wwood-bioruby 1.2.11

data/lib/bio/db/fasta/fasta_to_biosequence.rb

@@ -0,0 +1,63 @@
+#
+# = bio/db/fasta/fasta_to_biosequence.rb - Bio::FastaFormat to Bio::Sequence adapter module
+#
+# Copyright::   Copyright (C) 2008
+#               Naohisa Goto <ng@bioruby.org>,
+# License::     The Ruby License
+#
+# $Id:$
+#
+
+require 'bio/sequence'
+require 'bio/sequence/adapter'
+
+# Internal use only. Normal users should not use this module.
+#
+# Bio::FastaFormat to Bio::Sequence adapter module.
+# It is internally used in Bio::FastaFormat#to_biosequence.
+#
+module Bio::Sequence::Adapter::FastaFormat
+
+  extend Bio::Sequence::Adapter
+
+  private
+
+  def_biosequence_adapter :seq
+
+  # primary accession
+  def_biosequence_adapter :primary_accession do |orig|
+    orig.identifiers.accessions.first or orig.identifiers.entry_id
+  end
+
+  # secondary accessions
+  def_biosequence_adapter :secondary_accessions do |orig|
+    orig.identifiers.accessions[1..-1]
+  end
+
+  # entry_id
+  def_biosequence_adapter :entry_id do |orig|
+    orig.identifiers.locus or orig.identifiers.accessions.first or
+      orig.identifiers.entry_id
+  end
+
+  # NCBI GI is stored on other_seqids
+  def_biosequence_adapter :other_seqids do |orig|
+    other = []
+    if orig.identifiers.gi then
+      other.push Bio::Sequence::DBLink.new('GI', orig.identifiers.gi)
+    end
+    other.empty? ? nil : other
+  end
+
+  # definition
+  def_biosequence_adapter :definition do |orig|
+    if orig.identifiers.accessions.empty? and
+        !(orig.identifiers.gi) then
+      orig.definition
+    else
+      orig.identifiers.description
+    end
+  end
+
+end #module Bio::Sequence::Adapter::FastaFormat
+

data/lib/bio/db/fasta/format_fasta.rb

@@ -0,0 +1,97 @@
+#
+# = bio/db/fasta/format_fasta.rb - Fasta format generater
+#
+# Copyright::   Copyright (C) 2006-2008
+#               Toshiaki Katayama <k@bioruby.org>,
+#               Naohisa Goto <ng@bioruby.org>,
+#               Jan Aerts <jan.aerts@bbsrc.ac.uk>
+# License::    The Ruby License
+#
+# $Id: format_fasta.rb,v 1.1.2.1 2008/03/04 11:26:59 ngoto Exp $
+#
+
+require 'bio/sequence/format'
+
+module Bio::Sequence::Format::Formatter
+
+  # INTERNAL USE ONLY, YOU SHOULD NOT USE THIS CLASS.
+  # Simple Fasta format output class for Bio::Sequence.
+  class Fasta < Bio::Sequence::Format::FormatterBase
+
+    # INTERNAL USE ONLY, YOU SHOULD NOT CALL THIS METHOD.
+    #
+    # Creates a new Fasta format generater object from the sequence.
+    #
+    # ---
+    # *Arguments*:
+    # * _sequence_: Bio::Sequence object
+    # * (optional) :header => _header_: String (default nil)
+    # * (optional) :width => _width_: Fixnum (default 70)
+    def initialize; end if false # dummy for RDoc
+
+    # INTERNAL USE ONLY, YOU SHOULD NOT CALL THIS METHOD.
+    #
+    # Output the FASTA format string of the sequence.  
+    #
+    # Currently, this method is used in Bio::Sequence#output like so,
+    #
+    #   s = Bio::Sequence.new('atgc')
+    #   puts s.output(:fasta)                   #=> "> \natgc\n"
+    # ---
+    # *Returns*:: String object
+    def output
+      header = @options[:header]
+      width = @options.has_key?(:width) ? @options[:width] : 70
+      seq = @sequence.seq
+      entry_id = @sequence.entry_id || 
+        "#{@sequence.primary_accession}.#{@sequence.sequence_version}"
+      definition = @sequence.definition
+      header ||= "#{entry_id} #{definition}"
+
+      ">#{header}\n" +
+        if width
+          seq.to_s.gsub(Regexp.new(".{1,#{width}}"), "\\0\n")
+        else
+          seq.to_s + "\n"
+        end
+    end
+  end #class Fasta
+
+  # INTERNAL USE ONLY, YOU SHOULD NOT USE THIS CLASS.
+  # NCBI-Style Fasta format output class for Bio::Sequence.
+  # (like "ncbi" format in EMBOSS)
+  #
+  # Note that this class is under construction.
+  class Fasta_ncbi < Bio::Sequence::Format::FormatterBase
+
+    # INTERNAL USE ONLY, YOU SHOULD NOT CALL THIS METHOD.
+    #
+    # Output the FASTA format string of the sequence.  
+    #
+    # Currently, this method is used in Bio::Sequence#output like so,
+    #
+    #   s = Bio::Sequence.new('atgc')
+    #   puts s.output(:ncbi)                   #=> "> \natgc\n"
+    # ---
+    # *Returns*:: String object
+    def output
+      width = 70
+      seq = @sequence.seq
+      #gi = @sequence.gi_number
+      dbname = 'lcl'
+      if @sequence.primary_accession.to_s.empty? then
+        idstr = @sequence.entry_id
+      else
+        idstr = "#{@sequence.primary_accession}.#{@sequence.sequence_version}"
+      end
+
+      definition = @sequence.definition
+      header = "#{dbname}|#{idstr} #{definition}"
+
+      ">#{header}\n" + seq.to_s.gsub(Regexp.new(".{1,#{width}}"), "\\0\n")
+    end
+  end #class Ncbi
+
+end #module Bio::Sequence::Format::Formatter
+
+

data/lib/bio/db/genbank/common.rb

@@ -0,0 +1,307 @@
+#
+# = bio/db/genbank/common.rb - Common methods for GenBank style database classes
+#
+# Copyright::  Copyright (C) 2004 Toshiaki Katayama <k@bioruby.org>
+# License::    The Ruby License
+#
+# $Id: common.rb,v 1.11.2.5 2008/06/17 15:53:21 ngoto Exp $
+#
+
+require 'bio/db'
+
+module Bio
+class NCBIDB
+
+# == Description
+# 
+# This module defines a common framework among GenBank, GenPept, RefSeq, and
+# DDBJ.  For more details, see the documentations in each genbank/*.rb files.
+# 
+# == References
+# 
+# * ftp://ftp.ncbi.nih.gov/genbank/gbrel.txt
+# * http://www.ncbi.nlm.nih.gov/collab/FT/index.html
+#
+module Common
+
+  DELIMITER	= RS = "\n//\n"
+  TAGSIZE	= 12
+
+  def initialize(entry)
+    super(entry, TAGSIZE)
+  end
+
+  # LOCUS -- Locus class must be defined in child classes.
+  def locus
+    # must be overrided in each subclass
+  end
+
+  # DEFINITION -- Returns contents of the DEFINITION record as a String.
+  def definition
+    field_fetch('DEFINITION')
+  end
+
+
+  # ACCESSION -- Returns contents of the ACCESSION record as an Array.
+  def accessions
+    field_fetch('ACCESSION').strip.split(/\s+/)
+  end
+
+
+  # VERSION -- Returns contents of the VERSION record as an Array of Strings.
+  def versions
+    @data['VERSION'] ||= fetch('VERSION').split(/\s+/)
+  end
+
+  # Returns the first part of the VERSION record as "ACCESSION.VERSION" String.
+  def acc_version
+    versions.first.to_s
+  end
+
+  # Returns the ACCESSION part of the acc_version.
+  def accession
+    acc_version.split(/\./).first.to_s
+  end
+
+  # Returns the VERSION part of the acc_version as a Fixnum
+  def version
+    acc_version.split(/\./).last.to_i
+  end
+
+  # Returns the second part of the VERSION record as a "GI:#######" String.
+  def gi
+    versions.last
+  end
+
+
+  # NID -- Returns contents of the NID record as a String.
+  def nid
+    field_fetch('NID')
+  end
+
+
+  # KEYWORDS -- Returns contents of the KEYWORDS record as an Array of Strings.
+  def keywords
+    @data['KEYWORDS'] ||= fetch('KEYWORDS').chomp('.').split(/; /)
+  end
+
+
+  # SEGMENT -- Returns contents of the SEGMENT record as a "m/n" form String.
+  def segment
+    @data['SEGMENT'] ||= fetch('SEGMENT').scan(/\d+/).join("/")
+  end
+
+
+  # SOURCE -- Returns contents of the SOURCE record as a Hash.
+  def source
+    unless @data['SOURCE']
+      name, org = get('SOURCE').split('ORGANISM')
+      org ||= ""
+      if org[/\S+;/]
+        organism = $`
+        taxonomy = $& + $'
+      elsif org[/\S+\./]				# rs:NC_001741
+        organism = $`
+        taxonomy = $& + $'
+      else
+        organism = org
+        taxonomy = ''
+      end
+      @data['SOURCE'] = {
+        'common_name'	=> truncate(tag_cut(name)),
+        'organism'	=> truncate(organism),
+        'taxonomy'	=> truncate(taxonomy),
+      }
+      @data['SOURCE'].default = ''
+    end
+    @data['SOURCE']
+  end
+
+  def common_name
+    source['common_name']
+  end
+  alias vernacular_name common_name
+
+  def organism
+    source['organism']
+  end
+
+  def taxonomy
+    source['taxonomy']
+  end
+
+
+  # REFERENCE -- Returns contents of the REFERENCE records as an Array of
+  # Bio::Reference objects.
+  def references
+    unless @data['REFERENCE']
+      ary = []
+      toptag2array(get('REFERENCE')).each do |ref|
+        hash = Hash.new
+        subtag2array(ref).each do |field|
+          case tag_get(field)
+          when /REFERENCE/
+            if /(\d+)(\s*\((.+)\))?/m =~ tag_cut(field) then
+              hash['embl_gb_record_number'] = $1.to_i
+              if $3 and $3 != 'sites' then
+                seqpos = $3
+                seqpos.sub!(/\A\s*bases\s+/, '')
+                seqpos.gsub!(/(\d+)\s+to\s+(\d+)/, "\\1-\\2")
+                seqpos.gsub!(/\s*\;\s*/, ', ')
+                hash['sequence_position'] = seqpos
+              end
+            end
+          when /AUTHORS/
+            authors = truncate(tag_cut(field))
+            authors = authors.split(/, /)
+            authors[-1] = authors[-1].split(/\s+and\s+/) if authors[-1]
+            authors = authors.flatten.map { |a| a.sub(/,/, ', ') }
+            hash['authors']	= authors
+          when /TITLE/
+            hash['title']	= truncate(tag_cut(field)) + '.'
+          when /JOURNAL/
+            journal = truncate(tag_cut(field))
+            if journal =~ /(.*) (\d+) \((\d+)\), (\d+-\d+) \((\d+)\)$/
+      	hash['journal']	= $1
+      	hash['volume']	= $2
+      	hash['issue']	= $3
+      	hash['pages']	= $4
+      	hash['year']	= $5
+            else
+      	hash['journal'] = journal
+            end
+          when /MEDLINE/
+            hash['medline']	= truncate(tag_cut(field))
+          when /PUBMED/
+            hash['pubmed']	= truncate(tag_cut(field))
+          when /REMARK/
+            hash['comments'] ||= []
+            hash['comments'].push truncate(tag_cut(field))
+          end
+        end
+        ary.push(Reference.new(hash))
+      end
+      @data['REFERENCE'] = ary.extend(Bio::References::BackwardCompatibility)
+    end
+    if block_given?
+      @data['REFERENCE'].each do |r|
+        yield r
+      end
+    else
+      @data['REFERENCE']
+    end
+  end
+
+
+  # COMMENT -- Returns contents of the COMMENT record as a String.
+  def comment
+    str = get('COMMENT').to_s.sub(/\ACOMMENT     /, '')
+    str.gsub!(/^ {12}/, '')
+    str.chomp!
+    str
+  end
+
+
+  # FEATURES -- Returns contents of the FEATURES record as an array of
+  # Bio::Feature objects.
+  def features
+    unless @data['FEATURES']
+      ary = []
+      in_quote = false
+      get('FEATURES').each_line do |line|
+        next if line =~ /^FEATURES/
+
+        # feature type  (source, CDS, ...)
+        head = line[0,20].to_s.strip
+
+        # feature value (position or /qualifier=)
+        body = line[20,60].to_s.chomp
+
+        # sub-array [ feature type, position, /q="data", ... ]
+        if line =~ /^ {5}\S/
+          ary.push([ head, body ])
+
+        # feature qualifier start (/q="data..., /q="data...", /q=data, /q)
+        elsif body =~ /^ \// and not in_quote		# gb:IRO125195
+          ary.last.push(body)
+          
+          # flag for open quote (/q="data...)
+          if body =~ /="/ and body !~ /"$/
+            in_quote = true
+          end
+
+        # feature qualifier continued (...data..., ...data...")
+        else
+          ary.last.last << body
+
+          # flag for closing quote (/q="data... lines  ...")
+          if body =~ /"$/
+            in_quote = false
+          end
+        end
+      end
+
+      ary.collect! do |subary|
+        parse_qualifiers(subary)
+      end
+
+      @data['FEATURES'] = ary.extend(Bio::Features::BackwardCompatibility)
+    end
+    if block_given?
+      @data['FEATURES'].each do |f|
+        yield f
+      end
+    else
+      @data['FEATURES']
+    end
+  end
+
+
+  # ORIGIN -- Returns contents of the ORIGIN record as a String.
+  def origin
+    unless @data['ORIGIN']
+      ori, seqstr = get('ORIGIN').split("\n", 2)
+      seqstr ||= ""
+      @data['ORIGIN'] = truncate(tag_cut(ori))
+      @data['SEQUENCE'] = seqstr.tr("0-9 \t\n\r\/", '')
+    end
+    @data['ORIGIN']
+  end
+
+
+  ### private methods
+
+  private
+
+  def parse_qualifiers(ary)
+    feature = Feature.new
+
+    feature.feature = ary.shift
+    feature.position = ary.shift.gsub(/\s/, '')
+
+    ary.each do |f|
+      if f =~ %r{/([^=]+)=?"?([^"]*)"?}
+        qualifier, value = $1, $2
+
+        case qualifier
+        when 'translation'
+          value = Sequence::AA.new(value)
+        when 'codon_start'
+          value = value.to_i
+        else
+          value = true if value.empty?
+        end
+
+        feature.append(Feature::Qualifier.new(qualifier, value))
+      end
+    end
+
+    return feature
+  end
+
+end # Common 
+
+end # GenBank
+end # Bio
+
+