bio-rocker 0.2.5 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 14d82f0e8c6f1cf052b52c82de99c5616ebfd2a3
-  data.tar.gz: 175ffb75e14ecfa7f12073ef8fddff2d8f8bda5d
+  metadata.gz: b8a10cdc85d8b7b54c21d26f12b90c0b3dff4f82
+  data.tar.gz: c837b3c6687f6705dbfc7c959824dd530e7ee932
 SHA512:
-  metadata.gz: d8ce626b7731d7293339c74edbc8ac06005c6b4f988d7cb45e191b24ccc2194091d214d2352c5cfd9e34018abc4b94ed3ffa1b734a5438bc6246fc4f564f6abf
-  data.tar.gz: 0d7187d853a4ac73b91808f52757c94f7d61bc4f19762a7ba4ec97363c55c1eeb32791f78b31ae5d91a7dd72b6779b3bc4cd34b9783e3509bae72be33e4e523a
+  metadata.gz: 869cdadfed2dad125fc11c03133e2f56df53074a54b0f35d8ea5c6674029e7069332e4c35c486b1f39a417aeff01932a7eee30da44e15de036ce1a2d878d15d4
+  data.tar.gz: 823b30e7923c243f8dc8bb122f50426898e2cdda634516cc53e0325b150946b0699e0a9257a55c06ed8868a0f843f0274ed1d23bdf8ef2de9629fafb66f33552

data/bin/ROCker

@@ -16,8 +16,8 @@ require 'optparse'
 $t = {
    'build'   => 'Creates in silico metagenomes and training sets from reference genomes.',
    'compile' => 'Identifies the most discriminant bit-score per alignment position in a set of sequence.',
-   'filter'  => 'Uses a pre-compiled set of bit-score thresholds to filter a search result.',
    'search'  => 'Uses a ROCker compilation to identify reads putatively derived from a set of sequences.',
+   'filter'  => 'Uses a pre-compiled set of bit-score thresholds to filter a search result.',
    'plot'    => 'Generates a graphical representation of the alignment, the thresholds, and the hits.',
 }
 task = (ARGV.size > 0 ? ARGV.shift : '').downcase
@@ -43,49 +43,51 @@ opts = OptionParser.new do |opt|
       opt.on("-p", "--positive ID1,ID2,ID3", Array, "Comma-separated list of UniProtKB IDs corresponding to the 'positive' training set. Required unless -P or -a are used."){ |v| o[:posori]=v }
       opt.on("-n", "--negative ID1,ID2,ID3", Array, "Comma-separated list of UniProtKB IDs corresponding to the 'negative' training set. See also -N."){ |v| o[:negative]=v }
       opt.on("-o", "--baseout PATH", "Prefix for the output files to be generated. Required."){ |v| o[:baseout]=v }
-      #opt.on(      "--nucleotides", "If set, it assumes that the input sequences are in nucleotides (currently not implemented)."){ raise "--nucleotides: This option is currently not implemented." }
       opt.on("-t", "--threads INT", "Number of threads to use. By default: #{ROCker.default :thr}."){ |v| o[:thr]=v.to_i }
       opt.separator ""
       opt.separator "+ ADVANCED BUILDING ARGUMENTS"
       opt.on("-P", "--positive-file PATH", "File containing the positive set (see -p), one UniProtKB ID per line. If used, -p is not required."){ |v| o[:posfile]=v }
       opt.on("-N", "--negative-file PATH", "File containing the negative set (see -n), one UniProtKB ID per line."){ |v| o[:negfile]=v }
       opt.on("-a", "--alignment PATH", "Protein alignment of the reference sequences. The defline must contain UniProtKB ID. If used, -p is not required."){ |v| o[:aln]=v }
-      opt.on("-s", "--seqdepth NUMBER", "Sequencing depth to be used in building the in silico metagenome. By default: '#{ROCker.default :seqdepth}'."){ |v| o[:seqdepth]=v.to_f }
-      opt.on("-v", "--overlap NUMBER", "Minimum overlap with reference gene to tag a read as positive. By default: '#{ROCker.default :minovl}'."){ |v| o[:minovl]=v.to_f }
-      opt.on(      "--genome-frx NUMBER", "Fraction to subsample the positive set genomes to generate the metagenome. By default: #{ROCker.default :genomefrx}"){ |v| o[:genomefrx]=v.to_f }
-      opt.on(      "--per-taxon RANK", "If selected, only one genome per taxon is used to build the metagenome. Valid ranks include: species, genus, family, order, class, phylum.",
-	 "This option replaces --per-genus and --per-species from v0.1.*."){ |v| o[:pertaxon]=v.downcase }
-      opt.on(      "--nometagenome", "Do not create metagenome. Implies --noblast. By default, metagenome is created."){ |v| o[:nomg]=v }
-      opt.on(      "--noblast", "Do not execute BLAST. By default, BLAST is executed."){ |v| o[:noblast]=v }
+      opt.on("-s", "--seqdepth NUMBER", "Sequencing depth (reads/bp) to be used in building the in silico metagenome. By default: '#{ROCker.default :seqdepth}'."){ |v| o[:seqdepth]=v.to_f }
+      opt.on("-l", "--readlen INTEGER", "Average read length of in silico metagenome (in bp). By default: '#{ROCker.default :readlen}'."){ |v| o[:readlen]=v.to_i }
+      opt.on("-v", "--overlap INTEGER", "Minimum overlap (in bp) with reference gene to tag a read as positive. By default: '#{ROCker.default :minovl}'."){ |v| o[:minovl]=v.to_i }
+      opt.on(      "--per-taxon RANK", "If selected, only one genome per taxon is used to build the metagenome. Valid ranks include: species, genus, family, order, class, phylum."){ |v| o[:pertaxon]=v.downcase }
+      opt.on(      "--genome-frx NUMBER", "Fraction to subsample genomes to generate the metagenome. By default: #{ROCker.default :genomefrx}."){ |v| o[:genomefrx]=v.to_f }
+      opt.on(      "--nosimulate", "Do not simulate metagenome. Implies --nosearch. By default, metagenome is simulated."){ |v| o[:nosimulate]=v }
+      opt.on(      "--nosearch", "Do not execute similarity search. By default, it is executed."){ |v| o[:nosearch]=v }
       opt.on(      "--noalignment", "Do not align reference set. By default, references are aligned."){ |v| o[:noaln]=v }
       opt.on(      "--nocleanup", "Keep all intermediate files. By default, intermediate files are removed."){ |v| o[:noclean]=v }
       opt.on(      "--reuse-files", "Re-use existing result files. By default, existing files are ignored."){ |v| o[:reuse]=true }
       opt.separator ""
       opt.separator "+ EXTERNAL SOFTWARE OPTIONS"
-      opt.on("-G", "--grinder PATH", "Path to the grinder executable. By default: '#{ROCker.default :grinder}' (in the $PATH)."){ |v| o[:grinder]=v }
-      opt.on("-M", "--muscle PATH", "Path to the muscle executable. By default: '#{ROCker.default :muscle}' (in the $PATH)."){ |v| o[:muscle]=v }
-      opt.on("-B", "--blastbins PATH", "Path to the Blast+ executables. By default: '#{ROCker.default :blastbins}' (in the $PATH)."){ |v| o[:blastbins]=v }
-      opt.on(      "--grinder-cmd STR", "Command calling grinder, where %1$s: grinder bin, %2$s: input, %3$s: seq. depth, %4$s: output.",
-	 "By default: '#{ROCker.default :grindercmd}'."){ |v| o[:grindercmd]=v }
-      opt.on("--muscle-cmd STR", "Command calling muscle, where %1$s: muscle bin, %2$s: input, %3$s: output.",
-	 "By default: '#{ROCker.default :musclecmd}'."){ |v| o[:musclecmd]=v }
-      opt.on("--blast-cmd STR", "Command calling BLAST search, where %1$s: blast bins, %2$s: program, %3$s: input, %4$s: database, %5$s: output, %6$d: threads.",
-	 "By default: '#{ROCker.default :blastcmd}'."){ |v| o[:blastcmd]=v }
-      opt.on("--makedb-cmd STR", "Command calling BLAST format, where %1$s: blast bins, %2$s: dbtype, %3$s: input, %4$s: database.",
-	 "By default: '#{ROCker.default :makedbcmd}'."){ |v| o[:makedbcmd]=v }
+      opt.on(      "--search STR",	"Similarity search algorithm to use. Supported: 'blast' and 'diamond'. By default: '#{ROCker.default :search}'.")	{ |v| o[:search]=v.to_sym }
+      opt.on(      "--simulator STR",	"In silico metagenome simulator to use. Supported: 'grinder'. By default: '#{ROCker.default :simulator}'.")		{ |v| o[:ssimulator]=v.to_sym }
+      opt.on(      "--aligner STR",	"Multiple alignment algorithm to use. Supported: 'clustalo' and 'muscle'. By default: '#{ROCker.default :aligner}'.")	{ |v| o[:aligner]=v.to_sym }
+      opt.on(      "--search-bins PATH",	"Path to the similarity search executables. By default in the $PATH: '#{ROCker.default :searchbins}'.")		{ |v| o[:searchbins]=v }
+      opt.on(      "--simulator-bin PATH",	"Path to the simulator executable. By default in the $PATH: '#{ROCker.default(:simulatorbin).values.join("' or '")}'.")	{ |v| o[:simulatorbin]=v }
+      opt.on(      "--aligner-bin PATH",	"Path to the aligner executable. By default in the $PATH: '#{ROCker.default(:alignerbin).values.join("' or '")}'.")	{ |v| o[:alignerbin]=v }
+      opt.on(      "--search-cmd STR", "Command calling similarity search, where %1$s: binaries, %2$s: program, %3$s: input, %4$s: database, %5$s: output, %6$d: threads.",
+	 *ROCker.default(:searchcmd).keys.map{|k| "By default if --search #{k}: '#{ROCker.default(:searchcmd)[k]}'."}){ |v| o[:searchcmd]=v }
+      opt.on(      "--makedb-cmd STR", "Command calling database format for similarity search, where %1$s: binaries, %2$s: dbtype, %3$s: input, %4$s: database.",
+	 *ROCker.default(:makedbcmd).keys.map{|k| "By default if --search #{k}: '#{ROCker.default(:makedbcmd)[k]}'."}){ |v| o[:makedbcmd]=v }
+      opt.on(      "--simulator-cmd STR", "Command calling simulator, where %1$s: binary, %2$s: input, %3$s: seq. depth (X), %4$d: read len., %5$s: output.",
+	 *ROCker.default(:simulatorcmd).keys.map{|k| "By default if --simulator #{k}: '#{ROCker.default(:simulatorcmd)[k]}'."}){ |v| o[:simulatorcmd]=v }
+      opt.on("--aligner-cmd STR", "Command calling aligner, where %1$s: binary, %2$s: input, %3$s: output, %4$d: threads.",
+	 *ROCker.default(:alignercmd).keys.map{|k| "By default if --aligner #{k}: '#{ROCker.default(:alignercmd)[k]}'."}){ |v| o[:alignercmd]=v }
    when 'compile'
       opt.separator "+ COMPILATION ARGUMENTS"
       opt.on("-a", "--alignment PATH", "Protein alignment of the reference sequences. Required."){ |v| o[:aln]=v }
       opt.on("-b", "--ref-blast PATH",
       		"Tabular BLAST (blastx) of the test reads vs. the reference dataset. Required unless -t exists."){ |v| o[:blast]=v }
       opt.on("-k", "--rocker PATH", "ROCker file to be created. Required."){ |v| o[:rocker]=v }
-      opt.on(      "--nucleotides", "If set, it assumes that the input sequences are in nucleotides. By default, proteins are assumed."){ raise "--nucleotides: This option is currently not implemented." }
       opt.separator ""
       opt.separator "+ ADVANCED COMPILATION ARGUMENTS"
       opt.on("-t", "--table PATH", "Formated tabular file to be created (or reused). Required unless -b is provided."){ |v| o[:table]=v }
       opt.on(      "--min-score NUMBER", "Minimum Bit-Score to consider a hit. By default: #{ROCker.default :minscore}"){ |v| o[:minscore]=v.to_f }
       opt.on(      "--norefine", "Do not refine windows."){ o[:refine]=false }
       opt.on("-w", "--window INT", "Initial size of alignment windows (in number of AA columns). By default: #{ROCker.default :win}."){ |v| o[:win]=v.to_i }
+      opt.on(      "--reuse-files", "Re-use existing result files. By default, existing files are ignored."){ |v| o[:reuse]=true }
       opt.separator ""
       opt.separator "+ INPUT/OUTPUT"
       opt.separator "   o The input alignment (-a) MUST be in FastA format, and the IDs must"
@@ -107,15 +109,23 @@ opts = OptionParser.new do |opt|
       opt.separator "      5. Bit score threshold set for the window."
       opt.separator "     The file also contains the alignment (commented with #:)."
       opt.separator ""
+   when 'search'
+      opt.on("-k", "--rocker PATH", "ROCker file generated by the compile task (-k). Required."){ |v| o[:rocker]=v }
+      opt.on("-q", "--query PATH", "File containing the query sequences in FastA format. Required."){ |v| o[:query]=v }
+      opt.on("-o", "--out-blast PATH", "Filtered tabular BLAST to be created. Required."){ |v| o[:oblast]=v }
+      opt.separator ""
+      opt.separator "+ EXTERNAL SOFTWARE OPTIONS"
+      opt.on(      "--search STR",	"Similarity search algorithm to use. Supported: 'blast' and 'diamond'. By default: '#{ROCker.default :search}'.")	{ |v| o[:search]=v.to_sym }
+      opt.on(      "--search-bins PATH",	"Path to the similarity search executables. By default in the $PATH: '#{ROCker.default :searchbins}'.")		{ |v| o[:searchbins]=v }
+      opt.on(      "--search-cmd STR", "Command calling similarity search, where %1$s: binaries, %2$s: program, %3$s: input, %4$s: database, %5$s: output, %6$d: threads.",
+	 *ROCker.default(:searchcmd).keys.map{|k| "By default if --search #{k}: '#{ROCker.default(:searchcmd)[k]}'."}){ |v| o[:searchcmd]=v }
+      opt.on(      "--makedb-cmd STR", "Command calling database format for similarity search, where %1$s: binaries, %2$s: dbtype, %3$s: input, %4$s: database.",
+	 *ROCker.default(:makedbcmd).keys.map{|k| "By default if --search #{k}: '#{ROCker.default(:makedbcmd)[k]}'."}){ |v| o[:makedbcmd]=v }
    when 'filter'
       opt.separator "+ FILTERING ARGUMENTS"
       opt.on("-k", "--rocker PATH", "ROCker file generated by the compile task (-k). Required."){ |v| o[:rocker]=v }
       opt.on("-x", "--query-blast PATH", "Tabular BLAST (blastx) of the query reads vs. the reference dataset. Required."){ |v| o[:qblast]=v }
       opt.on("-o", "--out-blast PATH", "Filtered tabular BLAST to be created. Required."){ |v| o[:oblast]=v }
-   when 'search'
-      opt.on("-k", "--rocker PATH", "ROCker file generated by the compile task (-k). Required."){ |v| o[:rocker]=v }
-      opt.on("-q", "--query PATH", "File containing the query sequences in FastA format. Required."){ |v| o[:query]=v }
-      opt.on("-o", "--out-blast PATH", "Filtered tabular BLAST to be created. Required."){ |v| o[:oblast]=v }
    when 'plot'
       opt.separator "+ PLOTTING ARGUMENTS"
       opt.on("-k", "--rocker PATH", "ROCker file generated by the compile task (-k). Required."){ |v| o[:rocker]=v }
@@ -127,7 +137,7 @@ opts = OptionParser.new do |opt|
       opt.on("-t", "--table PATH", "Formated tabular file to be created (or reused). Required unless -b is provided."){ |v| o[:table]=v }
       opt.on(      "--color", "Color alignment by amino acid."){ o[:color]=true }
       opt.on(      "--no-transparency", "Do not use (semi-)transparencies."){ |v| o[:transparency] = v }
-      opt.on(      "--min-score NUMBER", "Minimum Bit-Score to consider a hit. By default: #{ROCker.default :minscore}"){ |v| o[:minscore]=v.to_f }
+      opt.on(      "--min-score NUMBER", "Minimum Bit-Score to consider a hit. By default: #{ROCker.default :minscore}."){ |v| o[:minscore]=v.to_f }
       opt.on(      "--stats-impact", "Plot impact on statistics, instead of absolute values per window."){ o[:impact]=true }
       opt.on(      "--stats-ylim STRING", "Limits of the Y-axis in the bottom panel. By default: '-2,.1' if --stats-impact is set, '50,100' otherwise."){ |v| o[:ylim]=v }
       opt.on("-s", "--subject SBJ1,SBJ2,...", Array,

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 May-14-2015
+# @update Jun-05-2015
 #
 
 require 'rocker/blasthit'
@@ -10,40 +10,20 @@ require 'rocker/rocdata'
 
 class ROCker
    #================================[ Class ]
-   @@EBIREST = 'http://www.ebi.ac.uk/Tools'
    @@DEFAULTS = {
       # General
-      :q=>false, :r=>'R', :nucl=>false, :debug=>false,
-      # Build
-      :positive=>[], :negative=>[], :thr=>2,:genomefrx=>1.0,
-	 # ext. software
-	 :grinder=>'grinder', :muscle=>'muscle', :blastbins=>'', :seqdepth=>3, :minovl=>0.75,
-	 :grindercmd=>'%1$s -reference_file "%2$s" -cf "%3$f" -base_name "%4$s" -dc \'-~*Nn\' -md "uniform 0.1" -mr "95 5" -rd "100 uniform 5"',
-	 :musclecmd=>'%1$s -in "%2$s" -out "%3$s" -quiet',
-	 :blastcmd=>'%1$s%2$s -query "%3$s" -db "%4$s" -out "%5$s" -num_threads %6$d -outfmt 6 -max_target_seqs 1',
-	 :makedbcmd=>'%1$smakeblastdb -dbtype %2$s -in "%3$s" -out "%4$s"',
-      # Compile
-      :refine=>true, :win=>20, :minscore=>0,
-      # Filter
-      :sbj=>[],
-      # Plot
-      :color=>false, :gformat=>'pdf', :width=>9, :height=>9, :impact=>false, :transparency=>true,
+      :q=>false, :r=>'R', :nucl=>false, :debug=>false,:thr=>2,:search=>:blast,
+      # External software
+      :searchbins=>'',
+      :searchcmd=>{
+	 :blast=>'%1$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"'}
    }
-   @@HAS_BUILD_GEMS = nil
-   def self.ebirest() @@EBIREST ; end
    def self.defaults() @@DEFAULTS ; end
    def self.default(k) @@DEFAULTS[k] ; end
-   def self.has_build_gems?
-      return @@HAS_BUILD_GEMS unless @@HAS_BUILD_GEMS.nil?
-      @@HAS_BUILD_GEMS = TRUE
-      begin
-	 require 'rubygems'
-	 require 'restclient'
-      rescue LoadError
-	 @@HAS_BUILD_GEMS = FALSE
-      end
-      @@HAS_BUILD_GEMS
-   end
 
    #================================[ Instance ]
    attr_reader :o
@@ -53,374 +33,6 @@ class ROCker
       RInterface.R_BIN = opts[:r] unless opts[:r].nil?
    end
    
-   #================================[ Build ]
-   def build!
-      # Check requirements
-      puts "Testing environment." unless @o[:q]
-      @o[:noblast]=true if @o[:nomg]
-      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 "-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"
-	 @o[:noaln] = true
-      end
-      self.bash "#{@o[:grinder]} --version", "-G/--grinder must be executable. Is Grinder installed?" unless @o[:nomg]
-      self.bash "#{@o[:muscle]} -version", "-M/--muscle must be executable. Is Muscle installed?" unless @o[:noaln]
-      self.bash "#{@o[:blastbins]}makeblastdb -version", "-B/--blastbins must contain executables. Is BLAST+ installed?" unless @o[:noblast]
-      # Download genes
-      puts "Downloading gene data." unless @o[:q]
-      f = File.open(@o[:baseout] + '.ref.fasta', 'w')
-      if @o[:posori].nil? and @o[:posfile].nil? and not @o[:aln].nil?
-	 puts "  * re-using aligned sequences as positive set." unless @o[:q]
-	 f.print aln.to_seq_s
-	 @o[:noaln] = true
-      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])
-	 while ids.size>0
-	    f.print ebiFetch(:uniprotkb, ids.shift(200), :fasta)
-	 end
-      end
-      f.close
-      genome_ids = {:positive=>[], :negative=>[]}
-      [:positive, :negative].each do |set|
-         unless @o[set].size==0
-	    puts "  * gathering genomes from #{@o[set].size} #{set.to_s} sequence(s)." unless @o[:q]
-	    $stderr.puts "   # #{@o[set]}" if @o[:debug]
-	    genome_ids[set] = genes2genomes(@o[set])
-	 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
-      
-      # Locate genes
-      puts "Analyzing genome data." unless @o[:q]
-      puts "  * downloading and parsing #{genome_ids[:positive].size} GFF3 document(s)." unless @o[:q]
-      $stderr.puts "   # #{genome_ids[:positive]}" if @o[:debug]
-      positive_coords = {}
-      genome_org = {}
-      i = 0
-      genome_ids[:positive].each do |genome_id|
-	 print "  * scanning #{(i+=1).ordinalize} genome out of #{genome_ids[:positive].size}. \r" unless @o[:q]
-	 unless @o[:pertaxon].nil?
-	    genome_taxon = genome2taxon(genome_id, @o[:pertaxon])
-	    next unless genome_org[ genome_taxon ].nil?
-	    genome_org[ genome_taxon ] = genome_id
-	 end
-	 $stderr.puts "   # Looking for any of #{@o[:positive]}" if @o[:debug]
-	 genome_file = @o[:baseout] + '.src.' + i.to_s + '.gff3'
-	 if @o[:reuse] and File.exist? genome_file
-	    puts "  * reusing existing file: #{genome_file}." unless @o[:q]
-	    ifh = File.open(genome_file, 'r')
-	    doc = ifh.readlines.grep(/^[^#]/)
-	    ifh.close
-	 else
-	    genome_file=nil unless @o[:noclean]
-	    res = ebiFetch(:embl, [genome_id], :gff3, genome_file)
-	    doc = res.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{ |p| @o[:positive].include? p }.first
-	    next if p.nil?
-	    positive_coords[ r[0] ] ||= []
-	    positive_coords[ r[0] ] << {
-	       #:strand	=> r[6],
-	       :prot_id	=> p,
-	       :from	=> r[3].to_i,
-	       :to	=> r[4].to_i
-	    }
-	 end
-      end
-      print "\n" unless @o[:q]
-      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]
-      end
-      all_genome_ids = genome_ids.values.reduce(:+).uniq
-      found = positive_coords.values.map{ |a| a.map{ |b| b[:prot_id] } }.reduce(:+)
-      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(',')}.\n\n" unless missing.size==0 or @o[:genomefrx]<1.0 or not @o[:pertaxon].nil?
-      
-      # Download genomes
-      genomes_file = @o[:baseout] + '.src.fasta'
-      if @o[:reuse] and File.exist? 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
-      end
-      
-      # Generate metagenome
-      unless @o[:nomg]
-	 puts "Generating in silico metagenome" unless @o[:q]
-	 if @o[:reuse] and File.exist? @o[:baseout] + ".mg.fasta"
-	    puts "  * reusing existing file: #{@o[:baseout]}.mg.fasta." unless @o[:q]
-	 else
-	    all_src = File.readlines("#{@o[:baseout]}.src.fasta").select{ |l| l =~ /^>/ }.size
-	    thrs = [@o[:thr], all_src].min
-	    puts "  * running grinder and tagging positive reads (#{thrs} threads)." unless @o[:q]
-	    $stderr.puts "   # #{positive_coords}" if @o[:debug]
-	    thr_obj = []
-	    seqs_per_thr = (all_src/thrs).ceil
-	    (0 .. (thrs-1)).each do |thr_i|
-	       thr_obj << Thread.new do
-		  Thread.current[:seqs_a] = thr_i*seqs_per_thr + 1
-		  Thread.current[:seqs_b] = [Thread.current[:seqs_a] + seqs_per_thr, all_src].min
-		  # Create sub-fasta
-		  Thread.current[:ofh] = File.open("#{@o[:baseout]}.src.fasta.#{thr_i.to_s}", 'w')
-		  Thread.current[:ifh] = File.open("#{@o[:baseout]}.src.fasta", 'r')
-		  Thread.current[:seq_i] = 0
-		  while Thread.current[:l] = Thread.current[:ifh].gets
-		     Thread.current[:seq_i]+=1 if Thread.current[:l] =~ /^>/
-		     break if Thread.current[:seq_i] > Thread.current[:seqs_b]
-		     Thread.current[:ofh].print Thread.current[:l] if Thread.current[:seq_i] >= Thread.current[:seqs_a]
-		  end
-		  Thread.current[:ifh].close
-		  Thread.current[:ofh].close
-		  bash sprintf(@o[:grindercmd], @o[:grinder], "#{@o[:baseout]}.src.fasta.#{thr_i.to_s}", @o[:seqdepth], "#{@o[:baseout]}.mg.tmp.#{thr_i.to_s}")
-		  # Tag positives
-		  puts "  * tagging positive reads." unless @o[:q]
-		  Thread.current[:ifh] = File.open(@o[:baseout] + ".mg.tmp.#{thr_i.to_s}-reads.fa", 'r')
-		  Thread.current[:ofh] = File.open(@o[:baseout] + ".mg.fasta.#{thr_i.to_s}", 'w')
-		  while Thread.current[:l]=Thread.current[:ifh].gets
-		     Thread.current[:rd] = /^>(?<id>\d+) reference=[A-Za-z]+\|(?<genome_id>[A-Za-z0-9_]+)\|.* position=(?<comp>complement\()?(?<from>\d+)\.\.(?<to>\d+)\)? /.match(Thread.current[:l])
-		     unless Thread.current[:rd].nil?
-			Thread.current[:positive] = false
-			positive_coords[Thread.current[:rd][:genome_id]] ||= []
-			positive_coords[Thread.current[:rd][:genome_id]].each do |gn|
-			   Thread.current[:left]  = Thread.current[:rd][:to].to_i - gn[:from]
-			   Thread.current[:right] = gn[:to] - Thread.current[:rd][:from].to_i
-			   if (Thread.current[:left]*Thread.current[:right] >= 0) and ([Thread.current[:left], Thread.current[:right]].min/(Thread.current[:rd][:to].to_i-Thread.current[:rd][:from].to_i) >= @o[:minovl])
-			      Thread.current[:positive] = true
-			      break
-			   end
-			end
-			Thread.current[:l] = ">#{Thread.current[:rd][:id]}#{Thread.current[:positive] ? "@%" : ""} ref=#{Thread.current[:rd][:genome_id]}:#{Thread.current[:rd][:from]}..#{Thread.current[:rd][:to]}#{(Thread.current[:rd][:comp]=='complement(')?'-':'+'}\n"
-		     end
-		     Thread.current[:ofh].print Thread.current[:l]
-		  end
-		  Thread.current[:ofh].close
-		  Thread.current[:ifh].close
-		  Thread.current[:output] = @o[:baseout] + ".mg.fasta.#{thr_i.to_s}"
-	       end # Thread.new do
-	    end # (1 .. thrs).each
-	    # Concatenate results
-	    ofh = File.open(@o[:baseout] + ".mg.fasta", 'w')
-	    thr_obj.each do |t|
-	       t.join
-	       raise "Thread failed without error trace: #{t}" if t[:output].nil?
-	       ifh = File.open(t[:output], 'r')
-	       while l = ifh.gets
-	          ofh.print l
-	       end
-	       ifh.close
-	       File.unlink t[:output]
-	    end
-	    ofh.close
-         end
-      end # unless @o[:nomg]
-      # Align references
-      unless @o[:noaln]
-	 puts "Aligning reference set." unless @o[:q]
-	 if @o[:reuse] and File.exist? "#{@o[:baseout]}.ref.aln"
-	    puts "  * reusing existing file: #{@o[:baseout]}.ref.aln." unless @o[:q]
-	 else
-	    bash sprintf(@o[:musclecmd], @o[:muscle], "#{@o[:baseout]}.ref.fasta", "#{@o[:baseout]}.ref.aln")
-	    puts "  +--\n  | IMPORTANT NOTE: Manually checking the alignment before\n  | the 'compile' step is *strongly* encouraged.\n  +--\n" unless @o[:q]
-	 end
-      end
-      # Run BLAST 
-      unless @o[:noblast]
-	 puts "Running homology search." unless @o[:q]
-	 if @o[:reuse] and File.exist? "#{@o[:baseout]}.ref.blast"
-	    puts "  * reusing existing file: #{@o[:baseout]}.ref.blast." unless @o[:q]
-	 else
-	    puts "  * preparing database." unless @o[:q]
-	    bash sprintf(@o[:makedbcmd], @o[:blastbins], (@o[:nucl]?'nucl':'prot'), "#{@o[:baseout]}.ref.fasta", "#{@o[:baseout]}.ref")
-	    puts "  * running BLAST." unless @o[:q]
-	    bash sprintf(@o[:blastcmd], @o[:blastbins], (@o[:nucl]?'blastn':'blastx'), "#{@o[:baseout]}.mg.fasta", "#{@o[:baseout]}.ref", "#{@o[:baseout]}.ref.blast", @o[:thr])
-	 end
-      end
-      # Clean
-      unless @o[:noclean]
-	 puts "Cleaning." unless @o[:q]
-	 sff  = %w{.src.xml .src.fasta}
-	 sff += %w{.mg.tmp-reads.fa .mg.tmp-ranks.txt} unless @o[:nomg]
-	 sff += %w{.ref.phr .ref.pin .ref.psq} unless @o[:noblast]
-	 sff.each { |sf| File.unlink @o[:baseout] + sf if File.exist? @o[:baseout] + sf }
-      end
-   end # build!
-
-   #================================[ Compile ]
-   def compile!
-      raise "-a/--alignment is mandatory." if @o[:aln].nil?
-      raise "-a/--alignment must exist." unless File.exist? @o[:aln]
-      if @o[:table].nil?
-	 raise "-t/--table is mandatory unless -b is provided." if @o[:blast].nil?
-	 @o[:table] = "#{@o[:blast]}.table"
-      end
-      raise "-b/--blast is mandatory unless -t exists." if @o[:blast].nil? and not File.exist? @o[:table]
-      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
-
-      puts "Testing environment." unless @o[:q]
-      bash "echo '' | #{@o[:r]} --vanilla", "-r/--path-to-r must be executable. Is R installed?"
-      bash "echo \"library('pROC')\" | #{@o[:r]} --vanilla", "Please install the 'pROC' library for R first."
-
-      puts "Reading files." unless @o[:q]
-      puts "  * loading alignment: #{@o[:aln]}." unless @o[:q]
-      aln = Alignment.new
-      aln.read_fasta @o[:aln]
-      
-      if File.exist? @o[:table]
-	 puts "  * reusing existing file: #{@o[:table]}." unless @o[:q]
-      else
-	 puts "  * generating table: #{@o[:table]}." unless @o[:q]
-	 blast2table(@o[:blast], @o[:table], aln, @o[:minscore])
-      end
-
-      puts "Analyzing data." unless @o[:q]
-      puts "  * computing windows." unless @o[:q]
-      data = ROCData.new(@o[:table], aln, @o[:win])
-      data.nucl = @o[:nucl]
-      if @o[:refine]
-	 puts "  * refining windows." unless @o[:q]
-	 warn "Insufficient hits to refine results." unless data.refine! @o[:table]
-      end
-      puts "  * saving ROCker file: #{@o[:rocker]}." unless @o[:q]
-      data.save @o[:rocker]
-   end # compile!
-   
-   #================================[ Filter ]
-   def filter!
-      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
-      raise "-x/--query-blast is mandatory." if @o[:qblast].nil?
-      raise "-o/--out-blast is mandatory." if @o[:oblast].nil?
-      
-      puts "Reading ROCker file." unless @o[:q]
-      data = ROCData.new @o[:rocker]
-
-      puts "Filtering BLAST." unless @o[:q]
-      ih = File.open(@o[:qblast], 'r')
-      oh = File.open(@o[:oblast], 'w')
-      while ln = ih.gets
-	 bh = BlastHit.new(ln, data.aln)
-	 oh.print ln if not(bh.sfrom.nil?) and bh.bits >= data.win_at_col(bh.midpoint).thr
-      end
-      ih.close
-      oh.close
-   end # filter!
-   #================================[ Search ]
-   def search!
-      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
-      raise "Code Under development..."
-      # ToDo
-      # [ ... ]
-   end # search!
-   
-   #================================[ Plot ]
-   def plot!
-      raise "-k/--rocker is mandatory." if o[:rocker].nil?
-      if @o[:table].nil?
-	 raise "-t/--table is mandatory unless -b is provided." if @o[:blast].nil?
-	 @o[:table] = "#{@o[:blast]}.table"
-      end
-      raise "-b/--blast is mandatory unless -t exists." if @o[:blast].nil? and not File.exist? @o[:table]
-
-      puts "Testing environment." unless @o[:q]
-      bash "echo '' | #{@o[:r]} --vanilla", "-r/--path-to-r must be executable. Is R installed?"
-
-      puts "Reading files." unless @o[:q]
-      puts "  * loding ROCker file: #{@o[:rocker]}." unless @o[:q]
-      data = ROCData.new @o[:rocker]
-      if File.exist? @o[:table]
-	 puts "  * reusing existing file: #{@o[:table]}." unless @o[:q]
-      else
-	 puts "  * generating table: #{@o[:table]}." unless @o[:q]
-	 blast2table(@o[:blast], @o[:table], data.aln, @o[:minscore])
-      end
-
-      puts "Plotting matches." unless @o[:q]
-      extra = @o[:gformat]=='pdf' ? "" : ", units='in', res=300"
-      @o[:gout] ||= "#{@o[:rocker]}.#{@o[:gformat]}"
-      data.rrun "#{@o[:gformat]}('#{@o[:gout]}', #{@o[:width]}, #{@o[:height]}#{extra});"
-      data.rrun "layout(c(2,1,3), heights=c(2-1/#{data.aln.size},3,1));"
-      some_thr = data.load_table! @o[:table], @o[:sbj], @o[:minscore]
-      data.rrun "par(mar=c(0,4,0,0.5)+.1);"
-      data.rrun "plot(1, t='n', xlim=c(0.5,#{data.aln.cols}+0.5), ylim=range(x$V4)+c(-0.04,0.04)*diff(range(x$V4)), xlab='', ylab='Bit score', xaxs='i', xaxt='n');"
-      data.rrun "noise <- runif(ncol(x),-.2,.2)"
-      data.rrun "arrows(x0=x$V2, x1=x$V3, y0=x$V4+noise, col=ifelse(x$V5==1, rgb(0,0,.5,#{@o[:transparency] ? ".2" : "1"}), rgb(.5,0,0,#{@o[:transparency] ? ".2" : "1"})), length=0);"
-      data.rrun "points(x$V6, x$V4+noise, col=ifelse(x$V5==1, rgb(0,0,.5,#{@o[:transparency] ? ".5" : "1"}), rgb(.5,0,0,#{@o[:transparency] ? ".5" : "1"})), pch=19, cex=1/4);"
-
-      puts "Plotting windows." unless @o[:q]
-      if some_thr
-	 data.rrun "arrows(x0=w$V1, x1=w$V2, y0=w$V5, lwd=2, length=0)"
-	 data.rrun "arrows(x0=w$V2[-nrow(w)], x1=w$V1[-1], y0=w$V5[-nrow(w)], y1=w$V5[-1], lwd=2, length=0)"
-      end
-      data.rrun "legend('bottomright',legend=c('Match span','Match mid-point','Reference','Non-reference')," +
-	 "lwd=c(1,NA,1,1),pch=c(NA,19,19,19),col=c('black','black','darkblue','darkred'),ncol=4,bty='n')"
-
-      puts "Plotting alignment." unless @o[:q]
-      data.rrun "par(mar=c(0,4,0.5,0.5)+0.1);"
-      data.rrun "plot(1, t='n', xlim=c(0,#{data.aln.cols}),ylim=c(1,#{data.aln.seqs.size}),xlab='',ylab='Alignment',xaxs='i',xaxt='n',yaxs='i',yaxt='n',bty='n');"
-      i = 0
-      data.rrun "clr <- rainbow(26, v=1/2, s=3/4);" if @o[:color]
-      data.aln.seqs.values.each do |s|
-         color = s.aln.split(//).map{|c| c=="-" ? "'grey80'" : (@o[:sbj].include?(s.id) ? "'red'" : (@o[:color] ? "clr[#{c.ord-64}]" : "'black'"))}.join(',')
-	 data.rrun "rect((1:#{data.aln.cols-1})-0.5, rep(#{i}, #{data.aln.cols-1}), (1:#{data.aln.cols-1})+0.5, rep(#{i+1}, #{data.aln.cols-1}), col=c(#{color}), border=NA);"
-	 i += 1
-      end
-
-      puts "Plotting statistics." unless @o[:q]
-      data.rrun "par(mar=c(5,4,0,0.5)+.1);"
-      data.rrun "plot(1, t='n', xlim=c(0,#{data.aln.cols}),ylim=c(#{@o[:ylim].nil? ? (@o[:impact] ? "-2,.1" : "50,100") : @o[:ylim]}),xlab='Alignment position (amino acids)',ylab='Precision',xaxs='i');"
-      if some_thr
-	 sn = data.rrun "100*sum(w$tp)/(sum(w$tp)+sum(w$fn))", :float
-	 sp = data.rrun "100*sum(w$tn)/(sum(w$fp)+sum(w$tn))", :float
-	 ac = data.rrun "100*(sum(w$tp)+sum(w$tn))/(sum(w$p)+sum(w$n))", :float
-	 unless @o[:q]
-	    puts "  * sensitivity: #{sn}%"
-	    puts "  * specificity: #{sp}%"
-	    puts "  * accuracy: #{ac}%"
-	 end
-	 data.rrun "pos <- (w$V1+w$V2)/2"
-	 if @o[:impact]
-	    data.rrun "lines(pos[!is.na(w$specificity)], (w$specificity[!is.na(w$specificity)]-#{sp})*w$tp[!is.na(w$specificity)]/sum(w$tp), col='darkred', lwd=2, t='o', cex=1/3, pch=19);"
-	    data.rrun "lines(pos[!is.na(w$sensitivity)], (w$sensitivity[!is.na(w$sensitivity)]-#{sn})*w$tn[!is.na(w$sensitivity)]/sum(w$tn), col='darkgreen', lwd=2, t='o', cex=1/3, pch=19);"
-	    data.rrun "lines(pos[!is.na(w$accuracy)], (w$accuracy[!is.na(w$accuracy)]-#{ac})*(w$tp+w$tn)[!is.na(w$accuracy)]/sum(c(w$tp, w$tn)), col='darkblue', lwd=2, t='o', cex=1/3, pch=19);"
-	 else
-	    data.rrun "lines(pos[!is.na(w$specificity)], w$specificity[!is.na(w$specificity)], col='darkred', lwd=2, t='o', cex=1/3, pch=19);"
-	    data.rrun "lines(pos[!is.na(w$sensitivity)], w$sensitivity[!is.na(w$sensitivity)], col='darkgreen', lwd=2, t='o', cex=1/3, pch=19);"
-	    data.rrun "lines(pos[!is.na(w$accuracy)], w$accuracy[!is.na(w$accuracy)], col='darkblue', lwd=2, t='o', cex=1/3, pch=19);"
-	 end
-	 #data.rrun "lines(pos[!is.na(w$precision)], w$precision[!is.na(w$precision)], col='purple', lwd=2, t='o', cex=1/3, pch=19);"
-      end
-      data.rrun "legend('bottomright',legend=c('Specificity','Sensitivity','Accuracy'),lwd=2,col=c('darkred','darkgreen','darkblue'),ncol=3,bty='n')"
-      data.rrun "dev.off();"
-   end # plot!
-
    #================================[ Utilities ]
    def blast2table(blast_f, table_f, aln, minscore)
       ifh = File.open(blast_f, "r")
@@ -432,39 +44,6 @@ class ROCker
       ifh.close
       ofh.close
    end
-   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{ |ln| ln.split('; ')[1] }
-      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)
-      response = RestClient.get url
-      raise "Unable to reach EBI REST client, error code #{response.code}." unless response.code == 200
-      unless outfile.nil?
-	 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}"
-      $stderr.puts "   # Calling: #{url}" if @o[:debug]
-      self.restcall url
-   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]=='{'
@@ -473,6 +52,14 @@ class ROCker
 end
 
 #================================[ Extensions ]
+# To ROCker
+require 'rocker/step/build'
+require 'rocker/step/compile'
+require 'rocker/step/search'
+require 'rocker/step/filter'
+require 'rocker/step/plot'
+
+# To other
 class Numeric
    def ordinalize
       n= self.to_s

data/lib/rocker/step/build.rb

@@ -0,0 +1,389 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-05-2015
+#
+
+require 'json'
+
+class ROCker
+   #================================[ Class ]
+   @@EBIREST = 'http://www.ebi.ac.uk/Tools'
+   @@DEFAULTS.merge!({:positive=>[], :negative=>[], :genomefrx=>1.0, :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'}
+   })
+   @@HAS_BUILD_GEMS = nil
+   def self.ebirest() @@EBIREST ; end
+   def self.has_build_gems?
+      return @@HAS_BUILD_GEMS unless @@HAS_BUILD_GEMS.nil?
+      @@HAS_BUILD_GEMS = TRUE
+      begin
+	 require 'rubygems'
+	 require 'restclient'
+      rescue LoadError
+	 @@HAS_BUILD_GEMS = FALSE
+      end
+      @@HAS_BUILD_GEMS
+   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
+      unless outfile.nil?
+	 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
+   end
+   def get_coords_from_gff3(genome_ids, protein_ids, transl_ids, thread_id, json_file)
+      positive_coords = {}
+      genomes_org = {}
+      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
+	 genome_file = @o[:baseout] + ".src." + genome_id + ".gff3"
+	 if @o[:reuse] and File.size? genome_file
+	    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(/^[^#]/)
+	 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
+	    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]
+	    }
+	 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.close
+   end
+   
+   #================================[ Build ]
+   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
+      @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 "-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"
+	 @o[:noaln] = true
+      end
+      unless @o[:nosimulate]
+	 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
+      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
+      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?
+	 puts "  * reusing aligned sequences as positive set." unless @o[:q]
+	 f = File.open(ref_file, "w")
+	 f.print aln.to_seq_s
+	 f.close
+	 @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
+      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
+	 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
+      
+      # 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}
+	 positive_coords = c[:positive_coords]
+	 genome_org = c[:genome_org]
+      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
+	    end
+	    o[:genomes_org].each_pair do |k,v|
+	       genomes_org[ k ] ||= []
+	       genomes_org[ k ] << v
+	    end
+	    File.unlink t
+	 end
+	 # Select one genome per taxon
+	 unless @o[:pertaxon].nil?
+	    genomes_org.each_pair{ |k,v| genome_org[ k ] = v.sample.first }
+	 end
+	 # Save coordinates
+	 ofh = File.open(coords_file, "w")
+	 ofh.print JSON.pretty_generate({:positive_coords=>positive_coords, :genome_org=>genome_org})
+	 ofh.close
+      end
+      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]
+      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
+      
+      # Download genomes
+      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
+      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]
+	 else
+	    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
+	    (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
+		  # Create sub-fasta
+		  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
+		     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
+
+		  # 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')
+		  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?
+			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])
+			      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"
+		     end
+		     ofh.print l
+		  end
+		  ofh.close
+		  ifh.close
+	       end # fork
+	    end # (1 .. thrs).each
+	    Process.waitall
+	    # Concatenate results
+	    ofh = File.open(@o[:baseout] + ".mg.fasta", 'w')
+	    thr_obj.each do |t|
+	       raise "Thread failed without error trace: #{t}" unless File.exist? t
+	       ifh = File.open(t, "r")
+	       while l = ifh.gets
+	          ofh.print l
+	       end
+	       ifh.close
+	       File.unlink t
+	    end
+	    ofh.close
+         end
+      end # unless @o[:nosimulate]
+      
+      # Align references
+      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]
+	 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]
+	 end
+      end
+      
+      # Run similarity search
+      unless @o[:nosearch]
+	 puts "Running homology 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]
+	 else
+	    puts "  * preparing database." unless @o[:q]
+	    bash sprintf(@o[:makedbcmd][@o[:search]], @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])
+	 end
+      end
+      
+      # Clean
+      unless @o[:noclean]
+	 puts "Cleaning." unless @o[:q]
+	 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 }
+      end
+   end # build!
+end # ROCker
+

data/lib/rocker/step/compile.rb

@@ -0,0 +1,53 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-05-2015
+#
+
+class ROCker
+   #================================[ Class ]
+   @@DEFAULTS.merge!({:refine=>true, :win=>20, :minscore=>0})
+   
+   #================================[ Compile ]
+   def compile!
+      raise "-a/--alignment is mandatory." if @o[:aln].nil?
+      raise "-a/--alignment must exist." unless File.exist? @o[:aln]
+      if @o[:table].nil?
+	 raise "-t/--table is mandatory unless -b is provided." if @o[:blast].nil? or not File.exist? @o[:blast]
+	 @o[:table] = "#{@o[:blast]}.table"
+      else
+	 @o[:reuse] = true
+      end
+      raise "-b/--blast is mandatory unless -t exists." if @o[:blast].nil? and not File.exist? @o[:table]
+      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
+
+      puts "Testing environment." unless @o[:q]
+      bash "echo '' | #{@o[:r]} --vanilla", "-r/--path-to-r must be executable. Is R installed?"
+      bash "echo \"library('pROC')\" | #{@o[:r]} --vanilla", "Please install the 'pROC' library for R first."
+
+      puts "Reading files." unless @o[:q]
+      puts "  * loading alignment: #{@o[:aln]}." unless @o[:q]
+      aln = Alignment.new
+      aln.read_fasta @o[:aln]
+      
+      if @o[:reuse] and File.exist? @o[:table]
+	 puts "  * reusing existing file: #{@o[:table]}." unless @o[:q]
+      else
+	 puts "  * generating table: #{@o[:table]}." unless @o[:q]
+	 blast2table(@o[:blast], @o[:table], aln, @o[:minscore])
+      end
+
+      puts "Analyzing data." unless @o[:q]
+      puts "  * computing windows." unless @o[:q]
+      data = ROCData.new(@o[:table], aln, @o[:win])
+      data.nucl = @o[:nucl]
+      if @o[:refine]
+	 puts "  * refining windows." unless @o[:q]
+	 warn "Insufficient hits to refine results." unless data.refine! @o[:table]
+      end
+      puts "  * saving ROCker file: #{@o[:rocker]}." unless @o[:q]
+      data.save @o[:rocker]
+   end # compile!
+end # ROCker
+

data/lib/rocker/step/filter.rb

@@ -0,0 +1,32 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-04-2015
+#
+
+class ROCker
+   #================================[ Class ]
+   #@@DEFAULTS.merge!({  })
+   
+   #================================[ Filter ]
+   def filter!
+      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
+      raise "-x/--query-blast is mandatory." if @o[:qblast].nil?
+      raise "-o/--out-blast is mandatory." if @o[:oblast].nil?
+      
+      puts "Reading ROCker file." unless @o[:q]
+      data = ROCData.new @o[:rocker]
+
+      puts "Filtering BLAST." unless @o[:q]
+      ih = File.open(@o[:qblast], 'r')
+      oh = File.open(@o[:oblast], 'w')
+      while ln = ih.gets
+	 bh = BlastHit.new(ln, data.aln)
+	 oh.print ln if not(bh.sfrom.nil?) and bh.bits >= data.win_at_col(bh.midpoint).thr
+      end
+      ih.close
+      oh.close
+   end # filter!
+end # ROCker
+

data/lib/rocker/step/plot.rb

@@ -0,0 +1,93 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-04-2015
+#
+
+class ROCker
+   #================================[ Class ]
+   @@DEFAULTS.merge!({:color=>false, :gformat=>'pdf', :width=>9, :height=>9, :impact=>false, :transparency=>true, :sbj=>[]})
+   
+   #================================[ Search ]
+   def plot!
+      raise "-k/--rocker is mandatory." if o[:rocker].nil?
+      if @o[:table].nil?
+	 raise "-t/--table is mandatory unless -b is provided." if @o[:blast].nil?
+	 @o[:table] = "#{@o[:blast]}.table"
+      end
+      raise "-b/--blast is mandatory unless -t exists." if @o[:blast].nil? and not File.exist? @o[:table]
+
+      puts "Testing environment." unless @o[:q]
+      bash "echo '' | #{@o[:r]} --vanilla", "-r/--path-to-r must be executable. Is R installed?"
+
+      puts "Reading files." unless @o[:q]
+      puts "  * loding ROCker file: #{@o[:rocker]}." unless @o[:q]
+      data = ROCData.new @o[:rocker]
+      if File.exist? @o[:table]
+	 puts "  * reusing existing file: #{@o[:table]}." unless @o[:q]
+      else
+	 puts "  * generating table: #{@o[:table]}." unless @o[:q]
+	 blast2table(@o[:blast], @o[:table], data.aln, @o[:minscore])
+      end
+
+      puts "Plotting matches." unless @o[:q]
+      extra = @o[:gformat]=='pdf' ? "" : ", units='in', res=300"
+      @o[:gout] ||= "#{@o[:rocker]}.#{@o[:gformat]}"
+      data.rrun "#{@o[:gformat]}('#{@o[:gout]}', #{@o[:width]}, #{@o[:height]}#{extra});"
+      data.rrun "layout(c(2,1,3), heights=c(2-1/#{data.aln.size},3,1));"
+      some_thr = data.load_table! @o[:table], @o[:sbj], @o[:minscore]
+      data.rrun "par(mar=c(0,4,0,0.5)+.1);"
+      data.rrun "plot(1, t='n', xlim=c(0.5,#{data.aln.cols}+0.5), ylim=range(x$V4)+c(-0.04,0.04)*diff(range(x$V4)), xlab='', ylab='Bit score', xaxs='i', xaxt='n');"
+      data.rrun "noise <- runif(ncol(x),-.2,.2)"
+      data.rrun "arrows(x0=x$V2, x1=x$V3, y0=x$V4+noise, col=ifelse(x$V5==1, rgb(0,0,.5,#{@o[:transparency] ? ".2" : "1"}), rgb(.5,0,0,#{@o[:transparency] ? ".2" : "1"})), length=0);"
+      data.rrun "points(x$V6, x$V4+noise, col=ifelse(x$V5==1, rgb(0,0,.5,#{@o[:transparency] ? ".5" : "1"}), rgb(.5,0,0,#{@o[:transparency] ? ".5" : "1"})), pch=19, cex=1/4);"
+
+      puts "Plotting windows." unless @o[:q]
+      if some_thr
+	 data.rrun "arrows(x0=w$V1, x1=w$V2, y0=w$V5, lwd=2, length=0)"
+	 data.rrun "arrows(x0=w$V2[-nrow(w)], x1=w$V1[-1], y0=w$V5[-nrow(w)], y1=w$V5[-1], lwd=2, length=0)"
+      end
+      data.rrun "legend('bottomright',legend=c('Match span','Match mid-point','Reference','Non-reference')," +
+	 "lwd=c(1,NA,1,1),pch=c(NA,19,19,19),col=c('black','black','darkblue','darkred'),ncol=4,bty='n')"
+
+      puts "Plotting alignment." unless @o[:q]
+      data.rrun "par(mar=c(0,4,0.5,0.5)+0.1);"
+      data.rrun "plot(1, t='n', xlim=c(0,#{data.aln.cols}),ylim=c(1,#{data.aln.seqs.size}),xlab='',ylab='Alignment',xaxs='i',xaxt='n',yaxs='i',yaxt='n',bty='n');"
+      i = 0
+      data.rrun "clr <- rainbow(26, v=1/2, s=3/4);" if @o[:color]
+      data.aln.seqs.values.each do |s|
+         color = s.aln.split(//).map{|c| c=="-" ? "'grey80'" : (@o[:sbj].include?(s.id) ? "'red'" : (@o[:color] ? "clr[#{c.ord-64}]" : "'black'"))}.join(',')
+	 data.rrun "rect((1:#{data.aln.cols-1})-0.5, rep(#{i}, #{data.aln.cols-1}), (1:#{data.aln.cols-1})+0.5, rep(#{i+1}, #{data.aln.cols-1}), col=c(#{color}), border=NA);"
+	 i += 1
+      end
+
+      puts "Plotting statistics." unless @o[:q]
+      data.rrun "par(mar=c(5,4,0,0.5)+.1);"
+      data.rrun "plot(1, t='n', xlim=c(0,#{data.aln.cols}),ylim=c(#{@o[:ylim].nil? ? (@o[:impact] ? "-2,.1" : "50,100") : @o[:ylim]}),xlab='Alignment position (amino acids)',ylab='Precision',xaxs='i');"
+      if some_thr
+	 sn = data.rrun "100*sum(w$tp)/(sum(w$tp)+sum(w$fn))", :float
+	 sp = data.rrun "100*sum(w$tn)/(sum(w$fp)+sum(w$tn))", :float
+	 ac = data.rrun "100*(sum(w$tp)+sum(w$tn))/(sum(w$p)+sum(w$n))", :float
+	 unless @o[:q]
+	    puts "  * sensitivity: #{sn}%"
+	    puts "  * specificity: #{sp}%"
+	    puts "  * accuracy: #{ac}%"
+	 end
+	 data.rrun "pos <- (w$V1+w$V2)/2"
+	 if @o[:impact]
+	    data.rrun "lines(pos[!is.na(w$specificity)], (w$specificity[!is.na(w$specificity)]-#{sp})*w$tp[!is.na(w$specificity)]/sum(w$tp), col='darkred', lwd=2, t='o', cex=1/3, pch=19);"
+	    data.rrun "lines(pos[!is.na(w$sensitivity)], (w$sensitivity[!is.na(w$sensitivity)]-#{sn})*w$tn[!is.na(w$sensitivity)]/sum(w$tn), col='darkgreen', lwd=2, t='o', cex=1/3, pch=19);"
+	    data.rrun "lines(pos[!is.na(w$accuracy)], (w$accuracy[!is.na(w$accuracy)]-#{ac})*(w$tp+w$tn)[!is.na(w$accuracy)]/sum(c(w$tp, w$tn)), col='darkblue', lwd=2, t='o', cex=1/3, pch=19);"
+	 else
+	    data.rrun "lines(pos[!is.na(w$specificity)], w$specificity[!is.na(w$specificity)], col='darkred', lwd=2, t='o', cex=1/3, pch=19);"
+	    data.rrun "lines(pos[!is.na(w$sensitivity)], w$sensitivity[!is.na(w$sensitivity)], col='darkgreen', lwd=2, t='o', cex=1/3, pch=19);"
+	    data.rrun "lines(pos[!is.na(w$accuracy)], w$accuracy[!is.na(w$accuracy)], col='darkblue', lwd=2, t='o', cex=1/3, pch=19);"
+	 end
+	 #data.rrun "lines(pos[!is.na(w$precision)], w$precision[!is.na(w$precision)], col='purple', lwd=2, t='o', cex=1/3, pch=19);"
+      end
+      data.rrun "legend('bottomright',legend=c('Specificity','Sensitivity','Accuracy'),lwd=2,col=c('darkred','darkgreen','darkblue'),ncol=3,bty='n')"
+      data.rrun "dev.off();"
+   end # plot!
+end # ROCker
+

data/lib/rocker/step/search.rb

@@ -0,0 +1,20 @@
+#
+# @author Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @author Luis (Coto) Orellana
+# @license artistic license 2.0
+# @update Jun-04-2015
+#
+
+class ROCker
+   #================================[ Class ]
+   #@@DEFAULTS.merge!({  })
+   
+   #================================[ Search ]
+   def search!
+      raise "-k/--rocker is mandatory." if @o[:rocker].nil?
+      raise "Code Under development..."
+      # ToDo
+      # [ ... ]
+   end # search!
+end # ROCker
+

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bio-rocker
 version: !ruby/object:Gem::Version
-  version: 0.2.5
+  version: 1.0.0
 platform: ruby
 authors:
 - Luis (Coto) Orellana
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-05-07 00:00:00.000000000 Z
+date: 2015-06-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rest-client
@@ -25,6 +25,20 @@ dependencies:
     - - ~>
       - !ruby/object:Gem::Version
         version: 1.7.3
+- !ruby/object:Gem::Dependency
+  name: json
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.8.1
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.8.1
 description: Detecting and quantifying functional genes in short-read metagenomic
   datasets
 email: lhorellana@gatech.edu
@@ -40,6 +54,11 @@ files:
 - lib/rocker/rocwindow.rb
 - lib/rocker/rocdata.rb
 - lib/rocker/rinterface.rb
+- lib/rocker/step/build.rb
+- lib/rocker/step/compile.rb
+- lib/rocker/step/search.rb
+- lib/rocker/step/filter.rb
+- lib/rocker/step/plot.rb
 - bin/ROCker
 homepage: http://enve-omics.ce.gatech.edu/rocker
 licenses:
@@ -51,9 +70,9 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ~>
     - !ruby/object:Gem::Version
-      version: '0'
+      version: '2.0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - '>='