miga-base 0.3.0.0 → 0.3.0.1

data/utils/enveomics/Scripts/FastQ.interpose.pl

@@ -0,0 +1,90 @@
+#!/usr/bin/env perl
+# 
+# @author Luis M. Rodriguez-R
+# @version 2.0
+# @update: Mar-23-2015
+# @license artistic license 2.0
+# 
+# Usage: FastQ.interpose.pl <output_fastq> <input_fastq_1> <input_fastq_2> [additional input files...]
+
+use strict;
+use warnings;
+use Symbol;
+
+my $HELP = <<HELP
+
+  Description:
+    Interposes sequences in FastQ format from two files into one output file.  If more than two files are
+    provided, the script will interpose all the input files.
+    Note that this script will check for the consistency of the names (assuming a pair of related reads
+    contains the same name varying only in a trailing slash (/) followed by a digit.  If you want to turn
+    this feature off just set the -T option to zero.  If you want to decrease the sampling period
+    (to speed the script up) or increase it (to make it more sensitive to errors) just change -T option
+    accordingly.
+     
+  Usage:
+     $0 [-T <int> ]<output_fastq> <input_fastq_1> <input_fastq_2> [additional input files...]
+
+  Where,
+     -T <int>		: Optional.  Integer indicating the sampling period for names evaluation (see
+     			  Description above).  By default: 1000.
+     output_fastq	: Output file
+     input_fastq_1	: First FastQ file
+     input_fastq_2	: Second FastQ file
+     ...		: Any additional FastQ files (or none)
+
+HELP
+;
+my $eval_T = 1000;
+if(exists $ARGV[0] and exists $ARGV[1] and $ARGV[0] eq '-T'){
+   $eval_T = $ARGV[1]+0;
+   shift @ARGV;
+   shift @ARGV;
+}
+my $out = shift @ARGV;
+my @in = @ARGV;
+
+die $HELP unless $out and $#in >= 1;
+open OUT, ">", $out or die "Unable to write on $out: $!\n";
+print "Output file: $out\n";
+
+my @in_fh = ();
+
+for my $k (0 .. $#in) {
+   $in_fh[$k] = gensym;
+   open $in_fh[$k], "<", $in[$k] or die "Unable to read $in[$k]: $!\n";
+   print "Input file: $in[$k]\n";
+}
+
+my $i = 0;
+my $frl;
+LINE: while(1){
+   my $name = "";
+   print STDERR "\rEntry: $i    " unless $i % 1000;
+   FILE: for my $k (0 .. $#in_fh){
+      my @ln = ();
+      for my $l (0 .. 3){
+	 $ln[$l] = readline($in_fh[$k]);
+	 last LINE if $k==0 and $l==0 and (not defined $ln[$l]);
+	 defined $ln[$l] or die "Impossible to read next entry (line $.) from $in[$k]: $!\n";
+	 chomp $ln[$l];
+      }
+      if($eval_T and not $i % $eval_T){
+	 $ln[0] =~ m/^\@(.*?)\/\d+\s*$/ or die "Impossible to evaluate names!\n offending entry:\n$ln[0]\n";
+	 $name ||= $1;
+	 die "Inconsistent name!\n base name is $name\n offending entry is:\n$ln[0]\n" unless $1 eq $name;
+      }
+      unless($frl){
+	 $ln[0] =~ /^@/ or die "Unexpected format! (missing @)\n offending entry: $ln[0].\n";
+         $ln[2] =~ /^\+/ or die "Unexpected format! (missing +)\n offending entry: $ln[0].\n";
+	 $frl = length $ln[1];
+      }
+      print OUT "".join("\n", @ln, "");
+   }
+   $i++;
+}
+print "\rNumber of entries: $i     \nFirst read length: $frl\n";
+close OUT;
+
+for my $k(0..$#in_fh){print "ALERT: The file $in[$k] contains trailing entries\n" if defined readline($in_fh[$k])}
+

data/utils/enveomics/Scripts/FastQ.offset.pl

@@ -0,0 +1,90 @@
+#!/usr/bin/env perl
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @update Mar-23-2015
+# @license artistic license 2.0
+#
+
+use warnings;
+use strict;
+
+my ($in, $off, $force) = @ARGV;
+$in or die "
+.Description:
+   There are several FastQ formats (see http://en.wikipedia.org/wiki/FASTQ_format).
+   This script takes a FastQ in any of them, identifies the type of FastQ (this is,
+   the offset), and generates a FastQ with the given offset.  Note that Solexa+64
+   FastQ can cause problematic values when using the offset 33, since there is no
+   equivalent in Phred+33 for negative values (the range of Solexa+64 is -5 to 40).
+
+.Usage:
+   $0 in.fastq[ offset[ force]] > out.fastq
+
+   in.fastq	Input file in FastQ format (range is automatically detected).
+   offset	(optional) Offset to use for the output.  Use 0 (zero) to detect
+   		the input format and exit.  By default: 33.
+   force	(optional) If true, turns errors into warnings and continues.
+   		Out-of-range values are set to the closest range limit.
+   out.fastq	Output file in FastQ format with the specified offset.
+
+";
+
+$off = 33 unless defined $off;
+
+my $in_off = 0;
+open IN, "<", $in or die "Cannot read file: $in: $!\n";
+GUESS_FORMAT: while(<IN>){
+   unless($.%4){
+      chomp;
+      for my $chr (split //){
+         my $o = ord $chr;
+	 if($o < 55){
+	    $in_off = 33;
+	    last GUESS_FORMAT;
+	 }elsif($o > 80){
+	    $in_off = 64;
+	    last GUESS_FORMAT;
+	 }
+      }
+   }
+}
+close IN;
+print STDERR "Detected input offset: Phred+$in_off\n";
+exit unless $off;
+
+my $Solexa64=0;
+die "Couldn't guess input format.\n" unless $in_off;
+open IN, "<", $in or die "Cannot read file: $in: $!\n";
+while(<IN>){
+   if($in_off==$off or $.%4){
+      print $_;
+   }else{
+      chomp;
+      for my $chr (split //){
+         my $score = ord($chr) - $in_off;
+	 my $err = '';
+	 if($score < -5){
+	    $err = "Out-of-range value $chr ($score) in line $..\n";
+	    $score = $off==64 ? -5 : 0;
+	 }elsif(!$Solexa64 and $score < 0){
+	    if($in_off==64){
+	       print STDERR "Format variant: Solexa+64\n";
+	       $Solexa64 = 1;
+	    }else{
+	       $err = "Out-of-range value $chr ($score) in line $..\n";
+	       $score = 0;
+	    }
+	 }elsif($score>41){
+	    $err = "Out-of-range value $chr ($score) in line $..\n";
+	    $score = 41;
+	 }
+	 if($err){
+	    if($force){ warn $err } else { die $err }
+	 }
+	 print chr( $score + $off );
+      }
+      print "\n";
+   }
+}
+close IN;
+

data/utils/enveomics/Scripts/FastQ.split.pl

@@ -0,0 +1,53 @@
+#!/usr/bin/env perl
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @license artistic license 2.0
+# @update Jul-05-2015
+#
+
+use warnings;
+use strict;
+use Symbol;
+
+my ($file, $base, $outN) = @ARGV;
+
+$outN ||= 2;
+($file and $base) or die "
+Usage
+   $0 in_file.fq out_base[ no_files]
+   
+   in_file.fq	Input file in FastA format.
+   out_base	Prefix for the name of the output files.  It will
+   		be appended with .<i>.fastq, where <i> is a consecutive
+		number starting in 1.
+   no_files	Number of files to generate.  By default: 2.
+
+";
+
+
+my @outSym = ();
+for my $i (1 .. $outN){
+   $outSym[$i-1] = gensym;
+   open $outSym[$i-1], ">", "$base.$i.fastq" or die "I can not create the file: $base.$i.fa: $!\n";
+}
+
+
+my($i, $seq) = (-1, '');
+open FILE, "<", $file or die "I can not read the file: $file: $!\n";
+while(my $ln=<FILE>){
+   if($.%4 == 1){
+      print { $outSym[$i % $outN] } $seq if $seq;
+      $i++;
+      $seq = '';
+   }
+   $seq.=$ln;
+}
+print { $outSym[$i % $outN] } $seq if $seq;
+close FILE;
+
+for(my $j=0; $j<$outN; $j++){
+   close $outSym[$j];
+}
+
+print STDERR "Sequences: ".($i+1)."\nFiles: $outN\n";
+

data/utils/enveomics/Scripts/FastQ.tag.rb

@@ -0,0 +1,63 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @update: Feb-06-2015
+# @license: artistic license 2.0
+#
+
+require 'optparse'
+
+o = {:q=>FALSE, :p=>"", :s=>""}
+ARGV << '-h' if ARGV.size==0
+OptionParser.new do |opts|
+   opts.banner = "
+Generates easy-to-parse tagged reads from FastQ files.
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-i", "--in FILE", "Path to the FastQ file containing the sequences."){ |v| o[:in] = v }
+   opts.on("-o", "--out FILE", "Path to the FastQ to create."){ |v| o[:out] = v }
+   opts.separator ""
+   opts.separator "ID options"
+   opts.on("-p", "--prefix STR", "Prefix to use in all IDs."){ |v| o[:p] = v }
+   opts.on("-s", "--suffix STR", "Suffix to use in all IDs."){ |v| o[:s] = v }
+   opts.separator ""
+   opts.separator "Other Options"
+   opts.on("-q", "--quiet", "Run quietly (no STDERR output)"){ o[:q] = TRUE }
+   opts.on("-h", "--help", "Display this screen") do
+      puts opts
+      exit
+   end
+   opts.separator ""
+end.parse!
+abort "-i is mandatory" if o[:in].nil?
+abort "-o is mandatory" if o[:out].nil?
+   
+begin
+   ifh = File.open(o[:in], 'r');
+   ofh = File.open(o[:out], 'w');
+   i=0
+   while ln=ifh.gets
+      ln.chomp!
+      if $.%4==1 and not /^@/.match(ln).nil?
+	 i+=1
+	 ofh.puts "@#{o[:p]}#{i}#{o[:s]}"
+      elsif $.%4==2 or $.%4==0
+         ofh.puts ln
+      elsif $.%4==3 and not /^\+/.match(ln).nil?
+         ofh.puts "+"
+      else
+         abort "Impossible to parse line #{$.}: #{ln}.\n"
+      end
+   end
+   ifh.close
+   ofh.close
+rescue => err
+   $stderr.puts "Exception: #{err}\n\n"
+   err.backtrace.each { |l| $stderr.puts l + "\n" }
+   err
+end
+
+

data/utils/enveomics/Scripts/FastQ.toFastA.awk

@@ -0,0 +1,24 @@
+#!/usr/bin/env awk -f
+#
+# @author  Luis M. Rodriguez-R
+# @update  Dec-26-2015
+# @license artistic license 2.0
+#
+
+BEGIN {
+  for (i = 0; i < ARGC; i++) {
+     if(ARGV[i] == "--help"){
+       print "Description:\n"
+       print "  Translates FastQ files into FastA.\n"
+       print "Usage:\n"
+       print "  FastQ.toFastA.awk < in.fq > out.fa\n"
+       exit
+     }
+  }
+}
+
+NR%4 == 1, NR%4 == 2 {
+   if(NR%4 == 1){ gsub(/^@/,">") }
+   print $0
+}
+

data/utils/enveomics/Scripts/GenBank.add_fields.rb

@@ -0,0 +1,84 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @update: Feb-06-2015
+# @license: artistic license 2.0
+#
+
+require 'optparse'
+
+o = {:q=>FALSE, :k=>1, :split=>"#"}
+ARGV << '-h' if ARGV.size==0
+OptionParser.new do |opts|
+   opts.banner = "
+Adds annotations to GenBank files.
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-g", "--genbank FILE", "Input GenBank file."){ |v| o[:gb]=v }
+   opts.on("-t", "--table FILE", "Input file containing the annotations. It must be a ",
+   				"tab-delimited raw table including a header row with ",
+				"the names of the fields."){ |v| o[:table]=v }
+   opts.on("-o", "--out FILE", "Output file containing the annotated GenBank."){ |v| o[:out]=v }
+   opts.separator ""
+   opts.separator "Other Options"
+   opts.on("-k", "--key NUMBER", "Key of the column to use as identifier. By default: #{o[:k]}"){ |v| o[:k] = v.to_i }
+   opts.on("-s", "--split STRING", "String that separates multiple entries in the annotation features. By default: \"#{o[:split]}\""){ |v| o[:k] = v.to_i }
+   opts.on("-q", "--quiet", "Run quietly (no STDERR output)."){ o[:q] = TRUE }
+   opts.on("-h", "--help", "Display this screen.") do
+      puts opts
+      exit
+   end
+   opts.separator ""
+end.parse!
+abort "-g is mandatory" if o[:gb].nil?
+abort "-t is mandatory" if o[:table].nil?
+abort "-o is mandatory" if o[:out].nil?
+
+##### MAIN:
+begin
+   puts "Reading annotation table: #{o[:table]}." unless o[:q]
+   ifh = File.open(o[:table], "r")
+   header = ifh.gets.chomp.split(/\t/)
+   puts "  * using #{header[ o[:k]-1 ]} column as feature identifier."
+   annot = {}
+   while ln=ifh.gets
+      row = ln.chomp.split(/\t/)
+      warn "WARNING: #{header[ o[:k]-1 ]} #{row[ o[:k]-1 ]} found more than once." unless annot[ row[ o[:k]-1 ] ].nil?
+      annot[ row[ o[:k]-1 ] ] = row
+   end
+   ifh.close
+   puts "  * found #{annot.size} annotation entries with #{header.size} fields." unless o[:q]
+   puts "Annotating GenBank." unless o[:q]
+   ifh = File.open(o[:gb], "r")
+   ofh = File.open(o[:out], "w")
+   found = 0
+   notfound = 0
+   while ln=ifh.gets
+      ofh.print ln
+      m = /^(?<sp>\s+)\/#{header[ o[:k]-1 ]}="(?<id>.+)"/.match(ln)
+      next if m.nil?
+      if annot[ m[:id] ].nil?
+	 notfound += 1
+	 next
+      end
+      found += 1
+      annot[ m[:id] ].each_index do |i|
+         next if i == o[:k]-1 or annot[ m[:id] ][i]==""
+	 annot[ m[:id] ][i].split(/#{o[:split]}/).each{ |v| ofh.puts "#{m[:sp]}/#{header[i]}=\"#{v}\"" }
+      end
+   end
+   ofh.close
+   ifh.close
+   puts "  * annotated #{found} features." unless o[:q]
+   puts "  * couldn't find #{notfound} features in the annotation table." unless o[:q] or notfound==0
+   $stderr.puts "Done.\n" 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/HMM.essential.rb

@@ -0,0 +1,254 @@
+#!/usr/bin/env ruby
+
+#
+# @author  Luis M. Rodriguez-R
+# @license artistic license 2.0
+# @update  Mar-23-2016
+#
+
+$:.push File.expand_path("../lib", __FILE__)
+require "enveomics_rb/enveomics"
+use "tmpdir"
+use "zlib"
+
+o = {bin:"", thr:2, q:false, stats:true, genes:true, bacteria:false,
+   archaea:false, genomeeq:false, metagenome:false, list:false}
+OptionParser.new do |opts|
+   opts.banner = "
+Finds and extracts a collection of essential proteins suitable for genome
+completeness evaluation and phylogenetic analyses. Important note: most complete
+bacterial genomes contain only 106/111 genes in this collection, therefore
+producing a completeness of 95.5%, and most archaeal genomes only contain 26/111
+genes, producing a completeness of 23.4%. Use the options --bacteria and/or 
+--archaea to ignore models often missing in one or both domains. Note that even
+with these options, some complete archaeal genomes result in very low values of
+completeness (e.g., Nanoarchaeum equitans returns 88.5%).
+
+Requires HMMer 3.0+ (http://hmmer.janelia.org/software).
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-i", "--in FILE",
+      "Path to the FastA file containing all the proteins in a genome."
+      ){ |v| o[:in] = v }
+   opts.separator ""
+   opts.separator "Report Options"
+   opts.on("-o", "--out FILE",
+      "Path to the output FastA file with the translated essential genes.",
+      "By default the file is not produced."){ |v| o[:out] = v }
+   opts.on("-m", "--per-model STR",
+      "Prefix of translated genes in independent files with the name of the",
+      "model appended. By default files are not produced."
+      ){ |v| o[:permodel] = v }
+   opts.on("-R", "--report FILE",
+      "Path to the report file. By default, the report is sent to the STDOUT."
+      ){ |v| o[:report] = v }
+   opts.on("-B", "--bacteria",
+      "If set, ignores models typically missing in Bacteria."
+      ){ |v| o[:bacteria] = v }
+   opts.on("-A", "--archaea",
+      "If set, ignores models typically missing in Archaea."
+      ){ |v| o[:archaea] = v }
+   opts.on("-G", "--genome-eq",
+      "If set, ignores models not suitable for genome-equivalents estimations.",
+      "See Rodriguez-R et al, 2015, ISME J 9(9):1928-1940."
+      ){ |v| o[:genomeeq] = v }
+   opts.on("-r", "--rename STR",
+      "If set, renames the sequences with the string provided and appends it",
+      "with pipe and the gene name (except in --per-model files)."
+      ){ |v| o[:rename]=v }
+   opts.on("-n", "--no-stats",
+      "If set, no statistics are reported on genome evaluation."
+      ){ |v| o[:stats] = v }
+   opts.on("-s", "--no-genes",
+      "If set, statistics won't include the lists of missing/multi-copy genes."
+      ){ |v| o[:genes] = v }
+   opts.on("-M", "--metagenome",
+      "If set, it allows for multiple copies of each gene and turns on",
+      "metagenomic report mode."){ |v| o[:metagenome] = v }
+   opts.separator ""
+   opts.separator "Other Options"
+   opts.on("-L", "--list-models",
+      "If set, it only lists the models and exits. Compatible with -A, -B, -G,",
+      "and -q; ignores all other parameters."){ |v| o[:list] = v }
+   opts.on("-b", "--bin DIR",
+      "Path to the directory containing the binaries of HMMer 3.0+."
+      ){ |v| o[:bin] = v }
+   opts.on("--model-file",
+      "External file containing models to search."){ |v| o[:model_file] = v }
+   opts.on("-t", "--threads INT",
+      "Number of parallel threads to be used.  By default: #{o[:thr]}."
+      ){ |v| o[:thr] = v.to_i }
+   opts.on("-q", "--quiet", "Run quietly (no STDERR output)."){ o[:q] = true }
+   opts.on("-h", "--help", "Display this screen.") do
+      puts opts
+      exit
+   end
+   opts.separator ""
+end.parse!
+abort "-i is mandatory" if o[:in].nil? and not o[:list]
+o[:bin] = o[:bin]+"/" if o[:bin].size > 0
+o[:rename] = nil if o[:metagenome]
+
+not_in_archaea = %w{GrpE Methyltransf_5 TIGR00001 TIGR00002 TIGR00009 TIGR00019
+TIGR00029 TIGR00043 TIGR00059 TIGR00060 TIGR00061 TIGR00062 TIGR00082 TIGR00086
+TIGR00092 TIGR00115 TIGR00116 TIGR00152 TIGR00158 TIGR00165 TIGR00166 TIGR00168
+TIGR00362 TIGR00388 TIGR00396 TIGR00409 TIGR00418 TIGR00420 TIGR00422 TIGR00436
+TIGR00459 TIGR00460 TIGR00472 TIGR00487 TIGR00496 TIGR00575 TIGR00631 TIGR00663
+TIGR00775 TIGR00810 TIGR00855 TIGR00922 TIGR00952 TIGR00959 TIGR00963 TIGR00964
+TIGR00967 TIGR00981 TIGR01009 TIGR01011 TIGR01017 TIGR01021 TIGR01024 TIGR01029
+TIGR01030 TIGR01031 TIGR01032 TIGR01044 TIGR01049 TIGR01050 TIGR01059 TIGR01063
+TIGR01066 TIGR01067 TIGR01071 TIGR01079 TIGR01164 TIGR01169 TIGR01171 TIGR01391
+TIGR01393 TIGR01632 TIGR01953 TIGR02012 TIGR02013 TIGR02027 TIGR02191 TIGR02350
+TIGR02386 TIGR02387 TIGR02397 TIGR02432 TIGR02729 TIGR03263 TIGR03594}
+not_in_bacteria = %w{TIGR00389 TIGR00408 TIGR00471 TIGR00775 TIGR02387}
+not_as_genomeeq = %w{TIGR02386 TIGR02387 TIGR00471 TIGR00472 TIGR00408 TIGR00409
+TIGR00389 TIGR00436 tRNA-synth_1d}
+
+begin
+   Dir.mktmpdir do |dir|
+      $stderr.puts "Temporal directory: #{dir}." unless o[:q]
+
+      # Create database.
+      $stderr.puts "Searching models." unless o[:q]
+      models = {}
+      model_id = nil
+      dbh = File.open("#{dir}/essential.hmm", "w")
+      o[:model_file] ||= File.expand_path("../lib/data/essential.hmm.gz",
+					     __FILE__)
+      mfh = (File.extname(o[:model_file])==".gz") ?
+	 Zlib::GzipReader.open(o[:model_file]) :
+	 File.open(o[:model_file],"r")
+      while ln = mfh.gets
+	 dbh.print ln
+	 ln.chomp!
+	 model_id = $1 if ln =~ /^NAME\s+(.+)/
+	 models[model_id] = $1 if ln =~ /^DESC\s+(.+)/
+      end
+      dbh.close
+      mfh.close
+      models.delete_if { |m| not_in_archaea.include? m  } if o[:archaea]
+      models.delete_if { |m| not_in_bacteria.include? m } if o[:bacteria]
+      models.delete_if { |m| not_as_genomeeq.include? m } if o[:genomeeq]
+      if o[:list]
+	 models.each_pair{ |id,desc| puts [id,desc].join("\t") }
+	 exit
+      end
+      
+      # Check HMMer version and run HMMsearch.
+      if `"#{o[:bin]}hmmsearch" -h`.lines[1] !~ /HMMER 3/
+	 raise "You have provided an unsupported version of HMMER. " +
+	    "This script requires HMMER 3.0+."
+      end
+      `"#{o[:bin]}hmmsearch" --cpu #{o[:thr]} --tblout "#{dir}/hmmsearch" \
+	 --cut_tc --notextw "#{dir}/essential.hmm" "#{o[:in]}" \
+	 > #{dir}/hmmsearch.log`
+      
+      # Parse output
+      $stderr.puts "Parsing results." unless o[:q]
+      resh = File.open("#{dir}/hmmsearch","r")
+      trash = []
+      genes = {}
+      while ln = resh.gets
+	 next if ln =~ /^#/
+	 r = ln.split /\s+/
+	 next unless models.include? r[2]
+	 if o[:metagenome]
+	    genes[ r[2] ] = [] if genes[ r[2] ].nil?
+	    genes[ r[2] ] << r[0]
+	 elsif genes[ r[2] ].nil?
+	    genes[ r[2] ] = r[0]
+	 else
+	    trash << r[2]
+	 end
+      end
+
+      # Report statistics
+      if o[:stats]
+	 reph = o[:report].nil? ? $stdout : File.open(o[:report], "w")
+	 if o[:metagenome]
+	    reph.printf "! Essential genes found: %d/%d.\n",
+	       genes.size, models.size
+	    gc = [0]*(models.size - genes.size) +
+	       genes.values.map{|g| g.length}.sort
+	    reph.printf "! Mean number of copies per model: %.3f.\n",
+	       gc.inject(:+).to_f/models.size
+	    reph.printf "! Median number of copies per model: %.1f.\n",
+	       gc.size.even? ? gc[gc.size/2,2].inject(:+).to_f/2 : gc[gc.size/2]
+	    if o[:genes] and genes.size != models.size
+	       reph.printf "! Missing genes: %s\n",
+		  ([""] +
+		     models.keys.select{|m| not genes.keys.include? m
+		     }.map{|m| "#{m}: #{models[m]}."}).join("\n!   ")
+	    end
+	 else
+	    reph.printf "! Essential genes found: %d/%d.\n",
+	       genes.size, models.size
+	    reph.printf "! Completeness: %.1f%%.\n",
+	       100.0*genes.size/models.size
+	    reph.printf "! Contamination: %.1f%%.\n",
+	       100.0*trash.size/models.size
+	    if o[:genes]
+	       reph.printf "! Multiple copies: %s\n",
+		  ([""] +
+		     trash.uniq.map{|m|
+		     "#{trash.count(m)+1} #{m}: #{models[m]}."}
+		  ).join("\n!   ") unless trash.empty?
+	       reph.printf "! Missing genes: %s\n",
+		  ([""] +
+		     models.keys.select{|m| not genes.keys.include? m
+		     }.map{|m| "#{m}: #{models[m]}."}
+		  ).join("\n!   ") unless genes.size==models.size
+	    end
+	 end
+	 reph.close unless o[:report].nil?
+      end
+      
+      # Extract sequences
+      unless o[:out].nil? and o[:permodel].nil?
+	 $stderr.puts "Extracting sequences." unless o[:q]
+	 faah = File.open(o[:in], "r")
+	 outh = o[:out].nil? ? nil : File.open(o[:out], "w")
+	 geneh = nil
+	 in_gene = nil
+	 unless o[:permodel].nil?
+	    genes.keys.each do |m|
+	       File.open("#{o[:permodel]}#{m}.faa", "w").close
+	    end
+	 end
+	 while ln = faah.gets
+	    if ln =~ /^>(\S+)/
+	       if o[:metagenome]
+		  in_gene = genes.keys.map{|k| genes[k].include?($1) ? k : nil
+		     }.compact.first
+		  in_gene = [in_gene, $1] unless in_gene.nil?
+	       else
+		  in_gene = genes.rassoc($1)
+	       end
+	       next if in_gene.nil?
+	       geneh.close unless geneh.nil?
+	       geneh = File.open("#{o[:permodel]}#{in_gene[0]}.faa",
+		  "a+") unless o[:permodel].nil?
+	       outh.print(o[:rename].nil? ?
+		  ln : ">#{o[:rename]}|#{in_gene[0]}\n") unless outh.nil?
+	       geneh.print(
+		  o[:rename].nil? ? ln : ">#{o[:rename]}\n") unless geneh.nil?
+	    else
+	       next if in_gene.nil?
+	       outh.print ln unless outh.nil?
+	       geneh.print ln unless geneh.nil?
+	    end
+	 end
+	 geneh.close unless geneh.nil?
+	 outh.close unless outh.nil?
+	 faah.close
+      end
+
+      $stderr.puts "Done." unless o[:q]
+   end # |dir|
+rescue => err
+   $stderr.puts "Exception: #{err}\n\n"
+   err.backtrace.each { |l| $stderr.puts l + "\n" }
+   err
+end