miga-base 0.7.26.0 → 0.7.26.1

data/utils/enveomics/Scripts/ogs.mcl.rb

@@ -0,0 +1,186 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @update: Sep-11-2015
+# @license: artistic license 2.0
+#
+
+$:.push File.expand_path(File.dirname(__FILE__) + "/lib")
+require 'enveomics_rb/og'
+require 'optparse'
+require 'tmpdir'
+
+o = {q:false, f:"(\\S+)-(\\S+)\\.rbm", mcl:"", inflation:1.5, blind:false,
+   evalue:false, thr:2, identity:false, bestmatch:false}
+ARGV << "-h" if ARGV.size==0
+OptionParser.new do |opts|
+   opts.banner = "
+Identifies Orthology Groups (OGs) in Reciprocal Best Matches (RBM)
+between all pairs in a collection of genomes, using the Markov Cluster
+Algorithm.
+
+Requires MCL (see http://www.micans.org/mcl).
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-o", "--out FILE",
+      "Output file containing the detected OGs."){ |v| o[:out]=v }
+   opts.on("-d", "--dir DIR",
+      "Directory containing the RBM files.",
+      "Becomes optional iff --abc is set to a non-empty file."){ |v| o[:dir]=v }
+   opts.separator ""
+   opts.separator "Other Options"
+   opts.on("-f", "--format STRING",
+      "Format of the filenames for the RBM files (within -d), using regex " +
+      "syntax.", "By default: '#{o[:f]}'."){ |v| o[:f]=v }
+   opts.on("-I", "--inflation FLOAT",
+      "Inflation parameter for MCL clustering. By default: #{o[:inflation]}."
+      ){ |v| o[:inflation]=v.to_f }
+   opts.on("-b", "--blind",
+      "If set, computes clusters without taking bitscore into account."
+      ){ |v| o[:blind]=v }
+   opts.on("-e", "--evalue",
+      "If set, uses the e-value to weight edges, instead of the default " +
+      "Bit-Score."){ |v| o[:evalue]=v }
+   opts.on("-i", "--identity",
+      "If set, uses the identity to weight edges, instead of the default " +
+      "Bit-Score."){ |v| o[:identity]=v }
+   opts.on("-B", "--best-match",
+      "If set, it assumes best-matches instead reciprocal best matches."
+      ){ |v| o[:bestmatch]=v }
+   opts.on("-m", "--mcl-bin DIR",
+      "Path to the directory containing the mcl binaries.",
+      "By default, assumed to be in the PATH."){ |v| o[:mcl]=v+"/" }
+   opts.on("--abc FILE",
+      "Use this abc file instead of a temporal file."){ |v| o[:abc] = v }
+   opts.on("-t", "--threads INT",
+      "Number of threads to use. 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 "-o is mandatory" if o[:out].nil?
+o[:evalue] = false if o[:identity]
+o[:evalue] = false if o[:blind]
+o[:identity] = false if o[:blind]
+
+##### MAIN:
+begin
+   Dir.mktmpdir do |dir|
+      o[:abc] = "#{dir}/rbms.abc" if o[:abc].nil?
+      abort "-d must exist and be a directory" unless
+	 File.size?(o[:abc]) or
+	 (!o[:dir].nil? and File.exists?(o[:dir]) and File.directory?(o[:dir]))
+      # Traverse the whole directory
+      if File.size? o[:abc]
+	 $stderr.puts "Reusing existing abc file '#{o[:abc]}'." unless o[:q]
+      else
+	 file_i = 0
+	 ln_i = 0
+	 $stderr.puts "Reading RBM files within '#{o[:dir]}'." unless o[:q]
+	 abc = File.open(o[:abc] + ".tmp", "w")
+	 Dir.entries(o[:dir]).each do |rbm_file|
+	    next unless File.file?(o[:dir]+"/"+rbm_file)
+	    # Parse the filename to identify the genomes
+	    m = /#{o[:f]}/.match(rbm_file)
+	    if m.nil? or m[2].nil?
+	       warn "Ignoring #{rbm_file}: doesn't match /#{o[:f]}/."
+	       next
+	    end
+	    file_i += 1
+	    # Read the RBMs list
+	    f = File.open(o[:dir]+"/"+rbm_file, "r")
+	    while ln = f.gets
+	       # Add the RBM to the abc file
+	       row = ln.split(/\t/)
+	       abc.puts [m[1]+">"+row[0], m[2]+">"+row[1],
+		  (o[:blind] ? "1" :
+		  (o[:evalue] ? row[10] :
+		  (o[:identity] ? row[2] : row[11])))].join("\t")
+	       ln_i += 1
+	    end
+	    f.close
+	    $stderr.print " Scanned files: #{file_i}. " +
+	       "Found RBMs: #{ln_i}.   \r" unless o[:q]
+	 end
+	 abc.close
+	 File.rename(o[:abc] + ".tmp", o[:abc])
+	 $stderr.print "\n" unless o[:q]
+      end # if File.size? o[:abc] ... else
+
+      # Build .mci file (mcxload) and compute .mccl file (mcl)
+      $stderr.puts "Markov-Clustering" unless o[:q]
+      `'#{o[:mcl]}mcxload' #{"--stream-mirror" unless o[:bestmatch]} \
+	 -abc '#{o[:abc]}' -o '#{dir}/rbms.mci' --write-binary \
+	 -write-tab '#{dir}/genes.tab' #{"--stream-neg-log10" if o[:evalue]} \
+	 &>/dev/null`
+      `'#{o[:mcl]}mcl' '#{dir}/rbms.mci' -V all -I #{o[:inflation].to_s} \
+	 -o '#{dir}/ogs.mcl' -te #{o[:thr].to_s}`
+
+      # Load .tab as Gene objects
+      $stderr.puts "Loading gene table from '#{dir}/genes.tab'." unless o[:q]
+      genes = []
+      tab = File.open("#{dir}/genes.tab", "r")
+      while ln = tab.gets
+	 ln.chomp!
+	 r = ln.split /\t|>/
+	 genes[ r[0].to_i ] = Gene.new(r[1], r[2])
+      end
+      tab.close
+      $stderr.puts " Got " + genes.size.to_s + " genes in " +
+	 Gene.genomes.size.to_s + " genomes." unless o[:q]
+      
+      # Load .mcl file as OGCollection
+      $stderr.puts "Loading clusters from '#{dir}/ogs.mcl'." unless o[:q]
+      collection = OGCollection.new
+      mcl = File.open("#{dir}/ogs.mcl", "r")
+      in_matrix = false
+      my_genes = nil
+      while ln = mcl.gets
+         ln.chomp!
+	 if ln =~ /^\(mclmatrix$/
+	    in_matrix = true
+	    next
+	 end
+	 next if ln =~ /^begin$/
+	 if in_matrix
+	    break if ln =~ /^\)$/
+	    if ln =~ /^\d+\s+/
+	       ln.sub!(/^\d+\s+/, "")
+	       my_genes = []
+	    end
+	    ln.sub!(/^\s+/, "")
+	    raise "Incomplete mcl matrix, offending line: #{$.}: #{ln}" if
+	       my_genes.nil?
+	    my_genes += ln.split(/\s/)
+	    if my_genes.last == "$"
+	       my_genes.pop
+	       og = OG.new
+	       my_genes.each{|i| og << genes[ i.to_i ]}
+	       collection << og
+	       my_genes = nil
+	    end
+	 end
+      end
+      mcl.close
+      $stderr.puts " Got #{collection.ogs.size} clusters." unless o[:q]
+      
+      # Save the output matrix
+      $stderr.puts "Saving matrix into '#{o[:out]}'." unless o[:q]
+      f = File.open(o[:out], "w")
+      f.puts collection.to_s
+      f.close
+      $stderr.puts "Done.\n" unless o[:q] 
+   end
+rescue => err
+   $stderr.puts "Exception: #{err}\n\n"
+   err.backtrace.each { |l| $stderr.puts l + "\n" }
+   err
+end
+
+

data/utils/enveomics/Scripts/ogs.rb

@@ -0,0 +1,104 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @update: Apr-29-2015
+# @license: artistic license 2.0
+#
+
+$:.push File.expand_path(File.dirname(__FILE__) + '/lib')
+require 'enveomics_rb/og'
+require 'optparse'
+
+o = {:q=>FALSE, :f=>"(\\S+)-(\\S+)\\.rbm", :consolidate=>TRUE, :pre=>[]}
+ARGV << '-h' if ARGV.size==0
+OptionParser.new do |opts|
+   opts.banner = "
+***IMPORTANT NOTE***
+This script suffers from chaining effect and is very sensitive to spurious connections,
+because it applies a greedy clustering algorithm. For most practical purposes, the use
+of this script is discouraged and `ogs.mcl.rb` should be preferred. [ Apr-29-2015 ]
+
+Identifies Orthology Groups (OGs) in Reciprocal Best Matches (RBM)
+between all pairs in a collection of genomes.
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-o", "--out FILE", "Output file containing the detected OGs."){ |v| o[:out]=v }
+   opts.separator ""
+   opts.separator "Other Options"
+   opts.on("-d", "--dir DIR", "Directory containing the RBM files."){ |v| o[:dir]=v }
+   opts.on("-p", "--pre-ogs FILE1,FILE2,...", Array, "Pre-computed OGs file(s), separated by commas."){ |v| o[:pre]=v }
+   opts.on("-n", "--unchecked", "Do not check internal redundancy in OGs."){ o[:consolidate]=FALSE }
+   opts.on("-f","--format STRING", "Format of the filenames for the RBM files (within -d), using regex syntax. By default: '#{o[:f]}'."){ |v| o[:f]=v }
+   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 "-o is mandatory" if o[:out].nil?
+
+##### MAIN:
+begin
+   # Initialize the collection of OGs.
+   collection = OGCollection.new
+   # Read the pre-computed OGs (if -p is passed).
+   o[:pre].each do |pre|
+      $stderr.puts "Reading pre-computed OGs in '#{pre}'." unless o[:q]
+      f = File.open(pre, 'r')
+      h = f.gets.chomp.split /\t/
+      while ln = f.gets
+	 collection << OG.new(h, ln.chomp.split(/\t/))
+      end
+      f.close
+      $stderr.puts " Loaded OGs: #{collection.ogs.length}." unless o[:q]
+   end
+   # Read the RBM files in the directory (if -d is passed).
+   unless o[:dir].nil?
+      abort "-d must exist and be a directory" unless File.exists?(o[:dir]) and File.directory?(o[:dir])
+      # Traverse the whole directory.
+      file_i = 0
+      $stderr.puts "Reading RBM files within '#{o[:dir]}'." unless o[:q] 
+      Dir.entries(o[:dir]).each do |rbm_file|
+	 next unless File.file?(o[:dir]+"/"+rbm_file)
+	 # Parse the filename to identify the genomes.
+	 m = /#{o[:f]}/.match(rbm_file)
+	 if m.nil? or m[2].nil?
+	    warn "Cannot parse filename: #{rbm_file} (doesn't match /#{o[:f]}/)."
+	    next
+	 end
+	 file_i += 1
+	 # Read the RBMs list
+	 f = File.open(o[:dir]+"/"+rbm_file, "r")
+	 while ln = f.gets
+	    # Add the RBM to the collection of OGs. Only the first two columns are used.
+	    row = ln.split(/\t/)
+	    collection.add_rbm( Gene.new(m[1],row[0]), Gene.new(m[2],row[1]) )
+	 end
+	 f.close
+	 $stderr.print " Scanned files: #{file_i}. Found OGs: #{collection.ogs.length}.   \r" unless o[:q]
+      end
+      $stderr.print "\n" unless o[:q]
+   end
+   # Evaluate internal consistency merging linked OGs (unless -n is passed).
+   if o[:consolidate]
+      $stderr.puts "Evaluating internal consistency." unless o[:q] 
+      collection.consolidate!
+      $stderr.puts " Final OGs: #{collection.ogs.length}." unless o[:q]
+   end
+   # Save the output matrix
+   $stderr.puts "Saving matrix into '#{o[:out]}'." unless o[:q]
+   f = File.open(o[:out], "w")
+   f.puts collection.to_s
+   f.close
+   $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/ogs.stats.rb

@@ -0,0 +1,131 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @license: Artistic-2.0
+#
+
+$:.push File.expand_path(File.dirname(__FILE__) + '/lib')
+require 'enveomics_rb/og'
+require 'optparse'
+require 'json'
+
+o = {q:false, a:false}
+ARGV << '-h' if ARGV.size==0
+OptionParser.new do |opts|
+  opts.banner = "
+Estimates some descriptive statistics on a set of Orthology Groups (OGs).
+
+Usage: #{$0} [options]"
+  opts.separator ""
+  opts.separator "Mandatory"
+  opts.on("-o", "--ogs FILE",
+    "Input file containing the precomputed OGs."){ |v| o[:ogs]=v }
+  opts.separator ""
+  opts.separator "Other Options"
+  opts.on("-j", "--json FILE", "Output file in JSON format."){ |v| o[:json]=v }
+  opts.on("-t", "--tab FILE","Output file in tabular format."){ |v| o[:tab]=v }
+  opts.on("-T", "--transposed-tab FILE",
+    "Output file in transposed tabular format."){ |v| o[:ttab]=v }
+  opts.on("-a", "--auto", "Run completely quietly (no STDERR or STDOUT)") do
+    o[:q] = true
+    o[:a] = true
+  end
+  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 "-o is mandatory" if o[:ogs].nil?
+
+##### MAIN:
+begin
+  # Initialize the collection of OGs.
+  collection = OGCollection.new
+   
+  # Read the pre-computed OGs
+  $stderr.puts "Reading pre-computed OGs in '#{o[:ogs]}'." unless o[:q]
+  f = File.open(o[:ogs], "r")
+  h = f.gets.chomp.split /\t/
+  while ln = f.gets
+    collection << OG.new(h, ln.chomp.split(/\t/))
+  end
+  f.close
+  $stderr.puts " Loaded OGs: #{collection.ogs.length}." unless o[:q]
+   
+  # Estimate descriptive stats
+  stat_name = {
+    genomes: "Number of genomes",
+    pan: "Pangenome (OGs)",
+    core: "Core genome (OGs)",
+    core90pc: "OGs in 90% of the genomes",
+    core80pc: "OGs in 80% of the genomes",
+    unus: "Unus genome, core genome discarding paralogs (OGs)",
+    avg: "Average number of OGs in a genome",
+    avg_pan: "Average genome (OGs) / Pangenome (OGs)",
+    core_avg: "Core genome (OGs) / Average genome (OGs)",
+    core_pan: "Core genome (OGs) / Pangenome (OGs)",
+    ogs_shannon: "Entropy of the OG frequencies (bits)"
+  }
+  stats = {}
+  stats[:genomes] = Gene.genomes.length
+  stats[:pan] = collection.ogs.length
+  stats[:core] = collection.ogs.map do |og|
+    (og.genomes.length == Gene.genomes.length) ? 1 : 0
+  end.inject(0,:+)
+  stats[:core90pc] = collection.ogs.map do |og|
+    (og.genomes.length >= 0.9*Gene.genomes.length) ? 1 : 0
+  end.inject(0,:+)
+  stats[:core80pc] = collection.ogs.map do |og|
+    (og.genomes.length >= 0.8*Gene.genomes.length) ? 1 : 0
+  end.inject(0,:+)
+  stats[:unus] = collection.ogs.map do |og|
+    (og.genomes.length != Gene.genomes.length) ? 0 :
+      (og.genes.all?{ |i| i.size==1 }) ? 1 : 0
+  end.inject(0,:+)
+  og_genomes = collection.ogs.map{ |og| og.genomes.length }.inject(0,:+)
+  stats[:avg] = og_genomes.to_f/Gene.genomes.length
+  stats[:avg_pan] = stats[:avg]/stats[:pan]
+  stats[:core_avg] = stats[:core].to_f/stats[:avg]
+  stats[:core_pan] = stats[:core].to_f/stats[:pan]
+  stats[:ogs_shannon] = -1 * collection.ogs.map do |og|
+    pi = og.genomes.length.to_f/Gene.genomes.length
+    pi * Math.log(pi)
+  end.inject(0.0,:+)
+
+  # Show result
+  $stderr.puts "Generating reports." unless o[:q]
+  stats.each_pair{ |k,v| puts " #{stat_name[k]}: #{v}" } unless o[:a]
+
+  # Save results in JSON
+  unless o[:json].nil?
+    ohf = File.open(o[:json], "w")
+    ohf.puts JSON.pretty_generate(stats)
+    ohf.close
+  end
+
+  # Save results in tab
+  unless o[:tab].nil?
+    ohf = File.open(o[:tab], "w")
+    stats.each_pair{ |k,v| ohf.puts "#{k}\t#{v}" }
+    ohf.close
+  end
+
+  # Save results in T(tab)
+  unless o[:ttab].nil?
+    ohf = File.open(o[:ttab], "w")
+    ohf.puts stats.keys.join("\t")
+    ohf.puts stats.values.join("\t")
+    ohf.close
+  end
+
+  $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/rbm.rb

@@ -0,0 +1,146 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @update: Aug-25-2015
+# @license: artistic license 2.0
+#
+
+require 'optparse'
+require 'tmpdir'
+
+o = {len:0, id:0, fract:0, score:0, q:false, bin:"", program:"blast+", thr:1,
+   nucl:false}
+ARGV << "-h" if ARGV.size==0
+OptionParser.new do |opts|
+   opts.banner = "
+Finds the reciprocal best matches between two sets of sequences.
+
+Usage: #{$0} [options]"
+   opts.separator ""
+   opts.separator "Mandatory"
+   opts.on("-1", "--seq1 FILE",
+      "Path to the FastA file containing the set 1."){ |v| o[:seq1] = v }
+   opts.on("-2", "--seq2 FILE",
+      "Path to the FastA file containing the set 2."){ |v| o[:seq2] = v }
+   opts.separator ""
+   opts.separator "Search Options"
+   opts.on("-n", "--nucl",
+      "Sequences are assumed to be nucleotides (proteins by default)."
+      ){ |v| o[:nucl] = true }
+   opts.on("-l", "--len INT",
+      "Minimum alignment length (in residues).  By default: #{o[:len]}."
+      ){ |v| o[:len] = v.to_i }
+   opts.on("-f", "--fract FLOAT",
+      "Minimum alignment length (as a fraction of the query).",
+      "If set, requires BLAST+ or Diamond (see -p).  By default: #{o[:fract]}."
+      ){ |v| o[:fract] = v.to_i }
+   opts.on("-i", "--id NUM",
+      "Minimum alignment identity (in %).  By default: #{o[:id].to_s}."
+      ){ |v| o[:id] = v.to_f }
+   opts.on("-s", "--score NUM",
+      "Minimum alignment score (in bits).  By default: #{o[:score]}."
+      ){ |v| o[:score] = v.to_f }
+   opts.separator ""
+   opts.separator "Software Options"
+   opts.on("-b", "--bin DIR",
+      "Path to the directory containing the binaries of the search program."
+      ){ |v| o[:bin] = v }
+   opts.on("-p", "--program STR",
+      "Search program to be used.  One of: blast+ (default), blast, diamond."
+      ){ |v| o[:program] = v }
+   opts.on("-t", "--threads INT",
+      "Number of parallel threads to be used.  By default: #{o[:thr]}."
+      ){ |v| o[:thr] = v.to_i }
+   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 "-1 is mandatory" if o[:seq1].nil?
+abort "-2 is mandatory" if o[:seq2].nil?
+abort '-p diamond is incompatible with -n' if o[:program]=='diamond' && o[:nucl]
+abort 'Argument -f/--fract requires -p blast+ or -p diamond' if
+   o[:fract]>0 and o[:program]!='blast+' and o[:program]!='diamond'
+o[:bin] = o[:bin]+"/" if o[:bin].size > 0
+
+Dir.mktmpdir do |dir|
+   $stderr.puts "Temporal directory: #{dir}." unless o[:q]
+
+   # Create databases.
+   $stderr.puts "Creating databases." unless o[:q]
+   [:seq1, :seq2].each do |seq|
+      case o[:program].downcase
+      when 'blast'
+         `"#{o[:bin]}formatdb" -i "#{o[seq]}" -n "#{dir}/#{seq}" \
+	 -p #{(o[:nucl]?"F":"T")}`
+      when 'blast+'
+         `"#{o[:bin]}makeblastdb" -in "#{o[seq]}" -out "#{dir}/#{seq}" \
+	 -dbtype #{(o[:nucl]?"nucl":"prot")}`
+      when 'diamond'
+         `"#{o[:bin]}diamond" makedb --in "#{dir}/#{seq}.fa" \
+         --db "#{dir}/#{seq}.fa.dmnd" --threads "#{o[:thr]}"`
+      else
+         abort "Unsupported program: #{o[:program]}."
+      end
+   end # |seq|
+
+   # Best-hits.
+   rbh = {}
+   n2 = 0
+   $stderr.puts " Running comparisons." unless o[:q]
+   [2,1].each do |i|
+      qry_seen = {}
+      q = o[:"seq#{i}"]
+      s = "#{dir}/seq#{i==1?2:1}"
+      $stderr.puts "  Query: #{q}." unless o[:q]
+      case o[:program].downcase
+      when 'blast'
+	 `"#{o[:bin]}blastall" -p #{o[:nucl]?"blastn":"blastp"} -d "#{s}" \
+	 -i "#{q}" -v 1 -b 1 -a #{o[:thr]} -m 8 -o "#{dir}/#{i}.tab"`
+      when 'blast+'
+	 `"#{o[:bin]}#{o[:nucl]?"blastn":"blastp"}" -db "#{s}" -query "#{q}" \
+	 -max_target_seqs 1 -num_threads #{o[:thr]} -out "#{dir}/#{i}.tab" \
+	 -outfmt "6 qseqid sseqid pident length mismatch gapopen qstart qend \
+	 sstart send evalue bitscore qlen slen"`
+      when 'diamond'
+         `"#{o[:bin]}diamond" blastp --threads "#{o[:thr]}" \
+         --outfmt "6 qseqid sseqid pident length mismatch gapopen qstart qend \
+         sstart send evalue bitscore qlen slen" --db "#{s}.dmnd" \
+         --query "#{q}" --out "#{dir}/#{i}.tab" --more-sensitive`
+      else
+	 abort "Unsupported program: #{o[:program]}."
+      end
+      fh = File.open("#{dir}/#{i}.tab", "r")
+      n = 0
+      fh.each_line do |ln|
+	 ln.chomp!
+	 row = ln.split(/\t/)
+	 row[12] = "1" unless %w[blast+ diamond].include? o[:program]
+	 if qry_seen[ row[0] ].nil? and row[3].to_i >= o[:len] and
+	       row[2].to_f >= o[:id] and row[11].to_f >= o[:score] and
+	       row[3].to_f/row[12].to_i >= o[:fract]
+	    qry_seen[ row[0] ] = 1
+	    n += 1
+	    if i==2
+	       rbh[ row[0] ] = row[1]
+	    else
+	       if !rbh[ row[1] ].nil? and rbh[ row[1] ]==row[0]
+		  puts ln
+		  n2 += 1
+	       end
+	    end
+	 end
+      end # |ln|
+      fh.close()
+      $stderr.puts "    #{n} sequences with hit." unless o[:q]
+   end # |i|
+   $stderr.puts "  #{n2} RBMs." unless o[:q]
+end # |dir|
+
+
+