genevalidator 1.6.12 → 2.1.3

data/lib/genevalidator/arg_validation.rb

@@ -16,7 +16,6 @@ module GeneValidator
 
       def validate_args
         @opt = opt
-        assert_output_dir_does_not_exist
         assert_file_present('input file', opt[:input_fasta_file])
         assert_input_file_probably_fasta
         assert_input_sequence
@@ -38,7 +37,7 @@ module GeneValidator
       private
 
       def assert_validations_arg
-        validations = %w(lenc lenr frame merge dup orf align)
+        validations = %w[lenc lenr frame merge dup orf align]
         if @opt[:validations]
           val = @opt[:validations].collect { |v| v.strip.downcase }
           validations = val unless val.include? 'all'
@@ -48,13 +47,13 @@ module GeneValidator
 
       def check_num_threads
         @opt[:num_threads] = Integer(@opt[:num_threads])
-        unless @opt[:num_threads] > 0
-          $stderr.puts 'Number of threads can not be lower than 0'
-          $stderr.puts 'Setting number of threads to 1'
+        unless @opt[:num_threads].positive?
+          warn 'Number of threads can not be lower than 0'
+          warn 'Setting number of threads to 1'
           @opt[:num_threads] = 1
         end
         return unless @opt[:num_threads] > 256
-        $stderr.puts "Number of threads set at #{@opt[:num_threads]} is" \
+        warn "Number of threads set at #{@opt[:num_threads]} is" \
                      ' unusually high.'
       end
 
@@ -68,45 +67,35 @@ module GeneValidator
         end
       end
 
-      def assert_output_dir_does_not_exist
-        output_dir = "#{@opt[:input_fasta_file]}.html"
-        return unless File.exist?(output_dir)
-        $stderr.puts 'The output directory already exists for this fasta file.'
-        $stderr.puts "\nPlease remove the following directory: #{output_dir}\n"
-        $stderr.puts "You can run the following command to remove the folder.\n"
-        $stderr.puts "\n   $ rm -r #{output_dir} \n"
-        exit 1
-      end
-
       def assert_tabular_options_exists
         return if @opt[:blast_tabular_options]
-        $stderr.puts '*** Error: BLAST tabular options (-o) have not been set.'
-        $stderr.puts '    Please set the "-o" option with the custom format'
-        $stderr.puts '    used in the BLAST -outfmt argument'
+        warn '*** Error: BLAST tabular options (-o) have not been set.'
+        warn '    Please set the "-o" option with the custom format'
+        warn '    used in the BLAST -outfmt argument'
         exit 1
       end
 
       def assert_input_file_probably_fasta
         File.open(@opt[:input_fasta_file], 'r') do |file_stream|
-          (file_stream.readline[0] == '>') ? true : false
+          file_stream.readline[0] == '>'
         end
       end
 
       def assert_file_present(desc, file, exit_code = 1)
         return if file && File.exist?(File.expand_path(file))
-        $stderr.puts "*** Error: Couldn't find the #{desc}: #{file}."
+        warn "*** Error: Couldn't find the #{desc}: #{file}."
         exit exit_code
       end
 
-      alias_method :assert_dir_present, :assert_file_present
+      alias assert_dir_present assert_file_present
 
       def assert_input_sequence
         fasta_content = IO.binread(@opt[:input_fasta_file])
         type = BlastUtils.type_of_sequences(fasta_content)
-        return if type == :nucleotide || type == :protein
-        $stderr.puts '*** Error: The input files does not contain just protein'
-        $stderr.puts '    or nucleotide data.'
-        $stderr.puts '    Please correct this and try again.'
+        return if %i[nucleotide protein].include? type
+        warn '*** Error: The input files does not contain just protein'
+        warn '    or nucleotide data.'
+        warn '    Please correct this and try again.'
         exit 1
       end
 
@@ -116,8 +105,8 @@ module GeneValidator
           if File.exist?(bin) && File.directory?(bin)
             add_to_path(bin)
           else
-            $stderr.puts '*** The following bin directory does not exist:'
-            $stderr.puts "    #{bin}"
+            warn '*** The following bin directory does not exist:'
+            warn "    #{bin}"
           end
         end
       end
@@ -131,10 +120,10 @@ module GeneValidator
 
       def assert_mafft_installation
         return if command?('mafft')
-        $stderr.puts '*** Could not find Mafft binaries.'
-        $stderr.puts '    Ignoring error and continuing - Please note that' \
+        warn '*** Could not find Mafft binaries.'
+        warn '    Ignoring error and continuing - Please note that' \
                      ' some validations may be skipped.'
-        $stderr.puts # a blank line
+        warn # a blank line
       end
     end
 
@@ -142,7 +131,7 @@ module GeneValidator
     class Blast
       class << self
         # Use a fixed minimum version of BLAST+
-        MINIMUM_BLAST_VERSION           = '2.2.30+'
+        MINIMUM_BLAST_VERSION           = '2.2.30+'.freeze
         # Use the following exit codes, or 1.
         EXIT_BLAST_NOT_INSTALLED        = 2
         EXIT_BLAST_NOT_COMPATIBLE       = 3
@@ -150,7 +139,6 @@ module GeneValidator
 
         def validate(opt)
           assert_blast_installation
-          warn_if_remote_database(opt)
           assert_local_blast_database_exists(opt[:db]) if opt[:db] !~ /remote/
         end
 
@@ -160,27 +148,10 @@ module GeneValidator
           assert_blast_compatible
         end
 
-        def warn_if_remote_database(opt)
-          return if opt[:db] !~ /remote/
-          $stderr.puts # a blank line
-          if !opt[:raw_sequences] &&
-              (opt[:validations].include?('align') ||
-               opt[:validations].include?('dup'))
-            $stderr.puts 'Warning: Hit sequences will be fetched from remote' \
-                         ' server.'
-          else
-            $stderr.puts 'Warning: BLAST will be carried out on remote server.'
-          end
-          $stderr.puts 'This may take quite a bit of time.'
-          $stderr.puts 'You may want to install a local BLAST database for' \
-                       ' faster analyses.'
-          $stderr.puts # a blank line
-        end
-
         def assert_local_blast_database_exists(db)
           return if system("blastdbcmd -db #{db} -info > /dev/null 2>&1")
-          $stderr.puts '*** No BLAST database found at the provided path.'
-          $stderr.puts '    Please ensure that the provided path is correct' \
+          warn '*** No BLAST database found at the provided path.'
+          warn '    Please ensure that the provided path is correct' \
                        ' and then try again.'
           exit EXIT_NO_BLAST_DATABASE
         end
@@ -189,15 +160,15 @@ module GeneValidator
 
         def assert_blast_installed
           return if GVArgValidation.command?('blastdbcmd')
-          $stderr.puts '*** Could not find BLAST+ binaries.'
+          warn '*** Could not find BLAST+ binaries.'
           exit EXIT_BLAST_NOT_INSTALLED
         end
 
         def assert_blast_compatible
           version = `blastdbcmd -version`.split[1]
           return if version >= MINIMUM_BLAST_VERSION
-          $stderr.puts "*** Your BLAST+ version #{version} is outdated."
-          $stderr.puts '    GeneValidator needs NCBI BLAST+ version' \
+          warn "*** Your BLAST+ version #{version} is outdated."
+          warn '    GeneValidator needs NCBI BLAST+ version' \
                        " #{MINIMUM_BLAST_VERSION} or higher."
           exit EXIT_BLAST_NOT_COMPATIBLE
         end

data/lib/genevalidator/blast.rb

@@ -12,34 +12,10 @@ module GeneValidator
   class BlastUtils
     class << self
       extend Forwardable
-      def_delegators GeneValidator, :opt, :config
+      def_delegators GeneValidator, :opt, :config, :dirs
 
       EVALUE = 1e-5
 
-      ##
-      # Calls blast from standard input with specific parameters
-      # Params:
-      # +blast_type+: blast command in String format (e.g 'blast(x/p)')
-      # +query+: String containing the the query in fasta format
-      # +db+: database
-      # +num_threads+: The number of threads to run BLAST with.
-      # Output:
-      # String with the blast xml output
-      def run_blast(query, db = opt[:db], seq_type = config[:type],
-                    num_threads = opt[:num_threads])
-
-        blast_type = (seq_type == :protein) ? 'blastp' : 'blastx'
-        # -num_threads is not supported on remote databases
-        threads = (db !~ /remote/) ? "-num_threads #{num_threads}" : ''
-
-        blastcmd = "#{blast_type} -db '#{db}' -evalue #{EVALUE} -outfmt 5" \
-                   " #{threads}"
-
-        cmd = "echo \"#{query}\" | #{blastcmd}"
-        `#{cmd} >/dev/null 2>&1`
-      end
-
-      ##
       # Runs BLAST on an input file
       # Params:
       # +blast_type+: blast command in String format (e.g 'blastx' or 'blastp')
@@ -49,32 +25,22 @@ module GeneValidator
       # +nr_hits+: max number of hits
       # Output:
       # XML file
-      def run_blast_on_input_file(input_file = opt[:input_fasta_file],
-                                  db = opt[:db], seq_type = config[:type],
-                                  num_threads = opt[:num_threads])
-        return if opt[:blast_xml_file] || opt[:blast_tabular_file]
+      def run_blast_on_input_file
+        remote = opt[:db].match?(/remote/) ? true : false
+        print_blast_info_text(remote)
 
-        $stderr.puts 'Running BLAST. This may take a while.'
-        opt[:blast_xml_file] = input_file + '.blast_xml'
+        log_file = File.join(dirs[:tmp_dir], 'blast_cmd_output.txt')
+        `#{blast_cmd(opt, config, remote)} > #{log_file} 2>&1`
 
-        blast_type = (seq_type == :protein) ? 'blastp' : 'blastx'
-        # -num_threads is not supported on remote databases
-        threads = (opt[:db] !~ /remote/) ? "-num_threads #{num_threads}" : ''
-
-        blastcmd = "#{blast_type} -query '#{input_file}'" \
-                   " -out '#{opt[:blast_xml_file]}' -db #{db} " \
-                   " -evalue #{EVALUE} -outfmt 5 #{threads}"
-
-        `#{blastcmd} >/dev/null 2>&1`
         return unless File.zero?(opt[:blast_xml_file])
-        $stderr.puts 'Blast failed to run on the input file.'
-        if opt[:db] !~ /remote/
-          $stderr.puts 'Please ensure that the BLAST database exists and try'
-          $stderr.puts 'again.'
+        warn 'Blast failed to run on the input file.'
+        if remote
+          warn 'You are using BLAST with a remote database. Please'
+          warn 'ensure that you have internet access and try again.'
         else
-          $stderr.puts 'You are using BLAST with a remote database. Please'
-          $stderr.puts 'ensure that you have internet access and try again.'
+          warn 'Please ensure that the BLAST database exists and try again.'
         end
+        exit 1
       end
 
       ##
@@ -84,66 +50,24 @@ module GeneValidator
       # +type+: the type of the sequence: :nucleotide or :protein
       # Outputs:
       # Array of +Sequence+ objects corresponding to the list of hits
-      def parse_next(iterator, type = config[:type])
-        hits = []
+      def parse_next(iterator)
         iter = iterator.next
 
         # parse blast the xml output and get the hits
         # hits obtained are proteins! (we use only blastp and blastx)
+        hits = []
         iter.each do |hit|
-          seq = Query.new
-
+          seq                = Query.new
           seq.length_protein = hit.len.to_i
           seq.type           = :protein
           seq.identifier     = hit.hit_id
           seq.definition     = hit.hit_def
-          seq.accession_no = hit.accession
-
-          # get all high-scoring segment pairs (hsp)
-          hsps = []
-
-          hit.hsps.each do |hsp|
-            current_hsp            = Hsp.new
-            current_hsp.hsp_evalue = format('%.0e', hsp.evalue)
+          seq.accession_no   = hit.accession
+          seq.hsp_list       = hit.hsps.map { |hsp| Hsp.new(xml_input: hsp) }
 
-            current_hsp.hit_from         = hsp.hit_from.to_i
-            current_hsp.hit_to           = hsp.hit_to.to_i
-            current_hsp.match_query_from = hsp.query_from.to_i
-            current_hsp.match_query_to   = hsp.query_to.to_i
-
-            if type == :nucleotide
-              current_hsp.match_query_from /= 3
-              current_hsp.match_query_to /= 3
-              current_hsp.match_query_from += 1
-              current_hsp.match_query_to += 1
-            end
-
-            current_hsp.query_reading_frame = hsp.query_frame.to_i
-
-            current_hsp.hit_alignment = hsp.hseq.to_s
-            seq_type = guess_sequence_type(current_hsp.hit_alignment)
-            fail SequenceTypeError unless seq_type == :protein || seq_type.nil?
-
-            current_hsp.query_alignment = hsp.qseq.to_s
-            seq_type = guess_sequence_type(current_hsp.query_alignment)
-            fail SequenceTypeError unless seq_type == :protein || seq_type.nil?
-
-            current_hsp.align_len = hsp.align_len.to_i
-            current_hsp.identity  = hsp.identity.to_i
-            current_hsp.pidentity = (100 * hsp.identity / hsp.align_len.to_f)
-              .round(2)
-
-            hsps.push(current_hsp)
-          end
-
-          seq.hsp_list = hsps
-          hits.push(seq)
+          hits << seq
         end
-
         hits
-      rescue SequenceTypeError => e
-        $stderr.puts e
-        exit 1
       rescue StopIteration
         nil
       end
@@ -164,7 +88,7 @@ module GeneValidator
         sequences = fasta_format_string.split(/^>.*$/).delete_if(&:empty?)
         # get all sequence types
         sequence_types = sequences.collect { |seq| guess_sequence_type(seq) }
-                         .uniq.compact
+                                  .uniq.compact
 
         return nil if sequence_types.empty?
         sequence_types.first if sequence_types.length == 1
@@ -184,7 +108,7 @@ module GeneValidator
         return nil if cleaned_sequence.length < 10 # conservative
 
         type = Bio::Sequence.new(cleaned_sequence).guess(0.9)
-        (type == Bio::Sequence::NA) ? :nucleotide : :protein
+        type == Bio::Sequence::NA ? :nucleotide : :protein
       end
 
       ##
@@ -192,11 +116,32 @@ module GeneValidator
       def guess_sequence_type_from_input_file(file = opt[:input_fasta_file])
         lines = File.foreach(file).first(10)
         seqs = ''
-        lines.each do |l|
-          seqs += l.chomp unless l[0] == '>'
-        end
+        lines.each { |l| seqs += l.chomp unless l[0] == '>' }
         guess_sequence_type(seqs)
       end
+
+      private
+
+      def blast_cmd(opt, config, remote)
+        blast_type = config[:type] == :protein ? 'blastp' : 'blastx'
+        # -num_threads is not supported on remote databases
+        threads = remote ? '' : "-num_threads #{opt[:num_threads]}"
+
+        "#{blast_type} -query '#{opt[:input_fasta_file]}'" \
+        " -db #{opt[:db]} -outfmt 5 -evalue #{EVALUE} #{threads}" \
+        " -out '#{opt[:blast_xml_file]}' #{opt[:blast_options]}"
+      end
+
+      def print_blast_info_text(remote)
+        warn '' # a blank line
+        if remote
+          warn '==> BLAST search and subsequent analysis will be done on a remote'
+          warn '    database. Please use a local database for larger analysis.'
+        else 
+          warn '==> Running BLAST. This may take a while.'
+        end
+        warn '' # a blank line
+      end
     end
   end
 end

data/lib/genevalidator/clusterization.rb

@@ -13,7 +13,7 @@ module GeneValidator
     end
 
     def print
-      $stderr.puts "Cluster: #{x} #{y}"
+      warn "Cluster: #{x} #{y}"
     end
 
     ##
@@ -44,7 +44,7 @@ module GeneValidator
     # Overload quality operator
     # Returns true if the pairs are equal, false otherwise
     def ==(other)
-      (other.x == x && other.y == y) ? true : false
+      other.x == x && other.y == y ? true : false
     end
 
     def eql?(other)
@@ -66,7 +66,7 @@ module GeneValidator
 
     def print
       objects.each do |elem|
-        $stderr.puts "(#{elem[0].x},#{elem[0].y}): #{elem[1]}"
+        warn "(#{elem[0].x},#{elem[0].y}): #{elem[1]}"
       end
     end
 
@@ -254,7 +254,7 @@ module GeneValidator
     # Real number
     def deviation(clusters, queryLength)
       hits = clusters.map { |c| c.lengths.map { |x| Array.new(x[1], x[0]) }.flatten }.flatten
-      raw_hits = clusters.map { |c| c.lengths.map { |x| Array.new(x[1], x[0]) }.flatten }.flatten.to_s.gsub('[', '').gsub(']', '')
+      raw_hits = clusters.map { |c| c.lengths.map { |x| Array.new(x[1], x[0]) }.flatten }.flatten.to_s.delete('[').delete(']')
       R.eval("sd = sd(c(#{raw_hits}))")
       sd = R.pull('sd')
       sd = standard_deviation(hits)
@@ -273,11 +273,11 @@ module GeneValidator
     ##
     # Prints the current cluster
     def print
-      $stderr.puts "Cluster: mean = #{mean}, density = #{density}"
+      warn "Cluster: mean = #{mean}, density = #{density}"
       lengths.sort { |a, b| a <=> b }.each do |elem|
-        $stderr.puts "#{elem[0]}, #{elem[1]}"
+        warn "#{elem[0]}, #{elem[1]}"
       end
-      $stderr.puts '--------------------------'
+      warn '--------------------------'
     end
 
     ##
@@ -337,7 +337,7 @@ module GeneValidator
       # clusters = array of clusters
       # initially each length belongs to a different cluster
       histogram.each do |e|
-        $stderr.puts "pair (#{e[0].x} #{e[0].y}) appears #{e[1]} times" if debug
+        warn "pair (#{e[0].x} #{e[0].y}) appears #{e[1]} times" if debug
         hash = { e[0] => e[1] }
         cluster = PairCluster.new(hash)
         clusters.push(cluster)
@@ -355,7 +355,7 @@ module GeneValidator
         break if no_clusters != 0 && clusters.length == no_clusters
 
         iteration += iteration
-        $stderr.puts "\nIteration #{iteration}" if debug
+        warn "\nIteration #{iteration}" if debug
 
         min_distance = 100_000_000
         cluster1     = 0
@@ -365,9 +365,7 @@ module GeneValidator
         [*(0..(clusters.length - 2))].each do |i|
           [*((i + 1)..(clusters.length - 1))].each do |j|
             dist = clusters[i].distance(clusters[j], distance_method)
-            if debug
-              $stderr.puts "distance between clusters #{i} and #{j} is #{dist}"
-            end
+            warn "distance between clusters #{i} and #{j} is #{dist}" if debug
             current_density = clusters[i].density + clusters[j].density
             if dist < min_distance
               min_distance = dist
@@ -383,14 +381,14 @@ module GeneValidator
         end
 
         # merge clusters 'cluster1' and 'cluster2'
-        $stderr.puts "clusters to merge #{cluster1} and #{cluster2}" if debug
+        warn "clusters to merge #{cluster1} and #{cluster2}" if debug
 
         clusters[cluster1].add(clusters[cluster2])
         clusters.delete_at(cluster2)
 
         if debug
           clusters.each_with_index do |elem, i|
-            $stderr.puts "cluster #{i}"
+            warn "cluster #{i}"
             elem.print
           end
         end
@@ -437,8 +435,8 @@ module GeneValidator
 
       # clusters = array of clusters
       # initially each length belongs to a different cluster
-      histogram.sort { |a, b| a[0] <=> b[0] }.each do |elem|
-        $stderr.puts "len #{elem[0]} appears #{elem[1]} times" if debug
+      histogram.sort_by { |a| a[0] }.each do |elem|
+        warn "len #{elem[0]} appears #{elem[1]} times" if debug
         hash = { elem[0] => elem[1] }
         cluster = Cluster.new(hash)
         clusters.push(cluster)
@@ -456,7 +454,7 @@ module GeneValidator
         break if no_clusters != 0 && clusters.length == no_clusters
 
         iteration += iteration
-        $stderr.puts "\nIteration #{iteration}" if debug
+        warn "\nIteration #{iteration}" if debug
 
         min_distance = 100_000_000
         cluster      = 0
@@ -464,9 +462,7 @@ module GeneValidator
 
         clusters[0..clusters.length - 2].each_with_index do |_item, i|
           dist = clusters[i].distance(clusters[i + 1], distance_method)
-          if debug
-            $stderr.puts "distance btwn clusters #{i} and #{i + 1} is #{dist}"
-          end
+          warn "distance btwn clusters #{i} and #{i + 1} is #{dist}" if debug
           current_density = clusters[i].density + clusters[i + 1].density
           if dist < min_distance
             min_distance = dist
@@ -485,14 +481,14 @@ module GeneValidator
         end
 
         # merge clusters 'cluster' and 'cluster'+1
-        $stderr.puts "clusters to merge #{cluster} and #{cluster + 1}" if debug
+        warn "clusters to merge #{cluster} and #{cluster + 1}" if debug
 
         clusters[cluster].add(clusters[cluster + 1])
         clusters.delete_at(cluster + 1)
 
         if debug
           clusters.each_with_index do |elem, i|
-            $stderr.puts "cluster #{i}"
+            warn "cluster #{i}"
             elem.print
           end
         end