miga-base 1.2.15.2 → 1.2.15.3

data/utils/enveomics/Scripts/Aln.cat.rb

@@ -1,221 +0,0 @@
-#!/usr/bin/env ruby
-
-# @author  Luis M. Rodriguez-R
-# @license artistic license 2.0
-
-$VERSION = 1.0
-$:.push File.expand_path('../lib', __FILE__)
-require 'enveomics_rb/enveomics'
-
-o = {
-  q: false, missing: '-', model: 'AUTO', removeinvar: false, undefined: '-.Xx?'
-}
-
-OptionParser.new do |opt|
-  cmd = File.basename($0)
-  opt.banner = <<~BANNER
-
-    [Enveomics Collection: #{cmd} v#{$VERSION}]
-
-    Concatenates several multiple alignments in FastA format into a single
-    multiple alignment.  The IDs of the sequences (or the ID prefixes, if using
-    --ignore-after) must coincide across files.
-  
-    Usage: #{cmd} [options] aln1.fa aln2.fa ... > aln.fa
-
-  BANNER
-  opt.on(
-    '-c', '--coords FILE',
-    'Output file of coordinates in RAxML-compliant format'
-  ) { |v| o[:coords] = v }
-  opt.on(
-    '-i', '--ignore-after STRING',
-    'Remove everything in the IDs after the specified string'
-  ) { |v| o[:ignoreafter] = v }
-  opt.on(
-    '-I', '--remove-invariable', 'Remove invariable sites',
-    'Note: Invariable sites are defined as columns with only one state and',
-    'undefined characters.  Additional ambiguous characters may exist and',
-    'should be declared using --undefined'
-  ) { |v| o[:removeinvar] = v }
-  opt.on(
-    '-u', '--missing-char CHAR',
-    "Character denoting missing data. By default: '#{o[:missing]}'"
-  ) do |v|
-    if v.length != 1
-      abort "-missing-char can only be denoted by single characters: #{v}"
-    end
-    o[:missing] = v
-  end
-  opt.on(
-    '-m', '--model STRING',
-    'Name of the model to use if --coords is used. See RAxML docs;',
-    'supported values in v8+ include:',
-    '~ For DNA alignments:',
-    '  "DNA[F|X]", or "DNA[F|X]/3" (to estimate rates per codon position,',
-    '  particular notation for this script)',
-    '~ General protein alignments:',
-    '  "AUTO" (default in this script), "DAYHOFF" (1978), "DCMUT" (MBE 2005;',
-    '  22(2):193-199), "JTT" (Nat 1992;358:86-89), "VT" (JCompBiol 2000;',
-    '  7(6):761-776), "BLOSUM62" (PNAS 1992;89:10915), and "LG" (MBE 2008;',
-    '  25(7):1307-1320)',
-    '~ Specialized protein alignments:',
-    '  "MTREV" (mitochondrial, JME 1996;42(4):459-468), "WAG" (globular, MBE',
-    '  2001;18(5):691-699), "RTREV" (retrovirus, JME 2002;55(1):65-73),',
-    '  "CPREV" (chloroplast, JME 2000;50(4):348-358), and "MTMAM" (nuclear',
-    '  mammal proteins, JME 1998;46(4):409-418)'
-  ) { |v| o[:model] = v }
-  opt.on(
-    '--undefined STRING',
-    'All characters to be regarded as "undefined". It should include all',
-    'ambiguous and missing data chars.  Ignored unless --remove-invariable',
-    "By default: '#{o[:undefined]}'"
-  ) { |v| o[:undefined] = v }
-  opt.on('-q', '--quiet', 'Run quietly (no STDERR output)') { o[:q] = true }
-  opt.on('-V', '--version', 'Returns version') { puts $VERSION ; exit }
-  opt.on('-h', '--help', 'Display this screen') { puts opt ; exit }
-  opt.separator ''
-end.parse!
-files = ARGV
-abort 'Alignment files are mandatory' if files.nil? || files.empty?
-$QUIET = o[:q]
-
-# Read individual gene alignments and return them as a single hash with genome
-# IDs as keys and arrays of single-line strings as values
-#
-# IDs are trimmed after the first occurrence of +ignoreafter+, if defined
-def read_alignments(files, ignoreafter = nil)
-  aln = {}
-  files.each_with_index do |file, i|
-    key = nil
-    File.open(file, 'r').each do |ln|
-      ln.chomp!
-      if ln =~ /^>(\S+)/
-        key = $1
-        key.sub!(/#{ignoreafter}.*/, '') if ignoreafter
-        aln[key] ||= []
-        aln[key][i] = ''
-      else
-        if key.nil?
-          abort "Invalid FastA file: #{file}: Leading line not a def-line"
-        end
-        ln.gsub!(/\s/, '')
-        aln[key][i] += ln
-      end
-    end
-    abort "Empty alignment file: #{file}" if key.nil?
-  end
-  aln
-end
-
-# Remove invariable sites from the alignment hash +aln+, using +undefined+ as
-# a string including all characters representing undefined positions (e.g., X)
-#
-# Returns number of columns removed
-def remove_invariable(aln, undefined)
-  invs = 0
-  lengths = aln.values.first.map(&:length)
-  undef_chars = undefined.chars
-
-  lengths.each_with_index do |len, i|
-    (0 .. len - 1).each do |pos|
-      chr = nil
-      inv = true
-      aln.each_key do |key|
-        next if aln[key][i].nil?
-        chr = aln[key][i][pos] if chr.nil? || undefined.chars.include?(chr)
-        if chr != aln[key][i][pos] && !undef_chars.include?(aln[key][i][pos])
-          inv = false
-          break
-        end
-      end
-      if inv
-        aln.each_key { |key| aln[key][i][pos] = '!' unless aln[key][i].nil? }
-        lengths[i] -= 1
-        invs += 1
-      end
-    end
-    aln.each_key { |key| aln[key][i].gsub!('!', '') unless aln[key][i].nil? }
-  end
-  invs
-end
-
-# Concatenate the alignments hash +aln+ using the character +missing+ to
-# indicate missing alignments, and send each entry in the concatenated alignment
-# to +blk+ as two variables: key (name) and value (alignment string)
-#
-# Returns an array with the lengths of each individual alignment
-def concatenate(aln, missing, &blk)
-  say 'Concatenating'
-  lengths = aln.values.first.map(&:length)
-  aln.each_key do |key|
-    # Pad missing entries
-    lengths.each_with_index { |len, i| aln[key][i] ||= missing * len }
-
-    # Check length
-    obs_len = aln[key].map(&:length)
-    unless lengths == obs_len
-      abort "Inconsistent lengths in '#{key}'\nexp: #{lengths}\nobs: #{obs_len}"
-    end
-
-    # Pass entry to the block and remove from alignment hash
-    blk[key, aln[key].join('')]
-    aln.delete(key)
-  end
-  lengths
-end
-
-# Save the coordinates in +file+ based on +files+ paths (for the names), and
-# using +lengths+ individual alignment lengths
-#
-# The saved format is RAxML coords, including the +model+ for each alignment
-def save_coords(file, names, lengths, model)
-  File.open(file, 'w') do |fh|
-    s = 0
-    names.each_with_index do |name, i|
-      l = lengths[i]
-      next unless l > 0
-      name += "_#{i}" while names.count(name) > 1
-      if model =~ /(DNA.?)\/3/
-        fh.puts "#{$1}, #{name}codon1 = #{s + 1}-#{s + l}\\3"
-        fh.puts "#{$1}, #{name}codon2 = #{s + 2}-#{s + l}\\3"
-        fh.puts "#{$1}, #{name}codon3 = #{s + 3}-#{s + l}\\3"
-      else
-        fh.puts "#{model}, #{name} = #{s + 1}-#{s + l}"
-      end
-      s += l
-    end
-  end
-end
-
-# ------ MAIN ------
-begin
-  say 'Reading'
-  alignments = read_alignments(files, o[:ignoreafter])
-
-  if o[:removeinvar]
-    say 'Removing invariable sites'
-    inv = remove_invariable(alignments, o[:undefined])
-    say "  Removed #{inv} sites"
-  end
-
-  lengths = concatenate(alignments, o[:missing]) do |name, seq|
-    puts ">#{name}", seq.gsub(/(.{1,60})/, "\\1\n")
-  end
-  say "  #{lengths.inject(:+)} columns"
-
-  unless o[:coords].nil?
-    say 'Generating coordinates'
-    names = files.map do |i|
-      File.basename(i).gsub(/\..*/, '').gsub(/[^A-Za-z0-9_]/, '_')
-    end
-    save_coords(o[:coords], names, lengths, o[:model])
-  end
-
-  $stderr.puts 'Done' unless o[:q] 
-rescue => err
-  $stderr.puts "Exception: #{err}\n\n"
-  err.backtrace.each { |l| $stderr.puts l + "\n" }
-  err
-end
-

data/utils/enveomics/Scripts/Aln.convert.pl

@@ -1,35 +0,0 @@
-#!/usr/bin/env perl
-
-#
-# @author: Luis M. Rodriguez-R
-# @update: Mar-23-2015
-# @license: artistic license 2.0
-#
-
-use Bio::AlignIO;
-
-my($iformat,$oformat) = @ARGV;
-($iformat and $oformat) or die "
-Usage:
-   $0 in-format out-format < in_file > output_file
-
-   in-format	Input file's format.
-   out-format	Output file's format.
-   in_file	Input file.
-   out_file	Output file.
-
-Example:
-   # Re-format example.fa into Stockholm
-   $0 fasta stockholm < example.fa > example.stk
-
-Supported formats are:
-   bl2seq, clustalw, emboss, fasta, maf, mase, mega,
-   meme, metafasta, msf, nexus, pfam, phylip, po,
-   prodom, psi, selex, stockholm, XMFA, arp
-
-";
-
-$in  = Bio::AlignIO->new(-fh => \*STDIN, -format => $iformat);
-$out = Bio::AlignIO->new(-fh => \*STDOUT, -format => $oformat);
-while ( my $aln = $in->next_aln ) { $out->write_aln($aln) }
-

data/utils/enveomics/Scripts/AlphaDiversity.pl

@@ -1,152 +0,0 @@
-#!/usr/bin/env perl
-
-#
-# @author: Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
-# @update: Nov-29-2015
-# @license: artistic license 2.0
-#
-
-use warnings;
-use strict;
-use Getopt::Std;
-use List::Util qw/min max sum/;
-
-sub VERSION_MESSAGE(){print "Alpha-diversity indices (enveomics)\n"}
-sub HELP_MESSAGE(){die "
-Description:
-   Takes a table of OTU abundance in one or more samples and calculates the Rao
-   (Q_alpha), Rao-Jost (Q_alpha_eqv), Shannon (Hprime), and inverse Simpson
-   (1_lambda) indices of alpha diversity for each sample.
-   
-   To use it with Qiime OTU Tables, run it as:
-   $0 -i OTU_Table.txt -h
-
-Usage:
-   $0 [opts]
-
-   -i <str>	* Input table (columns:samples, rows:OTUs, first column:OTU
-		names).
-   -r <int>	Number of rows to ignore.  By default: 0.
-   -c <int>	Number of columns to ignore after the first column (i.e.,
-		between the first column, containing the name of the categories,
-		and the first column of abundance values).  By default: 0.
-   -C <int>	Number of columns to ignore at the end of each line.
-		By default: 0.
-   -d <str>	Delimiter.  Supported escaped characters are: \"\\t\"
-		(tabulation), and \"\\0\" (null bit).  By default: \"\\t\".
-   -h		If set, the first row is assumed to have the names of the
-		samples.
-   -D <str>	Distances file.  A squared matrix (or bottom-left half matrix)
-		with the distances between categories (OTUs or functions).  The
-		first column must contain the names of the categories, and it
-		shouldn't have headers.  If not set, all distances are assumed
-		to be one.  Only used for Rao.
-   -R		Do not calculate Rao indices.  This significantly decreases the
-		total running time. Note that Rao indices are highly susceptible
-		to precision errors, and shouldn't be trusted for very big
-		numbers.
-   -q <int>	Estimate the qD index (true diversity order q).  By default: 0.
-   --help	This help message.
-
-   * Mandatory.
-
-"}
-
-# Input arguments
-my %o;
-getopts('i:c:C:d:r:hD:Rq:', \%o);
-
-#$o{B} and (eval("use bignum; 1") or die "Cannot use bignum.\n");
-&HELP_MESSAGE() unless $o{i};
-$o{c} ||= 0;
-$o{C} ||= 0;
-$o{r} ||= 0;
-$o{d} ||= "\\t";
-$o{q} ||= 0;
-
-$o{d}="\t" if $o{d} eq "\\t";
-$o{d}="\0" if $o{d} eq "\\0";
-
-# Distance matrix
-my $D = {};
-if($o{D} and not $o{R}){
-   my @Didx = ();
-   open DIST, "<", $o{D} or die "Cannot read file: $o{D}: $!\n";
-   while(<DIST>){
-      chomp;
-      my @d = split /\t/;
-      my $idx = shift @d;
-      push  @Didx,  $idx;
-      $D->{ $idx } ||= {};
-      $D->{ $idx }->{ $Didx[$_] } = $d[$_] for(0 .. $#d);
-   }
-   close DIST;
-   undef @Didx;
-}
-
-# Abundance matrix
-my @names  = ();
-my @cats   = ();
-my @values = ();
-open TABLE, "<", $o{i} or die "Cannot open file: ".$o{i}.": $!\n";
-<TABLE> for (1 .. $o{r});
-if($o{h}){
-   my $h = <TABLE>;
-   $h or die "Empty table!\n";
-   chomp $h;
-   @names = split $o{d}, $h;
-   shift @names for (0 .. $o{c});
-}
-
-while(<TABLE>){
-   chomp;
-   my @ln = split $o{d};
-   push @cats, shift(@ln);
-   shift @ln for (1 .. $o{c});
-   pop @ln for (1 .. $o{C});
-   push @{$values[$_] ||= []}, $ln[$_] for (0 .. $#ln);
-   push @{$values[$#ln+1]}, sum(@ln);
-}
-close TABLE;
-$names[$#values] = "gamma";
-
-if($o{R}){
-   print "".join($o{d}, qw/Sample Hprime 1_lambda qD/)."\n";
-}else{
-   print "".join($o{d}, qw/Sample Q_alpha Q_alpha_eqv Hprime 1_lambda qD/)."\n";
-}
-for my $i (0 .. $#values){
-   print "".(exists $names[$i] ? $names[$i] : $i).$o{d};
-   my $N = sum @{$values[$i]};
-   my $Q = 0;
-   my $H = 0;
-   my $l = 0;
-   my $qD = 0 unless $o{q}==1;
-   for my $ik (0 .. $#{$values[$i]}){
-      unless($o{R}){
-	 my $Qi = 0;
-	 for my $jk (0 .. $#{$values[$i]}){
-	    my $dij = (!$o{D}) ? 1 : 
-	       exists $D->{ $cats[$ik] }->{ $cats[$jk] } ?
-	       $D->{ $cats[$ik] }->{ $cats[$jk] } :
-	       exists $D->{ $cats[$jk] }->{ $cats[$ik] } ?
-	       $D->{ $cats[$jk] }->{ $cats[$ik] } :
-	       die "Cannot find distance between ".$cats[$ik].
-		  " and ".$cats[$jk].".\n";
-	    $Qi += $dij * ($values[$i]->[$ik]/$N) * ($values[$i]->[$jk]/$N);
-	 }
-	 $Q += $Qi;
-      }
-      my $pi = $N ? $values[$i]->[$ik]/$N : 0;
-      $H  -= $pi * log($pi) if $pi;
-      $l  += $pi**2;
-      $qD += $pi * ($pi**($o{q}-1)) unless $o{q}==1 or not $pi;
-   }
-   $qD = $o{q}==1 ? exp($H) : 1/($qD**(1/($o{q}-1)));
-   if($o{R}){
-      print "".join($o{d}, $H, $l ? 1/$l : "Inf", $qD)."\n";
-   }else{
-      print "".join($o{d}, $Q, ($Q==1 ? "NA" : 1/(1-$Q)), $H, 1/$l, $qD)."\n";
-   }
-}
-

data/utils/enveomics/Scripts/BedGraph.tad.rb

@@ -1,93 +0,0 @@
-#!/usr/bin/env ruby
-
-require 'optparse'
-
-o = {range: 0.5, perseq: false, length: false}
-ARGV << '-h' if ARGV.empty?
-OptionParser.new do |opt|
-  opt.banner = "
-  Estimates the truncated average sequencing depth (TAD) from a BedGraph file.
-
-  IMPORTANT: This script doesn't consider zero-coverage positions if missing
-  from the file. If you produce your BedGraph file with bedtools genomecov and
-  want to consider zero-coverage position, be sure to use -bga (not -bg).
-
-  Usage: #{$0} [options]"
-  opt.separator ''
-  opt.on('-i', '--input PATH',
-    'Input BedGraph file (mandatory).'){ |v| o[:i]=v }
-  opt.on('-r', '--range FLOAT',
-    'Central range to consider, between 0 and 1.',
-    "By default: #{o[:range]} (inter-quartile range)."
-    ){ |v| o[:range]=v.to_f }
-  opt.on('-s', '--per-seq',
-    'Calculate averages per reference sequence, not total.',
-    'Assumes a sorted BedGraph file.'
-    ){ |v| o[:perseq] = v }
-  opt.on('-l', '--length',
-    'Add sequence length to the output.'){ |v| o[:length] = v }
-  opt.on('-h', '--help', 'Display this screen.') do
-    puts opt
-    exit
-  end
-  opt.separator ''
-end.parse!
-abort '-i is mandatory.' if o[:i].nil?
-
-def pad(d, idx, r)
-  idx.each do |i|
-    next if d[i].nil?
-    d[i] -= r
-    break unless d[i] < 0
-    r = -d[i]
-    d[i] = nil
-  end
-  d
-end
-
-def report(sq, d, ln, o)
-  # Estimate padding ranges
-  pad = (1.0-o[:range])/2.0
-  r = (pad*ln).round
-
-  # Pad
-  d = pad(d, d.each_index.to_a, r+0)
-  d = pad(d, d.each_index.to_a.reverse, r+0)
-
-  # Average
-  y = [0.0]
-  unless d.compact.empty?
-    s = d.each_with_index.to_a.map{ |v,i| v.nil? ? 0 : i*v }.inject(0,:+)
-    y[0] = s.to_f/d.compact.inject(:+)
-  end
-
-  # Report
-  y.unshift(sq) if o[:perseq]
-  y << ln if o[:length]
-  puts y.join("\t")
-end
-
-# Read BedGraph
-d  = []
-ln = 0
-pre_sq = nil
-File.open(o[:i], "r") do |ifh|
-  ifh.each_line do |i|
-    next if i =~ /^#/
-    r  = i.chomp.split("\t")
-    sq = r.shift
-    if o[:perseq] and !pre_sq.nil? and pre_sq!=sq
-      report(pre_sq, d, ln, o)
-      d  = []
-      ln = 0
-    end
-    r.map! { |j| j.to_i }
-    l = r[1]-r[0]
-    d[ r[2] ] ||= 0
-    d[ r[2] ]  += l
-    ln += l
-    pre_sq = sq
-  end
-end
-report(pre_sq, d, ln, o)
-

data/utils/enveomics/Scripts/BedGraph.window.rb

@@ -1,71 +0,0 @@
-#!/usr/bin/env ruby
-
-require 'optparse'
-
-o = {win: 1000}
-ARGV << '-h' if ARGV.empty?
-OptionParser.new do |opt|
-  opt.banner = "
-  Estimates the sequencing depth per windows from a BedGraph file.
-
-  IMPORTANT: This script doesn't consider zero-coverage positions if missing
-  from the file. If you produce your BedGraph file with bedtools genomecov and
-  want to consider zero-coverage position, be sure to use -bga (not -bg).
-
-  Usage: #{$0} [options]"
-  opt.separator ''
-  opt.on('-i', '--input PATH',
-    'Input BedGraph file (mandatory).'){ |v| o[:i]=v }
-  opt.on('-w', '--win INT',
-    'Window size, in base pairs.', "By default: #{o[:win]}."
-    ){ |v| o[:win]=v.to_i }
-  opt.on('-h', '--help', 'Display this screen.') do
-    puts opt
-    exit
-  end
-  opt.separator ''
-end.parse!
-abort '-i is mandatory.' if o[:i].nil?
-
-def report(d, a, b, seqs)
-  # Average
-  y = 0.0
-  unless d.compact.empty?
-    s = d.each_with_index.to_a.map{ |v,i| v.nil? ? 0 : i*v }.inject(0,:+)
-    y = s.to_f/d.compact.inject(:+)
-  end
-
-  # Report
-  puts [a, b, y, seqs.keys.join(",")].join("\t")
-end
-
-# Read BedGraph
-d  = []
-ln = 0
-a = 1
-seqs = {}
-b = o[:win]
-File.open(o[:i], "r") do |ifh|
-  ifh.each_line do |i|
-    next if i =~ /^#/
-    r  = i.chomp.split("\t")
-    sq = r.shift
-    seqs[sq] = 1
-    r.map!{ |j| j.to_i }
-    l = r[1]-r[0]
-    d[ r[2] ] ||= 0
-    d[ r[2] ]  += l
-    ln += l
-    while ln >= b
-      d[ r[2] ] -= (ln-b)
-      report(d, a, b, seqs)
-      seqs = {}
-      seqs[ sq ] = 1 if ln > b
-      d = []
-      d[ r[2] ] = (ln-b)
-      a = b + 1
-      b = a + o[:win] - 1
-    end
-  end
-end
-

data/utils/enveomics/Scripts/BlastPairwise.AAsubs.pl

@@ -1,102 +0,0 @@
-#!/usr/bin/env perl
-# 
-# @authors Konstantinos Konstantinidis (initial version)
-#          modified to work with the BLASTp 2.2.25+ m0 output by
-#          Despina Tsementzi & Luis M. Rodriguez-R
-# @updated Dec-21-2015
-#
-
-
-$/ = "Lambda     ";
-use strict;
-my %hash_depth;
-
-my @query;
-my @subject;
-my @similarity;
-my $length = "0";
-
-my($cigar_chr, $blast) = @ARGV;
-
-($cigar_chr and $blast) or die "
-.Description:
-   Counts the different AA substitutions in the best hit blast alignments, from
-   a BLASTP pairwise format output (-outfmt 0 in BLAST+, -m 0 in legacy BLAST).
-
-.Usage: $0 cigar_char blast.m0.txt > aa-subs.list
-
-   cigar_char    Use '+' for similar substitutions, use '_' for non similar
-                 substitutions
-   blast.m0.txt  Blast in 'text' format (-outfmt/-m 0).
-   aa-subs.list  A tab-delimited raw file with one substitution per row and
-                 columns:
-                 (1) Name-of-query_Name-of-subject
-		 (2) AA-in-subject
-		 (3) AA-in-query
-		 (4) Total-Align-Length
-
-";
-
-# For each blast result (i.e., for each query)
-open BLAST, "<", $blast or die "Cannot read file: $blast: $!\n";
-while(my $data=<BLAST>) {
-   $data =~ s/\r//g;
-   my ($data_q, @array_matches) = split(/>/,$data);
-   my ($name_query) = ($data_q =~ /Query\= (\S+?)(?:_GENE|\s)/);
-   my ($length_query) = ($data_q =~ /\(([\d,]+) letters/ );
-   ($length_query) = ($data_q =~ /Length=([\d,]+)/) unless $length_query;
-   $length_query =~ tr/,//d;
-   
-   # For each alignment (i.e., for each HSP),
-   # note the "last" at the end of the block,
-   # so only the best match is considered
-   foreach my $data_f (@array_matches) {
-      # Capture statistics
-      my ($length_match) = ($data_f =~ /Identities = \d+\/(\d+)/);
-      my ($identity_match) = ($data_f =~ /Identities = \d+\/\d+ \((\d+)%/);
-      my ($target_name) = ($data_f =~ /^\s?(\S+)/);
-
-      # If the alignment meets minimum requirements
-      if ($length_query >30 && ($length_match/$length_query > 0.7) && $identity_match > 60) {
-	 $data_f =~ tr/ /_/;
-	 my @array = split ("\n", $data_f);
-	 my $blanks = 0;
-	 my $prefix_size = 0;
-
-	 # For each line in the alignment
-	 for my $data_fff (@array) {	
-	    if ($data_fff =~ /(Query[:_]_+\d+_+)([^_]+)/){
-	       # Query lines
-	       $prefix_size = length($1);
-	       $length = length($2);
-	       @query = split (//, $2);
-	    }elsif ($data_fff =~ /^_{11}/){
-	       # Cigar lines
-	       @similarity = split(//, substr($data_fff, $prefix_size, $length));
-	    }elsif ($data_fff =~ /Sbjct[:_]_+\d+_+([^_]+)/){
-	       # Subject lines
-	       @subject = split(//, $1);
-	       # For each alignment column
-	       for(my $i=0; $i <= $length; $i++){
-		  if ($similarity[$i] eq $cigar_chr) {
-		     print "$name_query\_$target_name\t$subject[$i]\t$query[$i]\t$length_match\n";
-		  }
-	       }
-	       undef @query;
-	       undef @similarity;
-	       undef @subject; 
-	    }
-	    
-	    # Remove secondary alignments
-	    if ($data_fff =~ /^$/){
-	       $blanks++;
-	       last if $blanks >= 3;
-	    }else{
-	       $blanks=0;
-	    }
-	 } # for my $data_fff (@array)
-      } # if ($length_query >30 ...
-      last; # <---- So it takes only the best match!
-   } # foreach my $data_f (@array_matches)
-} # while(my $data=<>)
-