bio-rocker 1.0.0 → 1.1.9

data/lib/rocker.rb

@@ -2,7 +2,7 @@
 # @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
 # @author Luis (Coto) Orellana
 # @license artistic license 2.0
-# @update Jun-05-2015
+# @update Sep-07-2015
 #
 
 require 'rocker/blasthit'
@@ -10,20 +10,30 @@ require 'rocker/rocdata'
 
 class ROCker
    #================================[ Class ]
+   @@VERSION = "1.1.9"
+   @@CITATION = "Orellana, Rodriguez-R, & Konstantinidis. Under review. " +
+      "Detecting and quantifying functional genes in short-read metagenomic " +
+      "datasets: method development and application to the nitrogen cycle " +
+      "genes."
    @@DEFAULTS = {
       # General
-      :q=>false, :r=>'R', :nucl=>false, :debug=>false,:thr=>2,:search=>:blast,
+      q: false, r: "R", nucl: false, debug: false, thr: 2, search: :blast,
       # External software
-      :searchbins=>'',
-      :searchcmd=>{
-	 :blast=>'%1$s%2$s -query "%3$s" -db "%4$s" -out "%5$s" -num_threads %6$d -outfmt 6 -max_target_seqs 1',
-	 :diamond=>'%1$sdiamond %2$s -q "%3$s" -d "%4$s" -o "%5$s" -t %6$d -k 1 --min-score 20 --sensitive'},
-      :makedbcmd=>{
-	 :blast=>'%1$smakeblastdb -dbtype %2$s -in "%3$s" -out "%4$s"',
-	 :diamond=>'%1$sdiamond makedb --in "%3$s" -d "%4$s"'}
+      searchbins: "",
+      searchcmd: {
+	 blast: '%1$s%2$s -query "%3$s" -db "%4$s" -out "%5$s" ' +
+	    '-num_threads %6$d -outfmt 6 -max_target_seqs 1',
+	 diamond: '%1$sdiamond %2$s -q "%3$s" -d "%4$s" -a "%5$s.daa" -p %6$d' +
+	    ' -k 1 --min-score 20 --sensitive && %1$sdiamond view -a "%5$s"' +
+	    ' -o "%5$s"'},
+      makedbcmd: {
+	 blast: '%1$smakeblastdb -dbtype %2$s -in "%3$s" -out "%4$s"',
+	 diamond: '%1$sdiamond makedb --in "%3$s" -d "%4$s"'}
    }
    def self.defaults() @@DEFAULTS ; end
    def self.default(k) @@DEFAULTS[k] ; end
+   def self.VERSION; @@VERSION ; end
+   def self.CITATION; @@CITATION ; end
 
    #================================[ Instance ]
    attr_reader :o
@@ -46,7 +56,8 @@ class ROCker
    end
    def bash(cmd, err_msg=nil)
       o = `#{cmd} 2>&1 && echo '{'`
-      raise (err_msg.nil? ? "Error executing: #{cmd}\n\n#{o}" : err_msg) unless o[-2]=='{'
+      raise (err_msg.nil? ? "Error executing: #{cmd}\n\n#{o}" : err_msg) unless
+	 o[-2]=="{"
       true
    end
 end
@@ -63,10 +74,10 @@ require 'rocker/step/plot'
 class Numeric
    def ordinalize
       n= self.to_s
-      s= n[-2]=='1' ? 'th' :
-	 n[-1]=='1' ? 'st' :
-	 n[-1]=='2' ? 'nd' :
-	 n[-1]=='3' ? 'rd' : 'th'
+      s= n[-2]=='1' ? "th" :
+	 n[-1]=='1' ? "st" :
+	 n[-1]=='2' ? "nd" :
+	 n[-1]=='3' ? "rd" : "th"
       n + s
    end
 end

data/lib/rocker/blasthit.rb

@@ -2,12 +2,13 @@
 # @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
 # @author Luis (Coto) Orellana
 # @license artistic license 2.0
-# @update Jan-22-2015
+# @update Sep-06-2015
 #
 
 class BlastHit
-   attr_reader :sbj, :sfrom, :sto, :bits, :istrue, :midpoint
-   # Initialize from BLAST using new(ln,aln), initialize from TABLE using new(ln)
+   attr_reader :sbj, :sfrom, :sto, :bits, :istrue, :isfalse, :midpoint
+   # Initialize from BLAST using new(ln,aln),
+   # initialize from TABLE using new(ln)
    def initialize(ln, aln=nil)
       l = ln.chomp.split(/\t/)
       if aln.nil?
@@ -16,6 +17,7 @@ class BlastHit
 	 @sto	= l[2].to_i
 	 @bits	= l[3].to_f
 	 @istrue = l[4]=='1'
+	 @istrue = l[4]=='-1'
 	 @midpoint = l[5].to_i
       else
 	 s = aln.seq(l[1])
@@ -27,13 +29,14 @@ class BlastHit
 	 @sto	= [a,b].max
 	 @bits	= l[11].to_f
 	 @istrue = ! /@%/.match(l[0]).nil?
+	 @isfalse = ! /@\$/.match(l[0]).nil?
 	 @midpoint = s.pos2col(((l[8].to_f+l[9].to_f)/2).ceil)
       end
    end
    def to_s
       self.sbj.nil? ? "" :
-	 [self.sbj, self.sfrom.to_s, self.sto.to_s, self.bits.to_s,
-	    self.istrue ? '1' : '0', self.midpoint].join("\t") + "\n"
+	 [sbj, sfrom.to_s, sto.to_s, bits.to_s,
+	    istrue ? "1" : (isfalse ? "-1" : "0"), midpoint].join("\t") + "\n"
    end
 end
 

data/lib/rocker/genome-set.rb

@@ -0,0 +1,70 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-23-2015
+#
+
+class GenomeSet
+   attr_reader :rocker, :ids, :taxa
+   def initialize(rocker, ids)
+      @rocker = rocker
+      @ids = ids
+      @ids = [] if ids.nil?
+      @taxa = {}
+      @all_taxa = {}
+   end
+   def download(file)
+      tmp_ids = Array.new(self.ids)
+      ofh = File.open(file, "w")
+      while tmp_ids.size>0
+	 ofh.print rocker.ebiFetch(:embl, tmp_ids.shift(200), :fasta)
+      end
+      ofh.close
+   end
+   def link_taxon(id, taxon)
+      @all_taxa[ taxon.to_sym ] ||= []
+      @all_taxa[ taxon.to_sym ] << id
+   end
+   def choose_genomes!(rank)
+      @taxa = {}
+      self.get_taxonomy! rank
+      @all_taxa.each_pair{ |taxon,ids| @taxa[taxon] = ids.sample }
+      @ids = @taxa.values
+   end
+   def get_taxonomy!(rank)
+      @all_taxa = {}
+      ids.each do |id|
+	 self.link_taxon(id, genome2taxon(id, rank))
+      end
+   end
+   def taxa=(hash)
+      @taxa = {}
+      hash.each_pair{ |taxon, id| @taxa[taxon] = id if self.ids.include? id }
+   end
+   def size() self.ids.size end
+   def empty?() self.ids.empty? end
+
+   #================================[ Utilities ]
+   def genome2taxon(genome_id, rank='species')
+      v = genome2taxid(genome_id)
+      unless v.nil?
+	 xml = rocker.ebiFetch('taxonomy', [v], 'enataxonomyxml').gsub(/\s*\n\s*/,'')
+	 v = xml.scan(/<taxon [^>]+>/).grep(/rank="#{rank}"/).first
+	 v.sub!(/.* taxId="(\d+)".*/,"\\1") unless v.nil?
+      end
+      return "no-taxon-#{(0...12).map { (65 + rand(26)).chr }.join}" if v.nil? or v !~ /^\d+$/
+      v
+   end
+   def genome2taxid(genome_id)
+      doc = rocker.ebiFetch('embl', [genome_id], 'annot').split(/[\n\r]/)
+      ln = doc.grep(/^FT\s+\/db_xref="taxon:/).first
+      ln = doc.grep(/^OX\s+NCBI_TaxID=/).first if ln.nil?
+      return nil if ln.nil?
+      ln.sub!(/.*(?:"taxon:|NCBI_TaxID=)(\d+)["; ].*/, "\\1")
+      return nil unless ln =~ /^\d+$/
+      ln
+   end
+end
+
+

data/lib/rocker/protein-set.rb

@@ -0,0 +1,90 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jul-20-2015
+#
+
+require 'rocker/alignment'
+
+class ProteinSet
+   attr_reader :rocker, :ids, :aln
+   def initialize(rocker, ids=nil, file=nil, aln_file=nil)
+      @genomes = {}
+      @tranids = {}
+      @aln = nil
+      @rocker = rocker
+      @ids = []
+      @ids += ids unless ids.nil?
+      @ids += File.readlines(file).map{ |l| l.chomp } unless file.nil?
+      unless aln_file.nil?
+	 aln = Alignment.new
+	 aln.read_fasta aln_file
+	 aln_ids = aln.get_ids
+	 @aln = aln if (@ids - aln_ids).empty?
+	 @ids += aln_ids
+      end
+      @ids.uniq!
+   end
+   def download(file)
+      tmp_ids = Array.new(self.ids)
+      f = File.open(file, "w")
+      while tmp_ids.size>0
+	 f.print rocker.ebiFetch(:uniprotkb, tmp_ids.shift(200), :fasta)
+      end
+      f.close
+   end
+   def get_from_aln(file, aln)
+      f = File.open(file, "w")
+      f.print aln.to_seq_s
+      f.close
+   end
+   def get_genomes!
+      self.ids.each do |id|
+	 doc = self.rocker.ebiFetch(:uniprotkb, [id], :annot).split("\n")
+	 doc.grep( /^DR\s+EMBL;/ ).map do |ln|
+	    r=ln.split('; ')
+	    self.link_genome(id, r[1])
+	    self.link_tranid(id, r[2])
+	 end
+      end
+   end
+   def link_genome(prot_id, genome_id)
+      @genomes[prot_id] ||= []
+      @genomes[prot_id] << genome_id
+      @genomes[prot_id].uniq!
+   end
+   def link_tranid(prot_id, transl_id)
+      @tranids[prot_id] ||= []
+      @tranids[prot_id] << transl_id
+      @tranids[prot_id].uniq!
+   end
+   def genomes
+      return [] if @genomes.empty?
+      @genomes.values.reduce(:+).uniq
+   end
+   def tranids
+      return [] if @tranids.empty?
+      @tranids.values.reduce(:+).uniq
+   end
+   def in_coords(coords)
+      coords.keys.map do |genome|
+	 locations = coords[ genome ]
+	 locations.map do |loc|
+	    if not loc[:prot_id].nil?
+	       loc[:prot_id] if self.include? loc[:prot_id]
+	    elsif not loc[:tran_id].nil? and not @tranids.rassoc(loc[:tran_id]).nil?
+	       @tranids.rassoc(loc[:tran_id]).first
+	    else
+	       warn "Warning: Impossible to resolve protein located in '#{genome}' at: #{loc}."
+	       nil
+	    end
+	 end
+      end.reduce([], :+).compact.uniq
+   end
+   def size() self.ids.size end
+   def empty?() self.ids.empty? end
+   def include?(id) self.ids.include?(id) end
+end
+
+

data/lib/rocker/rocdata.rb

@@ -2,7 +2,7 @@
 # @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
 # @author Luis (Coto) Orellana
 # @license artistic license 2.0
-# @update Jan-22-2015
+# @update Sep-07-2015
 #
 
 require 'rocker/rinterface'
@@ -11,11 +11,13 @@ require 'rocker/alignment'
 require 'tmpdir'
 
 class ROCData
-   attr_reader :aln, :windows, :r
-   # Use ROCData.new(table,aln,window) to re-compute from table, use ROCData.new(data) to load
+   attr_reader :aln, :windows, :r, :refined
+   # Use ROCData.new(table,aln,window) to re-compute from table, use
+   # ROCData.new(data) to load
    def initialize(val, aln=nil, window=nil)
       @r = RInterface.new
       @nucl = false
+      @refined = false
       if not aln.nil?
 	 @aln = aln
 	 self.rrun "library('pROC');"
@@ -33,7 +35,9 @@ class ROCData
 	 @aln.read_rocker(val)
       end
    end
-   def win_at_col(col) self.windows.select{|w| (w.from<=col) and (w.to>=col)}.first end
+   def win_at_col(col)
+      self.windows.select{|w| (w.from<=col) and (w.to>=col)}.first
+   end
    def in_nucl?() @nucl end
    def nucl=(nucl) @nucl=nucl end
    def refine! table
@@ -41,14 +45,17 @@ class ROCData
 	 return false unless self.load_table! table
 	 break if self._refine_iter(table)==0
       end
+      @refined = true
       return true
    end
+   def is_refined? ; @refined ; end
    def _refine_iter table
       to_refine = []
       self.windows.each do |w|
 	 next if w.almost_empty or w.length <= 5
 	 self.rrun "acc <- w$accuracy[w$V1==#{w.from}];"
-	 to_refine << w if self.rrun("ifelse(is.na(acc), 100, acc)", :float) < 95.0
+	 to_refine << w if
+	    self.rrun("ifelse(is.na(acc), 100, acc)", :float) < 95.0
       end
       n = to_refine.size
       return 0 unless n > 0
@@ -86,9 +93,17 @@ class ROCData
 	    win <- which( (m>=w$V1) & (m<=w$V2))[1];
 	    if(!is.na(win)){
 	       if(x$V4[i] >= w$V5[win]){
-		  if(x$V5[i]==1){ w$tp[win] <- w$tp[win]+1 }else{ w$fp[win] <- w$fp[win]+1 };
+		  if(x$V5[i]==1){
+		     w$tp[win] <- w$tp[win]+1
+		  } else {
+		     w$fp[win] <- w$fp[win]+1
+		  }
 	       }else{
-		  if(x$V5[i]==1){ w$fn[win] <- w$fn[win]+1 }else{ w$tn[win] <- w$tn[win]+1 };
+		  if(x$V5[i]==1){
+		     w$fn[win] <- w$fn[win]+1
+		  } else {
+		     w$tn[win] <- w$tn[win]+1
+		  };
 	       }
 	    }
 	 }
@@ -106,7 +121,9 @@ class ROCData
    end
    def init_windows!(size)
       @windows = []
-      1.step(self.aln.cols,size).each { |a| @windows << ROCWindow.new(self, a, a+size-1) }
+      1.step(self.aln.cols,size).each do |a|
+	 @windows << ROCWindow.new(self, a, a+size-1)
+      end
    end
    def rrun(cmd, type=nil) self.r.run cmd, type end
    def save(file)
@@ -115,7 +132,7 @@ class ROCData
       f.close
    end
    def to_s
-      o = ''
+      o = "#v ROCker " + ROCker.VERSION + "\n"
       self.windows.each{|w| o += w.to_s}
       o += self.aln.to_s
       return o

data/lib/rocker/rocwindow.rb

@@ -2,7 +2,7 @@
 # @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
 # @author Luis (Coto) Orellana
 # @license artistic license 2.0
-# @update Jan-22-2015
+# @update Sep-07-2015
 #
 
 class ROCWindow
@@ -22,16 +22,18 @@ class ROCWindow
 	 @from = [a,b].min
 	 @to = [a,b].max
 	 @thr = nil
-	 self.compute!
+	 compute!
       end
    end
    def compute!
-      self.load_hits
-      @hits = self.rrun "nrow(y);", :int
-      @tps = self.rrun "sum(y$V5);", :int
-      unless self.almost_empty
-	 self.rrun "rocobj <- roc(y$V5, y$V4);"
-	 thr = self.rrun 'coords(rocobj, "best", ret="threshold", best.method="youden", best.weights=c(0.5, sum(y$V5)/nrow(y)))[1];', :float
+      load_hits
+      @hits = rrun("nrow(y);", :int)
+      @tps = rrun("sum(y$V5==1);", :int)
+      unless almost_empty
+	 rrun "rocobj <- roc(as.numeric(y$V5==1), y$V4);"
+	 thr = rrun("coords(rocobj, 'best', ret='threshold', " +
+	    "best.method='youden', " +
+	    "best.weights=c(0.5, sum(y$V5==1)/nrow(y)))[1];", :float)
 	 @thr = thr.to_f
 	 @thr = nil if @thr==0.0 or @thr.infinite?
       end
@@ -48,16 +50,16 @@ class ROCWindow
       return nil if a.nil? and b.nil?
       return a.thr if b.nil?
       return b.thr if a.nil?
-      return (b.thr*(self.from-a.from) - a.thr*(self.from-b.from))/(b.from-a.from)
+      return (b.thr*(from-a.from) - a.thr*(from-b.from))/(b.from-a.from)
    end
-   def load_hits() self.rrun "y <- x[x$V6>=#{self.from} & x$V6<=#{self.to},];" end
-   def previous() (self.from == 1) ? nil : self.data.win_at_col(self.from - 1) end
-   def next() (self.to == self.data.aln.cols) ? nil : self.data.win_at_col(self.to + 1) end
-   def thr_notnil() (@thr.nil? or @thr.infinite?) ? self.around_thr : @thr end
-   def fps() self.hits - self.tps end
-   def almost_empty() self.fps < 3 or self.tps < 3 end
-   def length() self.to - self.from + 1 end
-   def rrun(cmd, type=nil) self.data.rrun cmd, type end
-   def to_s() [self.from, self.to, self.hits, self.tps, self.thr_notnil].join("\t") + "\n" end
+   def load_hits() self.rrun "y <- x[x$V6>=#{from} & x$V6<=#{to},];" end
+   def previous() (from == 1) ? nil : data.win_at_col(from - 1) end
+   def next() (to == data.aln.cols) ? nil : data.win_at_col(to + 1) end
+   def thr_notnil() (@thr.nil? or @thr.infinite?) ? around_thr : @thr end
+   def fps() hits - tps end
+   def almost_empty() fps < 3 or tps < 3 end
+   def length() to - from + 1 end
+   def rrun(cmd, type=nil) data.rrun(cmd, type) end
+   def to_s() [from, to, hits, tps, thr_notnil].join("\t") + "\n" end
 end
 

data/lib/rocker/step/build.rb

@@ -2,21 +2,27 @@
 # @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
 # @author Luis (Coto) Orellana
 # @license artistic license 2.0
-# @update Jun-05-2015
+# @update Sep-11-2015
 #
 
 require 'json'
+require 'rocker/protein-set'
+require 'rocker/genome-set'
 
 class ROCker
    #================================[ Class ]
-   @@EBIREST = 'http://www.ebi.ac.uk/Tools'
-   @@DEFAULTS.merge!({:positive=>[], :negative=>[], :genomefrx=>1.0, :seqdepth=>0.03, :readlen=>100, :minovl=>50,
+   @@EBIREST = "http://www.ebi.ac.uk/Tools"
+   @@DEFAULTS.merge!({positive:[], negative:[], seqdepth:0.03, readlen:100,
+      minovl:50,
       # Ext. Software
-      :aligner=>:clustalo, :simulator=>:grinder,
-      :simulatorbin=>{:grinder=>'grinder'},
-      :simulatorcmd=>{:grinder=>'%1$s -reference_file "%2$s" -cf "%3$f" -dc \'-~*NnKkMmRrYySsWwBbVvHhDdXx\' -md uniform 0.1 -mr 95 5 -rd %4$d uniform 5 -base_name "%5$s"'},
-      :alignerbin=>{:muscle=>'muscle', :clustalo=>'clustalo'},
-      :alignercmd=>{:muscle=>'%1$s -in "%2$s" -out "%3$s" -quiet', :clustalo=>'%1$s -i "%2$s" -o "%3$s" --threads=%4$d --force'}
+      aligner: :clustalo, simulator: :grinder,
+      simulatorbin:{grinder:"grinder"},
+      simulatorcmd:{grinder:"%1$s -reference_file \"%2$s\" -cf \"%3$f\" " +
+	 "-dc '-~*NnKkMmRrYySsWwBbVvHhDdXx' -md uniform 0.1 -mr 95 5 " +
+	 "-rd %4$d uniform 5 -base_name \"%5$s\""},
+      alignerbin:{muscle:"muscle", clustalo:"clustalo"},
+      alignercmd:{muscle:"%1$s -in \"%2$s\" -out \"%3$s\" -quiet",
+	 clustalo:"%1$s -i \"%2$s\" -o \"%3$s\" --threads=%4$d --force"}
    })
    @@HAS_BUILD_GEMS = nil
    def self.ebirest() @@EBIREST ; end
@@ -33,90 +39,65 @@ class ROCker
    end
 
    #================================[ Utilities ]
-   def genes2genomes(gene_ids)
-      genomes = []
-      ids = Array.new(gene_ids)
-      while ids.size>0
-	 doc = ebiFetch(:uniprotkb, ids.shift(200), :annot).split("\n")
-	 genomes += doc.grep( /^DR\s+EMBL;/ ).map do |ln|
-	    r=ln.split('; ')
-	    {:genome_id=>r[1], :transl_id=>r[2]}
-	 end
-      end
-      genomes.uniq
-   end
-   def genome2taxid(genome_id)
-      ln = ebiFetch('embl', [genome_id], 'annot').split(/[\n\r]/).grep(/^FT\s+\/db_xref="taxon:/).first
-      return ln if ln.nil?
-      ln.sub(/.*"taxon:(\d+)".*/, "\\1")
-   end
-   def genome2taxon(genome_id, rank='species')
-      xml = ebiFetch('taxonomy', [genome2taxid(genome_id)], 'enataxonomyxml').gsub(/\s*\n\s*/,'')
-      xml.scan(/<taxon [^>]+>/).grep(/rank="#{rank}"/).first.sub(/.* taxId="(\d+)".*/,"\\1")
-   end
    def restcall(url, outfile=nil)
       $stderr.puts "   # Calling: #{url}" if @o[:debug]
-      response = RestClient::Request.execute(:method=>:get,  :url=>url, :timeout=>600)
-      raise "Unable to reach EBI REST client, error code #{response.code}." unless response.code == 200
+      response = RestClient::Request.execute(:method=>:get, :url=>url,
+	 :timeout=>600)
+      raise "Unable to reach EBI REST client, error code " +
+	 response.code.to_s + "." unless response.code == 200
       unless outfile.nil?
-	 ohf = File.open(outfile, 'w')
+	 ohf = File.open(outfile, "w")
 	 ohf.print response.to_s
 	 ohf.close
       end
       response.to_s
    end
    def ebiFetch(db, ids, format, outfile=nil)
-      url = "#{ROCker.ebirest}/dbfetch/dbfetch/#{db.to_s}/#{ids.join(",")}/#{format.to_s}"
-      res = self.restcall url
-      unless outfile.nil?
-	 ohf = File.open(outfile, 'w')
-	 ohf.print res
-	 ohf.close
-      end
-      res
+      url = "#{ROCker.ebirest}/dbfetch/dbfetch/" +
+	 "#{db.to_s}/#{ids.join(",")}/#{format.to_s}"
+      self.restcall url, outfile
    end
-   def get_coords_from_gff3(genome_ids, protein_ids, transl_ids, thread_id, json_file)
-      positive_coords = {}
-      genomes_org = {}
+   def get_coords_from_gff3(genome_ids, pset, thread_id, json_file)
+      coords = {}
       i = 0
       genome_ids.each do |genome_id|
-	 print "  * scanning #{(i+=1).ordinalize} genome out of #{genome_ids.size} in first thread.  \r" if thread_id==0 and not @o[:q]
-	 unless @o[:pertaxon].nil?
-	    genome_taxon = genome2taxon(genome_id, @o[:pertaxon])
-	    genomes_org[ genome_taxon.to_sym ] ||= []
-	    genomes_org[ genome_taxon.to_sym ] << genome_id
-	 end
+	 print "  * scanning #{(i+=1).ordinalize} genome out of " +
+	    "#{genome_ids.size} in first thread.  \r" if
+	    thread_id==0 and not @o[:q]
 	 genome_file = @o[:baseout] + ".src." + genome_id + ".gff3"
 	 if @o[:reuse] and File.size? genome_file
-	    ifh = File.open(genome_file, 'r')
+	    ifh = File.open(genome_file, "r")
 	    doc = ifh.readlines.grep(/^[^#]/)
 	    ifh.close
 	 else
 	    genome_file=nil unless @o[:noclean]
-	    doc = ebiFetch(:embl, [genome_id], :gff3, genome_file).split("\n").grep(/^[^#]/)
+	    doc = ebiFetch(:embl, [genome_id], :gff3,
+	       genome_file).split("\n").grep(/^[^#]/)
 	 end
 	 doc.each do |ln|
 	    next if ln =~ /^#/
 	    r = ln.chomp.split /\t/
 	    next if r.size < 9
-	    prots = r[8].split(/;/).grep(/^db_xref=UniProtKB[\/A-Za-z-]*:/){ |xref| xref.split(/:/)[1] }
-	    p = prots.select{ |id| protein_ids.include? id }.first
-	    trans = r[8].split(/;/).grep(/^protein_id=/){ |pid| pid.split(/=/)[1] }
-	    t = trans.select{ |id|  transl_ids.include? id }.first
+	    prots = r[8].split(/;/).grep(
+	       /^db_xref=UniProtKB[\/A-Za-z-]*:/){ |xref| xref.split(/:/)[1] }
+	    p = prots.select{ |id| pset.ids.include? id }.first
+	    trans = r[8].split(/;/).grep(
+	       /^protein_id=/){ |pid| pid.split(/=/)[1] }
+	    t = trans.select{ |id| pset.tranids.include? id }.first
 	    next if p.nil? and t.nil?
-	    positive_coords[ r[0].to_sym ] ||= []
-	    positive_coords[ r[0].to_sym ] << {
-	       :prot_id	=> p,
-	       :tran_id => t,
-	       :from	=> r[3].to_i,
-	       :to	=> r[4].to_i,
-	       :strand	=> r[6]
+	    coords[ r[0].to_sym ] ||= []
+	    coords[ r[0].to_sym ] << {
+	       prot_id:	p,
+	       tran_id:	t,
+	       from:	r[3].to_i,
+	       to:	r[4].to_i,
+	       strand:	r[6]
 	    }
 	 end
       end
       print "\n" if thread_id==0 and not @o[:q]
-      ofh = File.open json_file, "w"
-      ofh.print({:positive_coords=>positive_coords, :genomes_org=>genomes_org}.to_json)
+      ofh = File.open(json_file, "w")
+      ofh.print({coords:coords}.to_json)
       ofh.close
    end
    
@@ -124,211 +105,251 @@ class ROCker
    def build!
       # Check requirements
       puts "Testing environment." unless @o[:q]
-      @o[:searchcmd] = @o[:searchcmd][@o[:search]] if @o[:searchcmd].is_a? Hash
-      @o[:makedbcmd] = @o[:makedbcmd][@o[:search]] if @o[:makedbcmd].is_a? Hash
-      @o[:alignercmd] = @o[:alignercmd][@o[:aligner]] if @o[:alignercmd].is_a? Hash
-      @o[:simulatorcmd] = @o[:simulatorcmd][@o[:simulator]] if @o[:simulatorcmd].is_a? Hash
-      @o[:alignerbin] = @o[:alignerbin][@o[:aligner]] if @o[:alignerbin].is_a? Hash
-      @o[:simulatorbin] = @o[:simulatorbin][@o[:simulator]] if @o[:simulatorbin].is_a? Hash
+      {  searchcmd: :search, makedbcmd: :search,
+	 alignercmd: :aligner, alignerbin: :aligner,
+	 simulatorcmd: :simulator, simulatorbin: :simulator
+      }.each_pair { |k,v| @o[k] = @o[k][@o[v]] if @o[k].is_a? Hash }
       @o[:nosearch]=true if @o[:nosimulate]
-      raise "Unsatisfied requirements, please see the help message (-h)." unless ROCker.has_build_gems?
-      @o[:positive] += @o[:posori] unless @o[:posori].nil?
-      @o[:positive] += File.readlines(@o[:posfile]).map{ |l| l.chomp } unless @o[:posfile].nil?
-      @o[:negative] += File.readlines(@o[:negfile]).map{ |l| l.chomp } unless @o[:negfile].nil?
-      unless @o[:aln].nil?
-         aln = Alignment.new
-	 aln.read_fasta @o[:aln]
-	 @o[:positive] += aln.get_ids
-      end
-      raise "-p or -P are mandatory." if @o[:positive].size==0
+      raise "Unsatisfied requirements, please see the help message (-h)." unless
+	 ROCker.has_build_gems?
+      protein_set = {}
+      protein_set[:+] = ProteinSet.new(self,@o[:positive],@o[:posfile],@o[:aln])
+      protein_set[:-] = ProteinSet.new(self,@o[:negative],@o[:negfile])
+      raise "-p, -P, or -a are mandatory." if protein_set[:+].empty?
       raise "-o/--baseout is mandatory." if @o[:baseout].nil?
-      if @o[:positive].size == 1 and not @o[:noaln]
-	 warn "\nWARNING: Positive set contains only one sequence, turning off alignment.\n\n"
+      if protein_set[:+].size==1 and not @o[:noaln]
+	 warn "\nWARNING: Positive set contains only one sequence, turning " +
+	    "off alignment.\n\n"
 	 @o[:noaln] = true
       end
       unless @o[:nosimulate]
-	 self.bash "#{@o[:simulatorbin]} --version", "--simulator-bin must be executable. Is Grinder installed?" if @o[:simulator]==:grinder
+	 self.bash("#{@o[:simulatorbin]} --version",
+	    "--simulator-bin must be executable. Is Grinder installed?") if
+	    @o[:simulator]==:grinder
       end
       unless @o[:noaln]
-	 self.bash "#{@o[:alignerbin]} -version", "--aligner-bin must be executable. Is Muscle installed?" if @o[:aligner]==:muscle
-	 self.bash "#{@o[:alignerbin]} --version", "--aligner-bin must be executable. Is ClustalOmega installed?" if @o[:aligner]==:clustalo
+	 self.bash("#{@o[:alignerbin]} -version",
+	    "--aligner-bin must be executable. Is Muscle installed?") if
+	    @o[:aligner]==:muscle
+	 self.bash("#{@o[:alignerbin]} --version",
+	    "--aligner-bin must be executable. Is ClustalOmega installed?") if
+	    @o[:aligner]==:clustalo
       end
       unless @o[:nosearch]
-	 self.bash "#{@o[:searchbins]}makeblastdb -version", "--search-bins must contain executables. Is BLAST+ installed?" if @o[:search]==:blast
-	 self.bash "#{@o[:searchbins]}diamond --help", "--search-bins must contain executables. Is DIAMOND installed?" if @o[:search]==:diamond
+	 self.bash("#{@o[:searchbins]}makeblastdb -version",
+	    "--search-bins must contain executables. Is BLAST+ installed?") if
+	    @o[:search]==:blast
+	 self.bash("#{@o[:searchbins]}diamond --help",
+	    "--search-bins must contain executables. Is DIAMOND installed?") if
+	    @o[:search]==:diamond
       end
 
       # Download genes
       puts "Downloading gene data." unless @o[:q]
       ref_file = @o[:baseout] + ".ref.fasta"
-      if @o[:posori].nil? and @o[:posfile].nil? and not @o[:aln].nil?
+      if not protein_set[:+].aln.nil?
 	 puts "  * reusing aligned sequences as positive set." unless @o[:q]
-	 f = File.open(ref_file, "w")
-	 f.print aln.to_seq_s
-	 f.close
+	 protein_set[:+].get_from_aln(ref_file, aln)
 	 @o[:noaln] = true
       elsif @o[:reuse] and File.size? ref_file
 	 puts "  * reusing positive set: #{ref_file}." unless @o[:q]
       else
-	 puts "  * downloading #{@o[:positive].size} sequence(s) in positive set." unless @o[:q]
-	 $stderr.puts "   # #{@o[:positive]}" if @o[:debug]
-	 ids = Array.new(@o[:positive])
-	 f = File.open(ref_file, "w")
-	 while ids.size>0
-	    f.print ebiFetch(:uniprotkb, ids.shift(200), :fasta)
-	 end
-	 f.close
+	 puts "  * downloading #{protein_set[:+].size} sequence(s) in " +
+	    "positive set." unless @o[:q]
+	 $stderr.puts "   # #{protein_set[:+].ids}" if @o[:debug]
+	 protein_set[:+].download(ref_file)
       end
-      genome_ids = {:positive=>[], :negative=>[]}
-      transl_ids = {:positive=>[], :negative=>[]}
-      [:positive, :negative].each do |set|
-         unless @o[set].size==0
-	    puts "  * linking genomes from #{@o[set].size} #{set.to_s} sequence(s)." unless @o[:q]
-	    $stderr.puts "   # #{@o[set]}" if @o[:debug]
-	    r = genes2genomes(@o[set])
-	    genome_ids[set] = r.map{|i| i[:genome_id]}.uniq
-	    transl_ids[set] = r.map{|i| i[:transl_id]}.uniq
+      [:+, :-].each do |set|
+         unless protein_set[set].empty?
+	    puts "  * linking genomes from #{protein_set[set].size} " +
+	       "[#{set.to_s}] sequence(s)." unless @o[:q]
+	    $stderr.puts "   # #{protein_set[set].ids}" if @o[:debug]
+	    protein_set[set].get_genomes!
 	 end
       end
-      raise "No genomes associated with the positive set." if genome_ids[:positive].size==0
-      genome_ids[:positive] = genome_ids[:positive].sample( (genome_ids[:positive].size*@o[:genomefrx]).round ) if @o[:genomefrx]
-      raise "No positive genomes selected for metagenome construction, is --genome-frx too small?" if genome_ids[:positive].empty?
-      all_genome_ids = genome_ids.values.reduce(:+).uniq
+      raise "No genomes associated with the positive set." if
+	 protein_set[:+].genomes.empty?
+      genome_set = {:+ => GenomeSet.new(self, protein_set[:+].genomes),
+	 :- => GenomeSet.new(self, protein_set[:-].genomes)}
       
       # Locate genes
       puts "Analyzing genome data." unless @o[:q]
       coords_file = @o[:baseout] + ".src.coords"
       if @o[:reuse] and File.size? coords_file
 	 puts "  * reusing coordinates: #{coords_file}." unless @o[:q]
-	 c = JSON.parse File.read(coords_file), {:symbolize_names=>true}
+	 c = JSON.parse File.read(coords_file), {symbolize_names:true}
 	 positive_coords = c[:positive_coords]
-	 genome_org = c[:genome_org]
+	 negative_coords = c[:negative_coords]
+	 genome_set[:+].taxa = c[:taxa_pos]
+	 genome_set[:-].taxa = c[:taxa_neg]
       else
-	 thrs = [@o[:thr], genome_ids[:positive].size].min
-	 puts "  * downloading and parsing #{genome_ids[:positive].size} GFF3 document(s) in #{thrs} threads." unless @o[:q]
-	 $stderr.puts "   # Looking for proteins: #{@o[:positive]}" if @o[:debug]
-	 $stderr.puts "   # Looking for translations: #{transl_ids[:positive]}" if @o[:debug]
-	 $stderr.puts "   # Looking into: #{genome_ids[:positive]}" if @o[:debug]
-	 thr_obj = []
-	 (0 .. (thrs-1)).each do |thr_i|
-	    ids_to_parse = []
-	    (0 .. (genome_ids[:positive].size-1)).each do |i|
-	       ids_to_parse << genome_ids[:positive][i] if (i % thrs)==thr_i
-	    end
-	    json_file = @o[:baseout] + ".src.coords." + thr_i.to_s
-	    thr_obj << json_file
-	    fork do
-	       get_coords_from_gff3(ids_to_parse, @o[:positive], transl_ids[:positive], thr_i, json_file)
-	    end
-	 end
-	 Process.waitall
-	 # Combine results
-	 positive_coords = {}
-	 genomes_org = {}
-	 genome_org = {}
-	 thr_obj.each do |t|
-	    raise "Thread failed without error trace: #{t}" unless File.exist? t
-	    o = JSON.parse File.read(t), {:symbolize_names=>true, :create_additions=>true}
-	    o[:positive_coords].each_pair do |k,v|
-	       positive_coords[ k ] ||= []
-	       positive_coords[ k ] += v
+	 all_coords = {}
+	 [:+, :-].each do |set_type|
+	    all_coords[set_type] = {}
+	    next if genome_set[set_type].empty?
+	    thrs = [@o[:thr], genome_set[set_type].size].min
+	    puts "  * downloading and parsing #{genome_set[set_type].size} " +
+	       "GFF3 document(s) in #{thrs} threads." unless @o[:q]
+	    $stderr.puts "   # Looking for translations: " +
+	       "#{protein_set[set_type].tranids}" if @o[:debug]
+	    $stderr.puts "   # Looking into: #{genome_set[set_type].ids}" if
+	       @o[:debug]
+	    # Launch threads
+	    thr_obj = []
+	    (0 .. (thrs-1)).each do |thr_i|
+	       ids_to_parse = []
+	       (0 .. (genome_set[set_type].size-1)).each do |i|
+		  ids_to_parse << protein_set[set_type].genomes[i] if
+		     (i % thrs) == thr_i
+	       end
+	       json_file = @o[:baseout] + ".src.coords." + thr_i.to_s + ".tmp"
+	       thr_obj << json_file
+	       fork do
+		  get_coords_from_gff3(ids_to_parse, protein_set[set_type],
+		     thr_i, json_file)
+	       end
 	    end
-	    o[:genomes_org].each_pair do |k,v|
-	       genomes_org[ k ] ||= []
-	       genomes_org[ k ] << v
+	    # Combine results
+	    Process.waitall
+	    thr_obj.each do |t|
+	       raise "Thread failed without error trace: #{t}" unless
+		  File.exist? t
+	       o = JSON.parse(File.read(t), {symbolize_names:true})
+	       o[:coords].each_pair do |k,v|
+		  all_coords[set_type][ k ] ||= []
+		  all_coords[set_type][ k ] += v
+	       end
+	       File.unlink t
 	    end
-	    File.unlink t
-	 end
+	 end # [:+, :-].each
+	 positive_coords = all_coords[:+]
+	 negative_coords = all_coords[:-]
 	 # Select one genome per taxon
 	 unless @o[:pertaxon].nil?
-	    genomes_org.each_pair{ |k,v| genome_org[ k ] = v.sample.first }
+	    puts "  Selecting genomes by #{@o[:pertaxon]}." unless @o[:q]
+	    [:+,:-].each{ |set| genome_set[set].choose_genomes! @o[:pertaxon] }
 	 end
-	 # Save coordinates
+	 # Save coordinates and taxa
 	 ofh = File.open(coords_file, "w")
-	 ofh.print JSON.pretty_generate({:positive_coords=>positive_coords, :genome_org=>genome_org})
+	 ofh.print JSON.pretty_generate({
+	    positive_coords:positive_coords,
+	    negative_coords:negative_coords,
+	    taxa_pos:genome_set[:+].taxa,
+	    taxa_neg:genome_set[:-].taxa})
 	 ofh.close
-      end
+      end # if @o[:reuse] and File.size? coords_file ... else
       unless @o[:pertaxon].nil?
-	 genome_ids[:positive] = genome_org.values
-	 puts "  Using #{genome_org.size} genome(s) after filtering by #{@o[:pertaxon]}." unless @o[:q]
+	 puts "  Using " +
+	    [:+,:-].map{ |set| genome_set[set].size }.reduce(:+).to_s +
+	    " genome(s) after filtering by #{@o[:pertaxon]}." unless @o[:q]
       end
-      all_genome_ids = genome_ids.values.reduce(:+).uniq
-      found = positive_coords.values.map{ |a| a.map{ |b| b[:prot_id] } }.reduce(:+).compact.uniq
-      unknown_pid = positive_coords.values.map{ |a| a.map{ |b| b[:prot_id].nil? ? b[:tran_id] : nil } }.reduce(:+).compact.uniq
-      raise "Cannot find the genomic location of any provided sequence." if found.nil?
-      missing = @o[:positive] - found
-      warn "\nWARNING: Cannot find genomic location of sequence(s) #{missing.join(',')}.\nMissing: #{missing.size}, Unlinked translations: #{unknown_pid.size}\n\n" unless missing.size==0 or missing.size==unknown_pid.size or @o[:genomefrx]<1.0
+      found = protein_set[:+].in_coords(positive_coords)
+      raise "Cannot find the genomic location of any provided sequence." if
+	 found.nil?
+      missing = protein_set[:+].ids - found
+      warn "\nWARNING: Cannot find genomic location of #{missing.size} " +
+	 "sequence(s) #{missing.join(",")}.\n\n" unless missing.empty?
       
       # Download genomes
-      genomes_file = @o[:baseout] + '.src.fasta'
+      genome_set[:all] = GenomeSet.new(self,
+	 genome_set[ :+ ].ids + genome_set[ :- ].ids)
+      genomes_file = @o[:baseout] + ".src.fasta"
       if @o[:reuse] and File.size? genomes_file
 	 puts "  * reusing existing file: #{genomes_file}." unless @o[:q]
       else
-	 puts "  * downloading #{all_genome_ids.size} genome(s) in FastA." unless @o[:q]
-	 $stderr.puts "   # #{all_genome_ids}" if @o[:debug]
-	 ids = Array.new(all_genome_ids)
-	 ofh = File.open(genomes_file, 'w')
-	 while ids.size>0
-	    ofh.print ebiFetch('embl', ids.shift(200), 'fasta')
-	 end
-	 ofh.close
+	 puts "  * downloading " + genome_set[:all].size.to_s +
+	    " genome(s) in FastA." unless @o[:q]
+	 $stderr.puts "   # #{genome_set[:all].ids}" if @o[:debug]
+	 genome_set[:all].download genomes_file
       end
       
       # Generate metagenome
       unless @o[:nosimulate]
 	 puts "Generating in silico metagenome" unless @o[:q]
 	 if @o[:reuse] and File.size? @o[:baseout] + ".mg.fasta"
-	    puts "  * reusing existing file: #{@o[:baseout]}.mg.fasta." unless @o[:q]
+	    puts "  * reusing existing file: #{@o[:baseout]}.mg.fasta." unless
+	       @o[:q]
 	 else
-	    all_src = File.readlines("#{@o[:baseout]}.src.fasta").select{ |l| l =~ /^>/ }.size
+	    all_src = File.readlines("#{@o[:baseout]}.src.fasta"
+	       ).select{ |l| l =~ /^>/ }.size
 	    thrs = [@o[:thr], all_src].min
-	    puts "  * simulating metagenomes and tagging positive reads in #{thrs} threads." unless @o[:q]
-	    $stderr.puts "   # #{positive_coords}" if @o[:debug]
 	    thr_obj = []
-	    seqs_per_thr = (all_src/thrs).ceil
+	    seqs_per_thr = (all_src.to_f/thrs).ceil
+	    thrs = (all_src.to_f/seqs_per_thr).ceil
+	    puts "  * simulating metagenomes and tagging positive reads in " +
+	       thrs.to_s + " threads." unless @o[:q]
+	    $stderr.puts "   # #{positive_coords}" if @o[:debug]
 	    (0 .. (thrs-1)).each do |thr_i|
 	       output = @o[:baseout] + ".mg.fasta.#{thr_i.to_s}"
 	       thr_obj << output
 	       fork do
 		  seqs_a = thr_i*seqs_per_thr + 1
-		  seqs_b = [seqs_a + seqs_per_thr, all_src].min
+		  seqs_b = [seqs_a + seqs_per_thr - 1, all_src].min
 		  # Create sub-fasta
-		  ofh = File.open("#{@o[:baseout]}.src.fasta.#{thr_i.to_s}", "w")
-		  ifh = File.open("#{@o[:baseout]}.src.fasta", "r")
+		  ofh = File.open("#{@o[:baseout]}.src.fasta.#{thr_i.to_s}","w")
+		  ifh = File.open("#{@o[:baseout]}.src.fasta","r")
 		  seq_i = 0
 		  while l = ifh.gets
 		     seq_i+=1 if l =~ /^>/
-		     break if seq_i > seqs_b
+			     break if seq_i > seqs_b
 		     ofh.print l if seq_i >= seqs_a
 		  end
 		  ifh.close
 		  ofh.close
 
-		  # Run simulator (except if the temporal file is already there and can be reused)
-		  unless @o[:reuse] and File.size? @o[:baseout] + ".mg.tmp.#{thr_i.to_s}-reads.fa"
-		     bash sprintf(@o[:simulatorcmd], @o[:simulatorbin], "#{@o[:baseout]}.src.fasta.#{thr_i.to_s}", @o[:seqdepth]*@o[:readlen].to_f, @o[:readlen], "#{@o[:baseout]}.mg.tmp.#{thr_i.to_s}")
-		  end
+		  # Run simulator (except if the temporal file is already
+		  # there and can be reused)
+		  bash sprintf(@o[:simulatorcmd], @o[:simulatorbin],
+		     "#{@o[:baseout]}.src.fasta.#{thr_i.to_s}",
+		     @o[:seqdepth]*@o[:readlen].to_f, @o[:readlen],
+		     "#{@o[:baseout]}.mg.tmp.#{thr_i.to_s}") unless
+			@o[:reuse] and
+			File.size? @o[:baseout] +
+			".mg.tmp.#{thr_i.to_s}-reads.fa"
 
-		  # Tag positives
-		  puts "  * tagging positive reads [thread #{thr_i.to_s}]." unless @o[:q]
-		  ifh = File.open(@o[:baseout] + ".mg.tmp.#{thr_i.to_s}-reads.fa", 'r')
-		  ofh = File.open(@o[:baseout] + ".mg.fasta.#{thr_i.to_s}", 'w')
+		  # Tag positive and negative reads
+		  puts "  * tagging reads [thread #{thr_i}]." unless
+		     @o[:q]
+		  ifh = File.open(@o[:baseout] + ".mg.tmp.#{thr_i}-reads.fa",
+		     "r")
+		  ofh = File.open(@o[:baseout] + ".mg.fasta.#{thr_i}", "w")
 		  while l = ifh.gets
 		     if l =~ /^>/
-			rd = /^>(?<id>\d+) reference=[A-Za-z]+\|(?<genome_id>[A-Za-z0-9_]+)\|.* position=(?<comp>complement\()?(?<from>\d+)\.\.(?<to>\d+)\)? /.match(l)
-			raise "Cannot parse simulated read's defline, are you using Grinder?: #{l}" if rd.nil?
+			rd = %r{
+			   ^>(?<id>\d+)\s
+			   reference=[A-Za-z]+\|
+			   (?<genome_id>[A-Za-z0-9_]+)\|.*\s
+			   position=(?<comp>complement\()?(?<from>\d+)\.\.
+			   (?<to>\d+)\)?\s
+			}x.match(l)
+			raise "Cannot parse simulated read's defline, are " +
+			   "you using Grinder?: #{l}" if rd.nil?
 			positive = false
 			positive_coords[rd[:genome_id].to_sym] ||= []
 			positive_coords[rd[:genome_id].to_sym].each do |gn|
 			   left  = rd[:to].to_i - gn[:from]
 			   right = gn[:to] - rd[:from].to_i
-			   if (left*right >= 0) and ([left, right].min >= @o[:minovl])
+			   if (left*right >= 0) and
+				 ([left, right].min >= @o[:minovl])
 			      positive = true
 			      break
 			   end
 			end
-			l = ">#{thr_i.to_s}_#{rd[:id]}#{positive ? "@%" : ""} " +
-			   "ref=#{rd[:genome_id]}:#{rd[:from]}..#{rd[:to]}#{(rd[:comp]=='complement(')?'-':'+'}\n"
+			negative = false
+			negative_coords[rd[:genome_id].to_sym] ||= []
+			negative_coords[rd[:genome_id].to_sym].each do |gn|
+			   left  = rd[:to].to_i - gn[:from]
+			   right = gn[:to] - rd[:from].to_i
+			   if (left*right >= 0) and
+				 ([left, right].min >= @o[:minovl])
+			      negative = true
+			      break
+			   end
+			end
+			l = ">#{thr_i.to_s}_#{rd[:id]}" +
+			   "#{positive ? "@%" : (negative ? "@$" : "")} " +
+			   "ref=#{rd[:genome_id]}:#{rd[:from]}..#{rd[:to]}" +
+			   "#{(rd[:comp]=="complement(") ? "-" : "+"}\n"
 		     end
 		     ofh.print l
 		  end
@@ -338,9 +359,10 @@ class ROCker
 	    end # (1 .. thrs).each
 	    Process.waitall
 	    # Concatenate results
-	    ofh = File.open(@o[:baseout] + ".mg.fasta", 'w')
+	    ofh = File.open(@o[:baseout] + ".mg.fasta", "w")
 	    thr_obj.each do |t|
-	       raise "Thread failed without error trace: #{t}" unless File.exist? t
+	       raise "Thread failed without error trace: #{t}" unless
+		  File.exist? t
 	       ifh = File.open(t, "r")
 	       while l = ifh.gets
 	          ofh.print l
@@ -356,23 +378,33 @@ class ROCker
       unless @o[:noaln]
 	 puts "Aligning reference set." unless @o[:q]
 	 if @o[:reuse] and File.size? "#{@o[:baseout]}.ref.aln"
-	    puts "  * reusing existing file: #{@o[:baseout]}.ref.aln." unless @o[:q]
+	    puts "  * reusing existing file: #{@o[:baseout]}.ref.aln." unless
+	       @o[:q]
 	 else
-	    bash sprintf(@o[:alignercmd], @o[:alignerbin], "#{@o[:baseout]}.ref.fasta", "#{@o[:baseout]}.ref.aln", @o[:thr])
-	    puts "  +--\n  | IMPORTANT NOTE: Manually checking the alignment before\n  | the 'compile' step is *strongly* encouraged.\n  +--\n" unless @o[:q]
+	    bash(sprintf(@o[:alignercmd],
+	       @o[:alignerbin], "#{@o[:baseout]}.ref.fasta",
+	       "#{@o[:baseout]}.ref.aln", @o[:thr]))
+	    puts "  +--\n  | IMPORTANT NOTE: Manually checking the alignment " +
+	       "before\n  | the 'compile' step is *strongly* encouraged.\n  " +
+	       "+--\n" unless @o[:q]
 	 end
       end
       
       # Run similarity search
       unless @o[:nosearch]
-	 puts "Running homology search." unless @o[:q]
+	 puts "Running similarity search." unless @o[:q]
 	 if @o[:reuse] and File.size? "#{@o[:baseout]}.ref.blast"
-	    puts "  * reusing existing file: #{@o[:baseout]}.ref.blast." unless @o[:q]
+	    puts "  * reusing existing file: #{@o[:baseout]}.ref.blast." unless
+	       @o[:q]
 	 else
 	    puts "  * preparing database." unless @o[:q]
-	    bash sprintf(@o[:makedbcmd][@o[:search]], @o[:searchbins], 'prot', "#{@o[:baseout]}.ref.fasta", "#{@o[:baseout]}.ref")
+	    bash(sprintf(@o[:makedbcmd],
+	       @o[:searchbins], "prot", "#{@o[:baseout]}.ref.fasta",
+	       "#{@o[:baseout]}.ref"))
 	    puts "  * running similarity search." unless @o[:q]
-	    bash sprintf(@o[:searchcmd][@o[:search]], @o[:searchbins], 'blastx', "#{@o[:baseout]}.mg.fasta", "#{@o[:baseout]}.ref", "#{@o[:baseout]}.ref.blast", @o[:thr])
+	    bash(sprintf(@o[:searchcmd],
+	       @o[:searchbins], "blastx", "#{@o[:baseout]}.mg.fasta",
+	       "#{@o[:baseout]}.ref", "#{@o[:baseout]}.ref.blast", @o[:thr]))
 	 end
       end
       
@@ -382,7 +414,8 @@ class ROCker
 	 sff  = %w{.src.xml .src.fasta}
 	 sff += %w{.mg.tmp-reads.fa .mg.tmp-ranks.txt} unless @o[:nosimulate]
 	 sff += %w{.ref.phr .ref.pin .ref.psq} unless @o[:nosearch]
-	 sff.each { |sf| File.unlink @o[:baseout] + sf if File.exist? @o[:baseout] + sf }
+	 sff.each { |sf| File.unlink @o[:baseout] + sf if
+	    File.exist? @o[:baseout] + sf }
       end
    end # build!
 end # ROCker