jandot-bio 1.2.1

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

@@ -0,0 +1,290 @@
+#
+# = 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 2007/04/05 23:35:40 trevor 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
+    accession.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 /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']	= $5
+            else
+      	hash['journal'] = journal
+            end
+          when /MEDLINE/
+            hash['medline']	= truncate(tag_cut(field))
+          when /PUBMED/
+            hash['pubmed']	= truncate(tag_cut(field))
+          end
+        end
+        ary.push(Reference.new(hash))
+      end
+      @data['REFERENCE'] = References.new(ary)
+    end
+    if block_given?
+      @data['REFERENCE'].each do |r|
+        yield r
+      end
+    else
+      @data['REFERENCE']
+    end
+  end
+
+
+  # COMMENT -- Returns contents of the COMMENT record as a String.
+  def comment
+    field_fetch('COMMENT')
+  end
+
+
+  # FEATURES -- Returns contents of the FEATURES record as a Bio::Features
+  # object.
+  def features
+    unless @data['FEATURES']
+      ary = []
+      in_quote = false
+      get('FEATURES').each_line do |line|
+        next if line =~ /^FEATURES/
+
+        # feature type  (source, CDS, ...)
+        head = line[0,20].to_s.strip
+
+        # feature value (position or /qualifier=)
+        body = line[20,60].to_s.chomp
+
+        # sub-array [ feature type, position, /q="data", ... ]
+        if line =~ /^ {5}\S/
+          ary.push([ head, body ])
+
+        # feature qualifier start (/q="data..., /q="data...", /q=data, /q)
+        elsif body =~ /^ \// and not in_quote		# gb:IRO125195
+          ary.last.push(body)
+          
+          # flag for open quote (/q="data...)
+          if body =~ /="/ and body !~ /"$/
+            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'] = Features.new(ary)
+    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
+
+

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

@@ -0,0 +1,22 @@
+#
+# = bio/db/genbank/ddbj.rb - DDBJ database class
+#
+# Copyright::  Copyright (C) 2000-2004 Toshiaki Katayama <k@bioruby.org>
+# License::    The Ruby License
+#
+# $Id: ddbj.rb,v 1.9 2007/04/05 23:35:40 trevor Exp $
+#
+
+require 'bio/db/genbank/genbank'
+
+module Bio
+
+class DDBJ < GenBank
+
+  autoload :XML,          'bio/io/ddbjxml'
+
+  # Nothing to do (DDBJ database format is completely same as GenBank)
+
+end # DDBJ
+
+end # Bio

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

@@ -0,0 +1,215 @@
+#
+# = bio/db/genbank/genbank.rb - GenBank database class
+#
+# 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 $
+#
+
+require 'bio/db'
+require 'bio/db/genbank/common'
+
+module Bio
+
+# == Description
+#
+# Parses a GenBank formatted database entry
+#
+# == Example
+#
+#   # entry is a string containing only one entry contents
+#   gb = Bio::GenBank.new(entry)
+#
+class GenBank < NCBIDB
+
+  include Bio::NCBIDB::Common
+
+  # Parses the LOCUS line and returns contents of the LOCUS record
+  # as a Bio::GenBank::Locus object.  Locus object is created automatically
+  # when Bio::GenBank#locus, entry_id etc. methods are called.
+  class Locus
+    def initialize(locus_line)
+      if locus_line.empty?
+        # do nothing (just for empty or incomplete entry string)
+      elsif locus_line.length > 75 			# after Rel 126.0
+        @entry_id = locus_line[12..27].strip
+        @length   = locus_line[29..39].to_i
+        @strand   = locus_line[44..46].strip
+        @natype   = locus_line[47..52].strip
+        @circular = locus_line[55..62].strip
+        @division = locus_line[63..66].strip
+        @date     = locus_line[68..78].strip
+      else
+        @entry_id = locus_line[12..21].strip
+        @length   = locus_line[22..29].to_i
+        @strand   = locus_line[33..35].strip
+        @natype   = locus_line[36..39].strip
+        @circular = locus_line[42..51].strip
+        @division = locus_line[52..54].strip
+        @date     = locus_line[62..72].strip
+      end
+    end
+    attr_accessor :entry_id, :length, :strand, :natype, :circular,
+      :division, :date
+  end
+
+  # Accessor methods for the contents of the LOCUS record.
+
+  def locus
+    @data['LOCUS'] ||= Locus.new(get('LOCUS'))
+  end
+
+  def entry_id;  locus.entry_id;  end
+  def length;    locus.length;    end
+  def circular;  locus.circular;  end
+  def division;  locus.division;  end
+  def date;      locus.date;      end
+
+  def strand;    locus.strand;    end
+  def natype;    locus.natype;    end
+
+
+  # FEATURES -- Iterate only for the 'CDS' portion of the Bio::Features.
+  def each_cds
+    features.each do |feature|
+      if feature.feature == 'CDS'
+        yield(feature)
+      end
+    end
+  end
+
+  # FEATURES -- Iterate only for the 'gene' portion of the Bio::Features.
+  def each_gene
+    features.each do |feature|
+      if feature.feature == 'gene'
+        yield(feature)
+      end
+    end
+  end
+
+
+  # BASE COUNT (this field is obsoleted after GenBank release 138.0) --
+  # Returns the BASE COUNT as a Hash.  When the base is specified, returns
+  # count of the base as a Fixnum.  The base can be one of 'a', 't', 'g',
+  # 'c', and 'o' (others).
+  def basecount(base = nil)
+    unless @data['BASE COUNT']
+      hash = Hash.new(0)
+      get('BASE COUNT').scan(/(\d+) (\w)/).each do |c, b|
+        hash[b] = c.to_i
+      end
+      @data['BASE COUNT'] = hash
+    end
+
+    if base
+      base.downcase!
+      @data['BASE COUNT'][base]
+    else
+      @data['BASE COUNT']
+    end
+  end
+
+  # ORIGIN -- Returns DNA sequence in the ORIGIN record as a
+  # Bio::Sequence::NA object.
+  def seq
+    unless @data['SEQUENCE']
+      origin
+    end
+    Bio::Sequence::NA.new(@data['SEQUENCE'])
+  end
+  alias naseq seq
+  alias nalen length
+
+  def seq_len
+    seq.length
+  end
+
+end # GenBank
+end # Bio
+
+
+
+if __FILE__ == $0
+
+  begin
+    require 'pp'
+    alias p pp
+  rescue LoadError
+  end
+
+  puts "### GenBank"
+  if ARGV.size > 0
+    gb = Bio::GenBank.new(ARGF.read)
+  else
+    require 'bio/io/fetch'
+    gb = Bio::GenBank.new(Bio::Fetch.query('gb', 'LPATOVGNS'))
+  end
+
+  puts "## LOCUS"
+  puts "# GenBank.locus"
+  p gb.locus
+  puts "# GenBank.entry_id"
+  p gb.entry_id
+  puts "# GenBank.nalen"
+  p gb.nalen
+  puts "# GenBank.strand"
+  p gb.strand
+  puts "# GenBank.natype"
+  p gb.natype
+  puts "# GenBank.circular"
+  p gb.circular
+  puts "# GenBank.division"
+  p gb.division
+  puts "# GenBank.date"
+  p gb.date
+
+  puts "## DEFINITION"
+  p gb.definition
+
+  puts "## ACCESSION"
+  p gb.accession
+
+  puts "## VERSION"
+  p gb.versions
+  p gb.version
+  p gb.gi
+
+  puts "## NID"
+  p gb.nid
+
+  puts "## KEYWORDS"
+  p gb.keywords
+
+  puts "## SEGMENT"
+  p gb.segment
+
+  puts "## SOURCE"
+  p gb.source
+  p gb.common_name
+  p gb.vernacular_name
+  p gb.organism
+  p gb.taxonomy
+
+  puts "## REFERENCE"
+  p gb.references
+
+  puts "## COMMENT"
+  p gb.comment
+
+  puts "## FEATURES"
+  p gb.features
+
+  puts "## BASE COUNT"
+  p gb.basecount
+  p gb.basecount('a')
+  p gb.basecount('A')
+
+  puts "## ORIGIN"
+  p gb.origin
+  p gb.naseq
+
+end
+
+
+