wwood-bioruby 1.2.11

data/lib/bio/appl/mafft/report.rb

@@ -0,0 +1,226 @@
+#
+# = bio/appl/mafft/report.rb - MAFFT report class
+#
+# Copyright:: Copyright (C) 2003, 2007  Naohisa Goto <ng@bioruby.org>
+# License::   The Ruby License
+#
+#  $Id: report.rb,v 1.13 2007/07/16 12:21:39 ngoto Exp $
+#
+# MAFFT result parser class.
+# MAFFT is a very fast multiple sequence alignment software.
+#
+# Since a result of MAFFT is simply a multiple-fasta format,
+# the significance of this class is to keep standard form and
+# interface between Bio::ClustalW::Report.
+#
+# Bio::Alignment::MultiFastaFormat is a generic data class for
+# fasta-formatted multiple sequence alignment data.
+# Bio::MAFFT::Report inherits Bio::Alignment::MultiFastaFormat.
+#
+# == References
+#
+# * K. Katoh, K. Misawa, K. Kuma and T. Miyata.
+#   MAFFT: a novel method for rapid multiple sequence alignment based
+#   on fast Fourier transform. Nucleic Acids Res. 30: 3059-3066, 2002.
+#   http://nar.oupjournals.org/cgi/content/abstract/30/14/3059
+# * http://www.biophys.kyoto-u.ac.jp/~katoh/programs/align/mafft/
+#
+
+require 'stringio'
+require 'bio/db/fasta'
+require 'bio/io/flatfile'
+require 'bio/alignment'
+require 'bio/appl/mafft'
+
+module Bio
+  module Alignment
+    # Data class for fasta-formatted multiple sequence alignment data,
+    # which is simply multiple entiries of fasta formatted sequences.
+    class MultiFastaFormat
+
+      # delimiter for flatfile
+      DELIMITER = RS = nil
+
+      # Creates a new data object.
+      # +str+ should be a (multi-)fasta formatted string.
+      def initialize(str)
+        ff = Bio::FlatFile.new(Bio::FastaFormat, StringIO.new(str))
+        @data = ff.to_a
+        @alignment = nil
+        @seq_method = nil
+      end
+
+      # Gets an multiple alignment.
+      # Returns a Bio::Alignment object.
+      # +method+ should be one of :naseq, :aaseq, :seq, or nil (default).
+      # nil means to automatically determine nucleotide or amino acid.
+      #
+      # This method returns previously parsed object
+      # if the same method is given (or guessed method is the same).
+      def alignment(method = nil)
+        m = determine_seq_method(@data, method)
+        if !@alignment or m != @seq_method then
+          @seq_method = m
+          @alignment = do_parse(@data, @seq_method)
+        end
+        @alignment
+      end
+
+      # Gets an array of the fasta formatted sequence objects.
+      # Returns an array of Bio::FastaFormat objects.
+      def entries
+        @data
+      end
+
+      private
+      # determines seqtype.
+      # if nil is given, try to guess DNA or protein.
+      def determine_seq_method(data, m = nil)
+        case m
+        when :aaseq
+          :aaseq
+        when :naseq
+          :naseq
+        when :seq
+          :seq
+        when nil
+          # auto-detection
+          score = 0
+          data[0, 3].each do |e|
+            k = e.to_seq.guess
+            if k == Bio::Sequence::NA then
+              score += 1
+            elsif k == Bio::Sequence::AA then
+              score -= 1
+            end
+          end
+          if score > 0 then
+            :naseq
+          elsif score < 0 then
+            :aaseq
+          else
+            :seq
+          end
+        else
+          raise 'one of :naseq, :aaseq, :seq, or nil should be given'
+        end
+      end
+
+      # Parses a result.
+      def do_parse(ary, seqmethod)
+        a = Bio::Alignment.new
+        a.add_sequences(ary) do |x|
+          [ x.__send__(seqmethod), x.definition ]
+        end
+        a
+      end
+    end #class MultiFastaFormat
+  end #module Alignment
+
+  class MAFFT
+
+    # MAFFT result parser class.
+    # MAFFT is a very fast multiple sequence alignment software.
+    #
+    # Since a result of MAFFT is simply a multiple-fasta format,
+    # the significance of this class is to keep standard form and
+    # interface between Bio::ClustalW::Report.
+    class Report < Bio::Alignment::MultiFastaFormat
+
+      # Creates a new Report object.
+      # +str+ should be multi-fasta formatted text as a string.
+      #
+      # Compatibility Note: the old usage (to get array of Bio::FastaFormat
+      # objects) is deprecated.
+      #
+      # Compatibility Note 2: the argument +seqclass+ is deprecated.
+      #
+      # +seqclass+ should be one of following:
+      # Class:  Bio::Sequence::AA, Bio::Sequence::NA, ...
+      # String: 'PROTEIN', 'DNA', ...
+      #
+      def initialize(str, seqclass = nil)
+        if str.is_a?(Array) then
+          warn "Array of Bio::FastaFormat objects will be no longer accepted."
+          @data = str
+        else
+          super(str)
+        end
+
+        if seqclass then
+          warn "the 2nd argument (seqclass) will be no deprecated."
+          case seqclass
+          when /PROTEIN/i
+            @seqclass = Bio::Sequence::AA
+          when /[DR]NA/i
+            @seqclass = Bio::Sequence::NA
+          else
+            if seqclass.is_a?(Module) then
+              @seqclass = seqclass
+            else
+              @seqclass = nil
+            end
+          end
+        end
+      end
+
+      # sequence data. Returns an array of Bio::FastaFormat.
+      attr_reader :data
+
+      # Sequence class (Bio::Sequence::AA, Bio::Sequence::NA, ...)
+      #
+      # Compatibility note: This method will be removed in the tufure.
+      attr_reader :seqclass
+
+      # Gets an multiple alignment.
+      # Returns a Bio::Alignment object.
+      def alignment(method = nil)
+        super
+      end
+
+      # This method will be deprecated. Instead, please use alignment.
+      #
+      # Gets an multiple alignment.
+      # Returns a Bio::Alignment object.
+      def align
+        warn "Bio::MAFFT::Report#align is deprecated. Please use \'alignment\'."
+        alignment
+      end
+
+      # This will be deprecated. Instead, please use alignment.output_fasta.
+      #
+      # Gets an fasta-format string of the sequences.
+      # Returns a string.
+      # Same as align.to_fasta.
+      # Please refer to Bio::Alignment#output_fasta for arguments.
+      def to_fasta(*arg)
+        warn "Bio::MAFFT::report#to_fasta is deprecated. Please use \'alignment.output_fasta\'"
+        alignment.output_fasta(*arg)
+      end
+
+      # Compatibility note: Behavior of the method will be changed
+      # in the future.
+      #
+      # Gets an array of the sequences.
+      # Returns an array of Bio::FastaFormat instances.
+      def to_a
+        @data
+      end
+
+      private
+      # Parsing a result.
+      def do_parse(ary, seqmethod)
+        if @seqclass then
+          a = Bio::Alignment.new
+          a.add_sequences(ary) do |x|
+            [ @seqclass.new(x.seq), x.definition ]
+          end
+        else
+          super(ary, seqmethod)
+        end
+      end
+
+    end #class Report
+  end #class MAFFT
+end #module Bio
+

data/lib/bio/appl/muscle.rb

@@ -0,0 +1,52 @@
+#
+# = bio/appl/muscle.rb - MUSCLE application wrapper class
+#
+# Copyright:: Copyright (C) 2006-2007
+#             Jeffrey Blakeslee and John Conery University of Oregon <jeffb@uoregon.edu>
+#             Naohisa Goto <ng@bioruby.org>
+# License::   The Ruby License
+#
+#  $Id: muscle.rb,v 1.1 2007/07/16 12:25:50 ngoto Exp $
+#
+#
+# Bio::Muscle is a wrapper class to execute MUSCLE.
+#
+# == References
+#
+# * http://www.drive5.com/muscle/
+# * Edgar R.C.
+#   MUSCLE: multiple sequence alignment with high accuracy and
+#   high throughput. Nucleic Acids Res. 32: 1792-1797, 2004.
+# * Edgar, R.C.
+#   MUSCLE: a multiple sequence alignment method with reduced time
+#   and space complexity. BMC Bioinformatics 5: 113, 2004.
+#
+
+module Bio
+
+  # Bio::Muscle is a wrapper class to execute MUSCLE.
+  #
+  # Please refer documents in bio/apple/muscle.rb for references.
+  class Muscle < Bio::Alignment::FactoryTemplate::StdinInFileOut
+
+    # default program name
+    DEFAULT_PROGRAM = 'muscle'.freeze
+
+    # default report parser
+    DEFAULT_PARSER = Bio::Alignment::MultiFastaFormat
+
+    private
+    # generates options specifying input filename.
+    # returns an array of string
+    def _option_input_file(fn)
+      [ '-in', fn ]
+    end
+
+    # generates options specifying output filename.
+    # returns an array of string
+    def _option_output_file(fn)
+      [ '-out', fn ]
+    end
+  end #class Muscle
+
+end #module Bio

data/lib/bio/appl/paml/baseml.rb

@@ -0,0 +1,95 @@
+#
+# = bio/appl/paml/baseml.rb - Wrapper for running PAML program baseml
+#
+# Copyright::  Copyright (C) 2008
+#              Naohisa Goto <ng@bioruby.org>
+#
+# License::    The Ruby License
+#
+# == Description
+#
+# This file contains Bio::PAML::Baseml, a wrapper class running baseml.
+#
+# == References
+#
+# * http://abacus.gene.ucl.ac.uk/software/paml.html
+#
+
+require 'bio/appl/paml/common'
+
+module Bio::PAML
+
+  # == Description
+  #
+  # Bio::PAML::Baseml is a wrapper for running PAML baseml program. 
+  #
+  # Because most of the methods in this class are inherited from
+  # Bio::PAML::Common, see documents of Bio::PAML::Common for details.
+  #
+  # == Examples
+  #
+  # Example 1:
+  #
+  #   require 'bio'
+  #   # Reads multi-fasta formatted file and gets a Bio::Alignment object.
+  #   alignment = Bio::FlatFile.open(Bio::Alignment::MultiFastaFormat,
+  #                                  'example.fst').alignment
+  #   # Reads newick tree from a file
+  #   tree = Bio::FlatFile.open(Bio::Newick, 'example.tree').tree
+  #   # Creates a Baseml object
+  #   baseml = Bio::PAML::Baseml.new
+  #   # Sets parameters
+  #   baseml.parameters[:runmode] = 0
+  #   baseml.parameters[:RateAncestor] = 1
+  #   # You can also set many parameters at a time.
+  #   baseml.parameters.update({ :alpha => 0.5, :fix_alpha => 0 })
+  #   # Executes baseml with the alignment and the tree
+  #   report = baseml.query(alignment, tree)
+  #
+  class Baseml < Common
+
+    autoload  :Report,  'bio/appl/paml/baseml/report'
+
+    # Default program name
+    DEFAULT_PROGRAM = 'baseml'.freeze
+
+    # Default parameters when running baseml.
+    #
+    # The parameters whose values are different from the baseml defalut
+    # value (described in pamlDOC.pdf) in PAML 4.1 are:
+    #  seqfile, outfile, treefile, ndata, noisy, verbose
+    #
+    DEFAULT_PARAMETERS = {
+      # Essential argumemts
+      :seqfile             => nil,
+      :outfile             => nil,
+      # Optional arguments
+      :treefile            => nil,
+      :noisy               => 0,
+      :verbose             => 1,
+      :runmode             => 0,
+      :model               => 5,
+      :Mgene               => 0,
+      :ndata               => 1,
+      :clock               => 0,
+      :fix_kappa           => 0,
+      :kappa               => 2.5,
+      :fix_alpha           => 1,
+      :alpha               => 0.0,
+      :Malpha              => 0,
+      :ncatG               => 5,
+      :fix_rho             => 1,
+      :rho                 => 0.0,
+      :nparK               => 0,
+      :nhomo               => 0,
+      :getSE               => 0,
+      :RateAncestor        => 0,
+      :Small_Diff          => 1e-6,
+      :cleandata           => 1,
+      :fix_blength         => 0,
+      :method              => 0
+    }
+
+  end #class Baseml
+end #module Bio::PAML
+

data/lib/bio/appl/paml/baseml/report.rb

@@ -0,0 +1,32 @@
+#
+# = bio/appl/paml/baseml/report.rb - parser class for PAML program baseml
+#
+# Copyright::  Copyright (C) 2008
+#              Naohisa Goto <ng@bioruby.org>
+#
+# License::    The Ruby License
+#
+# == Description
+#
+# This file contains Bio::PAML::Baseml::Report, a parser class for a result
+# of baseml.
+#
+# == References
+#
+# * http://abacus.gene.ucl.ac.uk/software/paml.html
+#
+
+require 'bio/appl/paml/baseml'
+
+module Bio::PAML
+  class Baseml
+
+    # UNDER CONSTRUCTION.
+    #
+    # Bio::PAML::Baseml::Report is a parser class for a baseml result.
+    #
+    class Report < Bio::PAML::Common::Report
+    end #class Report
+
+  end #class Baseml
+end #module Bio::PAML

data/lib/bio/appl/paml/codeml.rb

@@ -0,0 +1,242 @@
+#
+# = bio/appl/paml/codeml.rb - Wrapper for running PAML program codeml
+#
+# Copyright::  Copyright (C) 2008
+#              Michael D. Barton <mail@michaelbarton.me.uk>,
+#              Naohisa Goto <ng@bioruby.org>
+#
+# License::    The Ruby License
+#
+# == Description
+#
+# This file contains a wrapper for running the CODEML tool for estimating evolutionary rate 
+#
+# == References
+#
+# * http://abacus.gene.ucl.ac.uk/software/paml.html
+#
+
+require 'bio/appl/paml/common'
+
+module Bio
+module PAML
+
+  # == Description
+  #
+  # Bio::PAML::Codeml is a wrapper for estimating evolutionary rate using the CODEML 
+  # tool. The class provides methods for generating the necessary configuration 
+  # file, and running codeml with the specified binary. Codeml output is 
+  # returned when codeml is run. Bio::PAML::Codeml::Report and Bio::PAML::Codeml::Rates
+  # provide simple classes for parsing and accessing the Codeml report and
+  # rates files respectively.
+  #
+  # == Examples
+  #
+  # Example 1:
+  #
+  #   require 'bio'
+  #   # Reads multi-fasta formatted file and gets a Bio::Alignment object.
+  #   alignment = Bio::FlatFile.open(Bio::Alignment::MultiFastaFormat,
+  #                                  'example.fst').alignment
+  #   # Reads newick tree from a file
+  #   tree = Bio::FlatFile.open(Bio::Newick, 'example.tree').tree
+  #   # Creates a Codeml object
+  #   codeml = Bio::PAML::Codeml.new
+  #   # Sets parameters
+  #   codeml.parameters[:runmode] = 0
+  #   codeml.parameters[:RateAncestor] = 1
+  #   # You can also set many parameters at a time.
+  #   codeml.parameters.update({ :alpha => 0.5, :fix_alpha => 0 })
+  #   # Executes codeml with the alignment and the tree
+  #   report = codeml.query(alignment, tree)
+  #       
+  # Example 2 (Obsolete usage):
+  #
+  #   # Create a control file, setting some Codeml options
+  #   # Default parameters are used otherwise, see RDoc for defaults
+  #   # The names of the parameters correspond to those specified
+  #   # in the Codeml documentation
+  #   control_file = Tempfile.new('codeml_ctl')
+  #   control_file.close(false)
+  #   # Prepare output file as a temporary file
+  #   output_file = Tempfile.new('codeml_test')
+  #   output_file.close(false)
+  #   Bio::PAML::Codeml.create_control_file(config_file.path, {
+  #     :model       => 1,
+  #     :fix_kappa   => 1,
+  #     :aaRatefile  => TEST_DATA + '/wag.dat',
+  #     :seqfile     => TEST_DATA + '/abglobin.aa',
+  #     :treefile    => TEST_DATA + '/abglobin.trees',
+  #     :outfile     => output_file.path,
+  #   })
+  #   
+  #   # Create an instance of Codeml specifying where the codeml binary is
+  #   codeml = Bio::PAML::Codeml.new('/path/to/codeml')
+  #   
+  #   # Run codeml using a control file
+  #   # Returns the command line output
+  #   codeml_output = codeml.run(control_file)
+  #
+  class Codeml < Common
+
+    autoload  :Report,  'bio/appl/paml/codeml/report'
+    autoload  :Rates,   'bio/appl/paml/codeml/rates'
+
+    # Default program name
+    DEFAULT_PROGRAM = 'codeml'.freeze
+
+    # Default parameters when running codeml.
+    #
+    # The parameters whose values are different from the codeml defalut
+    # value (described in pamlDOC.pdf) in PAML 4.1 are:
+    #  seqfile, outfile, treefile, ndata, noisy, verbose, cleandata
+    #
+    DEFAULT_PARAMETERS = {
+      # Essential argumemts
+      :seqfile             => nil,
+      :outfile             => nil,
+      # Optional arguments
+      :treefile            => nil,
+      :noisy               => 0,
+      :verbose             => 1,
+      :runmode             => 0,
+      :seqtype             => 2,
+      :CodonFreq           => 2,
+      :ndata               => 1,
+      :clock               => 0,
+      :aaDist              => 0,
+      :aaRatefile          => 'wag.dat',
+      :model               => 2,
+      :NSsites             => 0,
+      :icode               => 0,
+      :Mgene               => 0,
+      :fix_kappa           => 0,
+      :kappa               => 2,
+      :fix_omega           => 0,
+      :omega               => 0.4,
+      :fix_alpha           => 0,
+      :alpha               => 0.0,
+      :Malpha              => 0,
+      :ncatG               => 3,
+      :fix_rho             => 1,
+      :rho                 => 0.0,
+      :getSE               => 0,
+      :RateAncestor        => 0,
+      :Small_Diff          => 0.5e-6,
+      :cleandata           => 1,
+      :fix_blength         => 0,
+      :method              => 0
+    }
+
+    # OBSOLETE. This method should not be used. 
+    # Instead, use parameters.
+    def options
+      warn 'The method Codeml#options will be changed to be used for command line arguments in the future. Instead, use Codeml#parameters.'
+      parameters
+    end
+
+    # OBSOLETE. This method should not be used. 
+    # Instead, use parameters=(hash).
+    def options=(hash)
+      warn 'The method Codeml#options=() will be changed to be used for command line arguments in the future. Instead, use Codeml#parameters=().'
+      self.parameters=(hash)
+    end
+
+    # Obsolete. This method will be removed in the future.
+    # Helper method for creating a codeml control file.
+    # Note that default parameters are automatically merged.
+    def self.create_control_file(parameters, filename)
+      parameters = DEFAULT_PARAMETERS.merge(parameters)
+      File.open(filename, 'w') do |file|
+        parameters.each do |key, value|
+          file.puts "#{key.to_s} = #{value.to_s}" if value
+        end
+      end
+      filename
+    end
+
+    # OBSOLETE. This method will soon be removed.
+    # Instead, use create_control_file(parameters, filename).
+    def self.create_config_file(parameters, filename)
+      warn "The method Codeml.create_config_file(parameters, filename) will soon be removed. Instead, use Codeml.create_control_file(filename, parameters)."
+      create_control_file(parameters, filename)
+    end
+
+
+    # Runs the program on the internal parameters with the specified
+    # sequence alignment and tree.
+    #
+    # Note that parameters[:seqfile] and parameters[:outfile]
+    # are always modified, and parameters[:treefile] and
+    # parameters[:aaRatefile] are modified when tree and aarate are
+    # specified respectively.
+    #
+    # For other important information, see the document of
+    # Bio::PAML::Common#query.
+    #
+    # ---
+    # *Arguments*:
+    # * (required) _alignment_: Bio::Alignment object or similar object
+    # * (optional) _tree_: Bio::Tree object
+    # * (optional) _aarate_: String or nil
+    # *Returns*:: Report object
+    def query(alignment, tree = nil, aarate = nil)
+      begin
+        aaratefile = prepare_aaratefile(aarate)
+        ret = super(alignment, tree)
+      ensure
+        finalize_aaratefile(aaratefile)
+      end
+      ret
+    end
+
+    # Runs the program on the internal parameters with the specified
+    # sequence alignment data string and tree data string.
+    #
+    # Note that parameters[:outfile] is always modified, and
+    # parameters[:seqfile], parameters[:treefile], and
+    # parameters[:aaRatefile] are modified when
+    # alignment, tree, and aarate are specified respectively.
+    #
+    # It raises RuntimeError if seqfile is not specified in the argument
+    # or in the parameter.
+    #
+    # For other important information, see the document of query method.
+    # 
+    # ---
+    # *Arguments*:
+    # * (optional) _alignment_: String
+    # * (optional) _tree_: String or nil
+    # * (optional) _aarate_: String or nil
+    # *Returns*:: contents of output file (String)
+    def query_by_string(alignment = nil, tree = nil, aarate = nil)
+      begin
+        aaratefile = prepare_aaratefile(aarate)
+        ret = super(alignment, tree)
+      ensure
+        finalize_aaratefile(aaratefile)
+      end
+      ret
+    end
+
+    private
+
+    # (private) prepares temporary file for aaRatefile if needed
+    def prepare_aaratefile(aarate)
+      if aarate then
+        aaratefile = Tempfile.new('codeml_aarate')
+        aaratefile.print aarate
+        aaratefile.close(false)
+        self.parameters[:aaRatefile] = aaratefile.path
+      end
+      aaratefile
+    end
+
+    # (private) removes temporary file for aaRatefile if needed
+    def finalize_aaratefile(aaratefile)
+      aaratefile.close(true) if aaratefile
+    end
+
+  end # End class Codeml
+end # End module PAML
+end # End module Bio