $file_strain_middle = -1; }
			$file_protein_pre = $protein_order[0];
			last;
		    }
		}
	    }
	}
# === === === === === END === === === === === search for proteins and gene direction from .faa file === === === === === END === === === === ===
	$temp_num = $#protein_order;
	for( $m=0;$m<3;$m++ )  # clear
	{
	    pop( @protein_order );
	    pop( @strain_order );
	}
	
	for( $j=$i+1;$j<=$threshold;$j++ )
	{
	    # same species & adjacent proteins
	    if( $Informations[$j]->[0] =~ $speciesID && ( $Informations[$j]->[1] =~ $file_protein_pre || $Informations[$j]->[1] =~ $file_protein_post ) )
	    {
		# filter out the overlap
		#if( ($Informations[$i]->[7] < $Informations[$j]->[7] && $Informations[$i]->[8] < $Informations[$j]->[7]) || ($Informations[$i]->[7] > $Informations[$j]->[7] && $Informations[$i]->[7] > $Informations[$j]->[8]) )
		#{
		    # gene order & gene direction
		    $K = $file_strain_middle * $Informations[$i]->[4];  # a parameter to see positive or negative strain
		    if( $K > 0 )  # read is positive strain
		    {
			if( $Informations[$i]->[7] > $Informations[$j]->[7] )  # j protein(2nd query) is ahead i protein(1st query)
			{
			    if( $K*$Informations[$j]->[4]*$file_strain_pre > 0 && $Informations[$j]->[1] =~ $file_protein_pre )  # if > 0(gene direction) & proteinID (correct gene order)
			    {
				$temp_min_E = $Informations[$i]->[3] * $Informations[$j]->[3];
				if( $temp_min_E <= $min_E )  # LCA
				{
				    $number_of_LCA++;
				    if( $min_E == 0 || $number_of_LCA >= 2 ) { $considerLCA = TRUE; }  # LCA
				    $min_E = $temp_min_E;
				    for( $k=0;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$i]->[$k]); }
				    for( $k=1;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$j]->[$k]); }
# === === === === === START === === === === === To know the taxa lineage === === === === === START === === === === ===
				    if( !$Taxon_ID{ $Min_Informations[0] } )
				    {
				    	#print FHD_without_family $species_ID . "\tcan not find taxon ID from summary.txt\n";
				    }
				    else
				    {
				    	$tax_ID = $Taxon_ID{ $Min_Informations[0] };
# === === === === === START === === === === === from Chien-Hao === === === === === START === === === === ===
					$query = $tax_ID;
					if( exists $merged{$query} )
					{
					    $query = $merged{$query};
					}
					until( $query eq '1' )
					{
					    $query = $parent{$query} ;
					    if( exists $merged{$query} )
					    {
					    	$query = $merged{$query};
					    }
					    if( $rank{$query} =~ 'species' )  { $species = $names{$query}; }
					    if( $rank{$query} =~ 'genus' )  { $genus = $names{$query}; }
					    if( $rank{$query} =~ 'family' )  { $family = $names{$query}; }
					    if( $rank{$query} =~ 'order' )  { $order = $names{$query}; }
					    if( $rank{$query} =~ 'class' )  { $class = $names{$query}; }
					    if( $rank{$query} =~ 'phylum' )  { $phylum = $names{$query}; }
					    if( $rank{$query} =~ 'superkingdom' )  { $superkingdom = $names{$query}; }
					}
					if( !$superkingdom )  { $superkingdom = "unclassified"; }
					if( !$phylum )  { $phylum = "unclassified"; }
					if( !$class )  { $class = "unclassified"; }
					if( !$order )  { $order = "unclassified"; }
					if( !$family )  { $family = "unclassified"; }
					if( !$genus )  { $genus = "unclassified"; }
					if( !$species )  { $species = "unclassified"; }
# === === === === === END === === === === ===== from Chien-Hao ===== === === === === END === === === === ===
					$Domain{ $superkingdom }++;
					$Phylum{ $phylum }++;
					$Class{ $class }++;
					$Order{ $order }++;
					$Family{ $family }++;
					$Genus{ $genus }++;
					$Species{ $species }++;
					$lineage = $superkingdom."\t".$phylum."\t".$class."\t".$order."\t".$family."\t".$genus."\t".$species;
					$superkingdom = $phylum = $class = $order = $family = $genus = $species = "";
					#print $lineage."\n";
				    }
# === === === === === END === === === === === To know the taxa lineage === === === === === END === === === === ===
				    $got_it = TRUE;
				    last;  # next!
				}
			    }
			}
			elsif( $Informations[$i]->[7] < $Informations[$j]->[7] )  # j protein(2nd query)is behind i protein(1st query)
			{
			    if( $K*$Informations[$j]->[4]*$file_strain_post > 0 && $Informations[$j]->[1] =~ $file_protein_post )  # if > 0(gene direction) & proteinID (correct gene order)
			    {
				$temp_min_E = $Informations[$i]->[3] * $Informations[$j]->[3];
				if( $temp_min_E <= $min_E )  # LCA
				{
				    $number_of_LCA++;
				    if( $min_E == 0 || $number_of_LCA >= 2 ) { $considerLCA = TRUE; }  # LCA
				    $min_E = $temp_min_E;
				    for( $k=0;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$i]->[$k]); }
				    for( $k=1;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$j]->[$k]); }
# === === === === === START === === === === === To know the taxa lineage === === === === === START === === === === ===
				    if( !$Taxon_ID{ $Min_Informations[0] } )
				    {
				    	#print FHD_without_family $species_ID . "\tcan not find taxon ID from summary.txt\n";
				    }
				    else
				    {
				    	$tax_ID = $Taxon_ID{ $Min_Informations[0] };
# === === === === === START === === === === === from Chien-Hao === === === === === START === === === === ===
					$query = $tax_ID;
					if( exists $merged{$query} )
					{
					    $query = $merged{$query};
					}
					until( $query eq '1' )
					{
					    $query = $parent{$query} ;
					    if( exists $merged{$query} )
					    {
					    	$query = $merged{$query};
					    }
					    if( $rank{$query} =~ 'species' )  { $species = $names{$query}; }
					    if( $rank{$query} =~ 'genus' )  { $genus = $names{$query}; }
					    if( $rank{$query} =~ 'family' )  { $family = $names{$query}; }
					    if( $rank{$query} =~ 'order' )  { $order = $names{$query}; }
					    if( $rank{$query} =~ 'class' )  { $class = $names{$query}; }
					    if( $rank{$query} =~ 'phylum' )  { $phylum = $names{$query}; }
					    if( $rank{$query} =~ 'superkingdom' )  { $superkingdom = $names{$query}; }
					}
					if( !$superkingdom )  { $superkingdom = "unclassified"; }
					if( !$phylum )  { $phylum = "unclassified"; }
					if( !$class )  { $class = "unclassified"; }
					if( !$order )  { $order = "unclassified"; }
					if( !$family )  { $family = "unclassified"; }
					if( !$genus )  { $genus = "unclassified"; }
					if( !$species )  { $species = "unclassified"; }
# === === === === === END === === === === ===== from Chien-Hao ===== === === === === END === === === === ===
					$Domain{ $superkingdom }++;
					$Phylum{ $phylum }++;
					$Class{ $class }++;
					$Order{ $order }++;
					$Family{ $family }++;
					$Genus{ $genus }++;
					$Species{ $species }++;
					$lineage = $superkingdom."\t".$phylum."\t".$class."\t".$order."\t".$family."\t".$genus."\t".$species;
					$superkingdom = $phylum = $class = $order = $family = $genus = $species = "";
					#print $lineage."\n";
				    }
# === === === === === END === === === === === To know the taxa lineage === === === === === END === === === === ===
				    $got_it = TRUE;
				    last;  # next!
				}
			    }
			}
		    }
		    elsif( $K < 0 )  # positive strain
		    {
			if( $Informations[$i]->[7] > $Informations[$j]->[7] )  # j protein(2nd query) is ahead i protein(1st query)
			{
			    if( $K*$Informations[$j]->[4]*$file_strain_post > 0 && $Informations[$j]->[1] =~ $file_protein_post )  # if > 0(gene direction) & proteinID (correct gene order)
			    {
				$temp_min_E = $Informations[$i]->[3] * $Informations[$j]->[3];
				if( $temp_min_E <= $min_E )  # LCA
				{
				    $number_of_LCA++;
				    if( $min_E == 0 || $number_of_LCA >= 2 ) { $considerLCA = TRUE; }  # LCA
				    $min_E = $temp_min_E;
				    for( $k=0;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$i]->[$k]); }
				    for( $k=1;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$j]->[$k]); }
# === === === === === START === === === === === To know the taxa lineage === === === === === START === === === === ===
				    if( !$Taxon_ID{ $Min_Informations[0] } )
				    {
				    	#print FHD_without_family $species_ID . "\tcan not find taxon ID from summary.txt\n";
				    }
				    else
				    {
				    	$tax_ID = $Taxon_ID{ $Min_Informations[0] };
# === === === === === START === === === === === from Chien-Hao === === === === === START === === === === ===
					$query = $tax_ID;
					if( exists $merged{$query} )
					{
					    $query = $merged{$query};
					}
					until( $query eq '1' )
					{
					    $query = $parent{$query} ;
					    if( exists $merged{$query} )
					    {
					    	$query = $merged{$query};
					    }
					    if( $rank{$query} =~ 'species' )  { $species = $names{$query}; }
					    if( $rank{$query} =~ 'genus' )  { $genus = $names{$query}; }
					    if( $rank{$query} =~ 'family' )  { $family = $names{$query}; }
					    if( $rank{$query} =~ 'order' )  { $order = $names{$query}; }
					    if( $rank{$query} =~ 'class' )  { $class = $names{$query}; }
					    if( $rank{$query} =~ 'phylum' )  { $phylum = $names{$query}; }
					    if( $rank{$query} =~ 'superkingdom' )  { $superkingdom = $names{$query}; }
					}
					if( !$superkingdom )  { $superkingdom = "unclassified"; }
					if( !$phylum )  { $phylum = "unclassified"; }
					if( !$class )  { $class = "unclassified"; }
					if( !$order )  { $order = "unclassified"; }
					if( !$family )  { $family = "unclassified"; }
					if( !$genus )  { $genus = "unclassified"; }
					if( !$species )  { $species = "unclassified"; }
# === === === === === END === === === === ===== from Chien-Hao ===== === === === === END === === === === ===
					$Domain{ $superkingdom }++;
					$Phylum{ $phylum }++;
					$Class{ $class }++;
					$Order{ $order }++;
					$Family{ $family }++;
					$Genus{ $genus }++;
					$Species{ $species }++;
					$lineage = $superkingdom."\t".$phylum."\t".$class."\t".$order."\t".$family."\t".$genus."\t".$species;
					$superkingdom = $phylum = $class = $order = $family = $genus = $species = "";
					#print $lineage."\n";
				    }
# === === === === === END === === === === === To know the taxa lineage === === === === === END === === === === ===
				    $got_it = TRUE;
				    last;  # next!
				}
			    }
			}
			elsif( $Informations[$i]->[7] < $Informations[$j]->[7] )  # j protein(2nd query) is behind i protein(1st query)
			{
			    if( $K*$Informations[$j]->[4]*$file_strain_pre > 0 && $Informations[$j]->[1] =~ $file_protein_pre )  # if > 0(gene direction) & proteinID (correct gene order)
			    {
				$temp_min_E = $Informations[$i]->[3] * $Informations[$j]->[3];
				if( $temp_min_E <= $min_E )  # LCA
				{
				    $number_of_LCA++;
				    if( $min_E == 0 || $number_of_LCA >= 2 ) { $considerLCA = TRUE; }  # LCA
				    $min_E = $temp_min_E;
				    for( $k=0;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$i]->[$k]); }
				    for( $k=1;$k<11;$k++ )  { shift(@Min_Informations);  push(@Min_Informations, $Informations[$j]->[$k]); }
# === === === === === START === === === === === To know the taxa lineage === === === === === START === === === === ===
				    if( !$Taxon_ID{ $Min_Informations[0] } )
				    {
				    	#print FHD_without_family $species_ID . "\tcan not find taxon ID from summary.txt\n";
				    }
				    else
				    {
				    	$tax_ID = $Taxon_ID{ $Min_Informations[0] };
# === === === === === START === === === === === from Chien-Hao === === === === === START === === === === ===
					$query = $tax_ID;
					if( exists $merged{$query} )
					{
					    $query = $merged{$query};
					}
					until( $query eq '1' )
					{
					    $query = $parent{$query} ;
					    if( exists $merged{$query} )
					    {
					    	$query = $merged{$query};
					    }
					    if( $rank{$query} =~ 'species' )  { $species = $names{$query}; }
					    if( $rank{$query} =~ 'genus' )  { $genus = $names{$query}; }
					    if( $rank{$query} =~ 'family' )  { $family = $names{$query}; }
					    if( $rank{$query} =~ 'order' )  { $order = $names{$query}; }
					    if( $rank{$query} =~ 'class' )  { $class = $names{$query}; }
					    if( $rank{$query} =~ 'phylum' )  { $phylum = $names{$query}; }
					    if( $rank{$query} =~ 'superkingdom' )  { $superkingdom = $names{$query}; }
					}
					if( !$superkingdom )  { $superkingdom = "unclassified"; }
					if( !$phylum )  { $phylum = "unclassified"; }
					if( !$class )  { $class = "unclassified"; }
					if( !$order )  { $order = "unclassified"; }
					if( !$family )  { $family = "unclassified"; }
					if( !$genus )  { $genus = "unclassified"; }
					if( !$species )  { $species = "unclassified"; }
# === === === === === END === === === === ===== from Chien-Hao ===== === === === === END === === === === ===
					$Domain{ $superkingdom }++;
					$Phylum{ $phylum }++;
					$Class{ $class }++;
					$Order{ $order }++;
					$Family{ $family }++;
					$Genus{ $genus }++;
					$Species{ $species }++;
					$lineage = $superkingdom."\t".$phylum."\t".$class."\t".$order."\t".$family."\t".$genus."\t".$species;
					$superkingdom = $phylum = $class = $order = $family = $genus = $species = "";
					#print $lineage."\n";
				    }
# === === === === === END === === === === === To know the taxa lineage === === === === === END === === === === ===
				    $got_it = TRUE;
				    last;  # next!
				}
			    }
			}
		    }
		#}
	    }
	}
    }
    
    if( $got_it =~ TRUE )
    {
	print FHD_result ">".$read_ID."\t";
	
	# decide to use LCA or not
	if( $considerLCA =~ TRUE )
	{
	    print FHD_result "LCA\t";
	    for( $i=0;$i<20;$i++ )
	    {
	        print FHD_result $Min_Informations[$i]."\t";
	    }
	    print FHD_result $Min_Informations[20]."\t";
	    
	    @key_Domain = keys %Domain;
	    @key_Phylum = keys %Phylum;
	    @key_Class = keys %Class;
	    @key_Order = keys %Order;
	    @key_Family = keys %Family;
	    @key_Genus = keys %Genus;
	    @key_Species = keys %Species;
	    if( $#key_Species == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\t".$key_Class[0]."\t".$key_Order[0]."\t".$key_Family[0]."\t".$key_Genus[0]."\t".$key_Species[0]."\n"; }
	    elsif( $#key_Genus == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\t".$key_Class[0]."\t".$key_Order[0]."\t".$key_Family[0]."\t".$key_Genus[0]."\tdifferentTaxa\n"; }
	    elsif( $#key_Family == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\t".$key_Class[0]."\t".$key_Order[0]."\t".$key_Family[0]."\tdifferentTaxa\tdifferentTaxa\n"; }
	    elsif( $#key_Order == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\t".$key_Class[0]."\t".$key_Order[0]."\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\n"; }
	    elsif( $#key_Class == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\t".$key_Class[0]."\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\n"; }
	    elsif( $#key_Phylum == 0 )  { print FHD_result $key_Domain[0]."\t".$key_Phylum[0]."\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\n"; }
	    elsif( $#key_Domain == 0 )  { print FHD_result $key_Domain[0]."\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\tdifferentTaxa\n"; }
	    else  { print FHD_result "no any LCA\n"; }
	}
	else
	{
	    print FHD_result "noLCA\t";
	    for( $i=0;$i<20;$i++ )
	    {
	        print FHD_result $Min_Informations[$i]."\t";
	    }
	    print FHD_result $Min_Informations[20]."\t".$lineage."\n";
	}
	# clear Tables
	for( $i=0;$i<=$#key_Domain;$i++ ) { delete $Domain{ $key_Domain[$i] }; }
	for( $i=0;$i<=$#key_Phylum;$i++ ) { delete $Phylum{ $key_Phylum[$i] }; }
	for( $i=0;$i<=$#key_Class;$i++ ) { delete $Class{ $key_Class[$i] }; }
	for( $i=0;$i<=$#key_Order;$i++ ) { delete $Order{ $key_Order[$i] }; }
	for( $i=0;$i<=$#key_Family;$i++ ) { delete $Family{ $key_Family[$i] }; }
	for( $i=0;$i<=$#key_Genus;$i++ ) { delete $Genus{ $key_Genus[$i] }; }
	for( $i=0;$i<=$#key_Species;$i++ ) { delete $Species{ $key_Species[$i] }; }
    }
    else
    {
	print FHD_no_result $read_ID . "\t" . $read_des . "\n";
    }
    
    # clear
    $temp_num = $#Informations;
    for( $i=0;$i<=$temp_num;$i++ )  { shift( @Informations ); }  # ²MªÅ @Informations
    for( $k=0;$k<21;$k++ )  { shift(@Min_Informations); }  # ²MªÅ "Min_Informations"
    
    # log
    $count_reads++;
    if( $count_reads%100 == 0 )
    {
	print $count_reads . " reads have been analised...\n";
    }
}
close( FHD_result );
close( FHD_no_result );
close( FHD_input_reads_log );
close( FHD_ID_log );