bio 1.2.1 → 1.3.0

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

@@ -4,7 +4,7 @@
 # Copyright::  Copyright (C) 2004 Toshiaki Katayama <k@bioruby.org>
 # License::    The Ruby License
 #
-# $Id: common.rb,v 1.11 2007/04/05 23:35:40 trevor Exp $
+# $Id: common.rb,v 1.11.2.5 2008/06/17 15:53:21 ngoto Exp $
 #
 
 require 'bio/db'
@@ -44,7 +44,7 @@ module Common
 
   # ACCESSION -- Returns contents of the ACCESSION record as an Array.
   def accessions
-    accession.split(/\s+/)
+    field_fetch('ACCESSION').strip.split(/\s+/)
   end
 
 
@@ -137,9 +137,20 @@ module Common
     unless @data['REFERENCE']
       ary = []
       toptag2array(get('REFERENCE')).each do |ref|
-        hash = Hash.new('')
+        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(/, /)
@@ -163,11 +174,14 @@ module Common
             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'] = References.new(ary)
+      @data['REFERENCE'] = ary.extend(Bio::References::BackwardCompatibility)
     end
     if block_given?
       @data['REFERENCE'].each do |r|
@@ -181,12 +195,15 @@ module Common
 
   # COMMENT -- Returns contents of the COMMENT record as a String.
   def comment
-    field_fetch('COMMENT')
+    str = get('COMMENT').to_s.sub(/\ACOMMENT     /, '')
+    str.gsub!(/^ {12}/, '')
+    str.chomp!
+    str
   end
 
 
-  # FEATURES -- Returns contents of the FEATURES record as a Bio::Features
-  # object.
+  # FEATURES -- Returns contents of the FEATURES record as an array of
+  # Bio::Feature objects.
   def features
     unless @data['FEATURES']
       ary = []
@@ -228,7 +245,7 @@ module Common
         parse_qualifiers(subary)
       end
 
-      @data['FEATURES'] = Features.new(ary)
+      @data['FEATURES'] = ary.extend(Bio::Features::BackwardCompatibility)
     end
     if block_given?
       @data['FEATURES'].each do |f|

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

@@ -0,0 +1,187 @@
+#
+# = bio/db/genbank/format_genbank.rb - GenBank format generater
+#
+# Copyright::  Copyright (C) 2008 Naohisa Goto <ng@bioruby.org>
+# License::    The Ruby License
+#
+# $Id: format_genbank.rb,v 1.1.2.5 2008/06/17 15:59:24 ngoto Exp $
+#
+
+require 'bio/sequence/format'
+
+module Bio::Sequence::Format::NucFormatter
+
+  # INTERNAL USE ONLY, YOU SHOULD NOT USE THIS CLASS.
+  # GenBank format output class for Bio::Sequence.
+  class Genbank < Bio::Sequence::Format::FormatterBase
+
+    # helper methods
+    include Bio::Sequence::Format::INSDFeatureHelper
+    
+    private
+
+    # string wrapper for GenBank format
+    def genbank_wrap(str)
+      wrap(str.to_s, 67).gsub(/\n/, "\n" + " " * 12)
+    end
+
+    # string wrap with adding a dot at the end of the string
+    def genbank_wrap_dot(str)
+      str = str.to_s
+      str = str + '.' unless /\.\z/ =~ str
+      genbank_wrap(str)
+    end
+
+    # Given words (an Array of String) are wrapping with EMBL style.
+    # Each word is never splitted inside the word.
+    def genbank_wrap_words(array)
+      width = 67
+      result = []
+      str = nil
+      array.each do |x|
+        if str then
+          if str.length + 1 + x.length > width then
+            str = nil
+          else
+            str.concat ' '
+            str.concat x
+          end
+        end
+        unless str then
+          str = "#{x}"
+          result.push str
+        end
+      end
+      result.join("\n" + " " * 12)
+    end
+
+    # formats references
+    def reference_format_genbank(ref, num)
+      pos = ref.sequence_position.to_s.gsub(/\s/, '')
+      pos.gsub!(/(\d+)\-(\d+)/, "\\1 to \\2")
+      pos.gsub!(/\s*\,\s*/, '; ')
+      if pos.empty?
+        pos = ''
+      else
+        pos = " (bases #{pos})"
+      end
+      volissue = "#{ref.volume.to_s}"
+      volissue += " (#{ref.issue})" unless ref.issue.to_s.empty? 
+      journal = "#{ref.journal.to_s}"
+      journal += " #{volissue}" unless volissue.empty? 
+      journal += ", #{ref.pages}" unless ref.pages.to_s.empty?
+      journal += " (#{ref.year})" unless ref.year.to_s.empty?
+
+      alist = ref.authors.collect do |x|
+        y = x.to_s.strip.split(/\, *([^\,]+)\z/)
+        y[1].gsub!(/\. +/, '.') if y[1]
+        y.join(',')
+      end
+      lastauthor = alist.pop
+      last2author = alist.pop
+      alist.each { |x| x.concat ',' }
+      alist.push last2author if last2author
+      alist.push "and" unless alist.empty?
+      alist.push lastauthor.to_s
+      result = <<__END_OF_REFERENCE__
+REFERENCE   #{ genbank_wrap(sprintf('%-2d%s', num, pos))}
+  AUTHORS   #{ genbank_wrap_words(alist) }
+  TITLE     #{ genbank_wrap(ref.title.to_s) }
+  JOURNAL   #{ genbank_wrap(journal) }
+__END_OF_REFERENCE__
+      unless ref.pubmed.to_s.empty? then
+        result.concat "   PUBMED   #{ genbank_wrap(ref.pubmed) }\n"
+      end
+      if ref.comments and !(ref.comments.empty?) then
+        ref.comments.each do |c|
+          result.concat "  REMARK    #{ genbank_wrap(c) }\n"
+        end
+      end
+      result
+    end
+
+    # formats comments lines as GenBank
+    def comments_format_genbank(cmnts)
+      return '' if !cmnts or cmnts.empty?
+      cmnts = [ cmnts ] unless cmnts.kind_of?(Array)
+      a = []
+      cmnts.each do |str|
+        a.push "COMMENT     #{ genbank_wrap(str) }\n"
+      end
+      a.join('')
+    end
+
+    # formats sequence lines as GenBank
+    def seq_format_genbank(str)
+      i = 1
+      result = str.gsub(/.{1,60}/) do |s|
+        s = s.gsub(/.{1,10}/, ' \0')
+        y = sprintf("%9d%s\n", i, s)
+        i += 60
+        y
+      end
+      result
+    end
+
+    # formats date
+    def date_format_genbank
+      date_modified || date_created || null_date
+    end
+
+    # moleculue type
+    def mol_type_genbank
+      if /(DNA|(t|r|m|u|sn|sno)?RNA)/i =~ molecule_type.to_s then
+        $1.sub(/[DR]NA/) { |x| x.upcase }
+      else
+        'NA'
+      end
+    end
+
+    # NCBI GI number
+    def ncbi_gi_number
+      ids = other_seqids
+      if ids and r = ids.find { |x| x.database == 'GI' } then
+        r.id
+      else
+        nil
+      end
+    end
+
+    # strandedness
+    def strandedness_genbank
+      return nil unless strandedness
+      case strandedness
+      when 'single'; 'ss-'; 
+      when 'double'; 'ds-'; 
+      when 'mixed';  'ms-'; 
+      else; nil
+      end
+    end
+
+    # Erb template of GenBank format for Bio::Sequence
+    erb_template <<'__END_OF_TEMPLATE__'
+LOCUS       <%= sprintf("%-16s", entry_id) %> <%= sprintf("%11d", length) %> bp <%= sprintf("%3s", strandedness_genbank) %><%= sprintf("%-6s", mol_type_genbank) %>  <%= sprintf("%-8s", topology) %><%= sprintf("%4s", division) %> <%= date_format_genbank %>
+DEFINITION  <%= genbank_wrap_dot(definition.to_s) %>
+ACCESSION   <%= genbank_wrap(([ primary_accession ] + (secondary_accessions or [])).join(" ")) %>
+VERSION     <%= primary_accession %>.<%= sequence_version %><% if gi = ncbi_gi_number then %>  GI:<%= gi %><% end %>
+KEYWORDS    <%= genbank_wrap_dot((keywords or []).join('; ')) %>
+SOURCE      <%= genbank_wrap(species) %>
+  ORGANISM  <%= genbank_wrap(species) %>
+            <%= genbank_wrap_dot((classification or []).join('; ')) %>
+<% 
+    n = 0
+    (references or []).each do |ref|
+      n += 1
+%><%= reference_format_genbank(ref, n) %><%
+    end
+%><%= comments_format_genbank(comments)
+%>FEATURES             Location/Qualifiers
+<%= format_features_genbank(features || [])
+ %>ORIGIN
+<%= seq_format_genbank(seq)
+ %>//
+__END_OF_TEMPLATE__
+
+  end #class Genbank
+end #module Bio::Sequence::Format::NucFormatter
+

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

@@ -4,11 +4,14 @@
 # Copyright::  Copyright (C) 2000-2005 Toshiaki Katayama <k@bioruby.org>
 # License::    The Ruby License
 #
-# $Id: genbank.rb,v 0.40 2007/04/05 23:35:40 trevor Exp $
+# $Id: genbank.rb,v 0.40.2.4 2008/06/17 15:56:18 ngoto Exp $
 #
 
+require 'date'
 require 'bio/db'
 require 'bio/db/genbank/common'
+require 'bio/sequence'
+require 'bio/sequence/dblink'
 
 module Bio
 
@@ -121,10 +124,42 @@ class GenBank < NCBIDB
   alias naseq seq
   alias nalen length
 
+  # (obsolete???) length of the sequence
   def seq_len
     seq.length
   end
 
+  # modified date. Returns Date object, String or nil.
+  def date_modified
+    begin
+      Date.parse(self.date)
+    rescue ArgumentError, TypeError, NoMethodError, NameError
+      self.date
+    end
+  end
+
+  # Taxonomy classfication. Returns an array of strings.
+  def classification
+    self.taxonomy.to_s.sub(/\.\z/, '').split(/\s*\;\s*/)
+  end
+
+  # Strandedness. Returns one of 'single', 'double', 'mixed', or nil.
+  def strandedness
+    case self.strand.to_s.downcase
+    when 'ss-'; 'single'
+    when 'ds-'; 'double'
+    when 'ms-'; 'mixed'
+    else nil; end
+  end
+
+  # converts Bio::GenBank to Bio::Sequence
+  # ---
+  # *Arguments*: 
+  # *Returns*:: Bio::Sequence object
+  def to_biosequence
+    Bio::Sequence.adapter(self, Bio::Sequence::Adapter::GenBank)
+  end
+
 end # GenBank
 end # Bio