bio 1.4.3.0001 → 1.5.0

data/lib/bio/db/embl/swissprot.rb

@@ -1,41 +1,24 @@
 #
-# = bio/db/embl/swissprot.rb - SwissProt database class
+# = bio/db/embl/swissprot.rb - (deprecated) SwissProt database class
 # 
-# Copyright::   Copyright (C) 2001, 2002 Toshiaki Katayama <k@bioruby.org>
+# Copyright::   Copyright (C) 2013 BioRuby Project
 # License::     The Ruby License
 #
-#  $Id: swissprot.rb,v 1.7 2007/04/05 23:35:40 trevor Exp $
-#
 
-require 'bio/db/embl/sptr'
+warn "Bio::SwissProt is deprecated. Use Bio::UniProtKB."
 
 module Bio
 
-# == Description
-# 
-# Parser class for SwissProt database entry. See also Bio::SPTR class.
-# This class holds name space for SwissProt specific methods.
-#
-# SwissProt (before UniProtKB/SwissProt) specific methods are defined in 
-# this class. Shared methods for UniProtKB/SwissProt and TrEMBL classes 
-# are defined in Bio::SPTR class.
-#
-# == Examples
-#
-#   str = File.read("p53_human.swiss")
-#   obj = Bio::SwissProt.new(str)
-#   obj.entry_id #=> "P53_HUMAN"
-#
-# == Referencees
-#
-# * Swiss-Prot Protein knowledgebase
-#   http://au.expasy.org/sprot/
-#
-# * Swiss-Prot Protein Knowledgebase User Manual
-#   http://au.expasy.org/sprot/userman.html
-# 
+require 'bio/db/embl/uniprotkb' unless const_defined?(:UniProtKB)
+
+# Bio::SwissProt is deprecated. Use Bio::UniProtKB.
 class SwissProt < SPTR
-  # Nothing to do (SwissProt format is abstracted in SPTR)
+
+  # Bio::SwissProt is deprecated. Use Bio::UniProtKB.
+  def initialize(str)
+    warn "Bio::SwissProt is deprecated. Use Bio::UniProtKB."
+    super(str)
+  end
 end
 
 end

data/lib/bio/db/embl/trembl.rb

@@ -1,41 +1,24 @@
 #
-# = bio/db/embl/trembl.rb - TrEMBL database class
+# = bio/db/embl/trembl.rb - (deprecated) TrEMBL database class
 # 
-# Copyright::  Copyright (C) 2001, 2002 Toshiaki Katayama <k@bioruby.org>
+# Copyright::  Copyright (C) 2013 BioRuby Project
 # License::    The Ruby License
 #
-# $Id: trembl.rb,v 1.7 2007/04/05 23:35:40 trevor Exp $
-#
 
-require 'bio/db/embl/sptr'
+warn "Bio::TrEMBL is deprecated. Use Bio::UniProtKB."
 
 module Bio
 
-# == Description
-#
-# Parser class for TrEMBL database entry. See also Bio::SPTR class.
-# This class holds name space for TrEMBL specific methods.
-#
-# UniProtKB/SwissProt specific methods are defined in this class. 
-# Shared methods for UniProtKB/SwissProt and TrEMBL classes are 
-# defined in Bio::SPTR class.
-#
-# == Examples
-#
-#   str = File.read("Q2UNG2_ASPOR.trembl")
-#   obj = Bio::TrEMBL.new(str)
-#   obj.entry_id #=> "Q2UNG2_ASPOR"
-#
-# == Referencees
-#
-# * TrEMBL Computer-annotated supplement to Swiss-Prot	
-#   http://au.expasy.org/sprot/
-#
-# * TrEMBL Computer-annotated supplement to Swiss-Prot User Manual
-#   http://au.expasy.org/sprot/userman.html
-# 
-class TrEMBL < SPTR
-  # Nothing to do (TrEMBL format is abstracted in SPTR)
+require 'bio/db/embl/uniprotkb' unless const_defined?(:UniProtKB)
+
+# Bio::TrEMBL is deprecated. Use Bio::UniProtKB.
+class TrEMBL < UniProtKB
+
+  # Bio::TrEMBL is deprecated. Use Bio::UniProtKB.
+  def initialize(str)
+    warn "Bio::TrEMBL is deprecated. Use Bio::UniProtKB."
+    super(str)
+  end
 end
 
 end

data/lib/bio/db/embl/uniprot.rb

@@ -1,42 +1,25 @@
 #
 # = bio/db/embl/uniprot.rb - UniProt database class
 # 
-# Copyright::  Copyright (C) 2005 Toshiaki Katayama <k@bioruby.org>
+# Copyright::  Copyright (C) 2013 BioRuby Project
 # License::    The Ruby License
 #
-#  $Id: uniprot.rb,v 1.5 2007/04/05 23:35:40 trevor Exp $
 #
 
-require 'bio/db/embl/sptr'
+warn "Bio::UniProt is an alias of Bio::UniProtKB. Please use Bio::UniProtKB. Bio::UniProt may be deprecated in the future." if $VERBOSE
 
 module Bio
 
-# == Description
-# 
-# Parser class for SwissProt database entry.# See also Bio::SPTR class.
-# This class holds name space for UniProtKB/SwissProt specific methods.
-#
-# UniProtKB/SwissProt specific methods are defined in this class. 
-# Shared methods for UniProtKB/SwissProt and TrEMBL classes are 
-# defined in Bio::SPTR class.
-#
-# == Examples
-#
-#   str = File.read("p53_human.swiss")
-#   obj = Bio::UniProt.new(str)
-#   obj.entry_id #=> "P53_HUMAN"
-#
-# == Referencees
-#
-# * UniProt
-#   http://uniprot.org/
-#
-# * The UniProtKB/SwissProt/TrEMBL User Manual
-#   http://www.expasy.org/sprot/userman.html
-#
-class UniProt < SPTR
-  # Nothing to do (UniProt format is abstracted in SPTR)
-end
+require 'bio/db/embl/uniprotkb' unless const_defined?(:UniProtKB)
+
+  # Bio::UniProt is changed to an alias of Bio::UniProtKB.
+  # Please use Bio::UniProtKB.
+  # Bio::UniProt may be deprecated in the future.
+  #
+  # Note that Bio::SPTR have been renamed to Bio::UniProtKB and
+  # is also an alias of Bio::UniProtKB.
+  #
+  UniProt = UniProtKB
 
 end
 

data/lib/bio/db/embl/uniprotkb.rb

@@ -0,0 +1,1455 @@
+#
+# = bio/db/embl/uniprotkb.rb - UniProtKB data parser class
+# 
+# Copyright::   Copyright (C) 2001-2006  Mitsuteru C. Nakao <n@bioruby.org>
+# License::     The Ruby License
+#
+#
+# == Description
+# 
+# See Bio::UniProtKB documents.
+#
+
+require 'bio/db'
+require 'bio/db/embl/common'
+
+module Bio
+
+# == Description
+#
+# Parser class for UniProtKB/SwissProt and TrEMBL database entry.
+# 
+# See the UniProtKB document files and manuals.
+# 
+# == Examples
+#
+#   str = File.read("p53_human.swiss")
+#   obj = Bio::UniProtKB.new(str)
+#   obj.entry_id #=> "P53_HUMAN"
+# 
+# == References
+# 
+# * The UniProt Knowledgebase (UniProtKB)
+#   http://www.uniprot.org/help/uniprotkb
+#
+# * The Universal Protein Resource (UniProt)
+#   http://uniprot.org/
+#
+# * The UniProtKB/SwissProt/TrEMBL User Manual
+#   http://www.uniprot.org/docs/userman.html
+#
+class UniProtKB < EMBLDB
+  include Bio::EMBLDB::Common
+    
+  @@entry_regrexp = /[A-Z0-9]{1,4}_[A-Z0-9]{1,5}/
+  @@data_class = ["STANDARD", "PRELIMINARY"]
+
+  # returns a Hash of the ID line.
+  #
+  # returns a content (Int or String) of the ID line by a given key.
+  # Hash keys: ['ENTRY_NAME', 'DATA_CLASS', 'MODECULE_TYPE', 'SEQUENCE_LENGTH']
+  #
+  # === ID Line (since UniProtKB release 9.0 of 31-Oct-2006)
+  #   ID   P53_HUMAN               Reviewed;         393 AA.
+  #   #"ID  #{ENTRY_NAME} #{DATA_CLASS}; #{SEQUENCE_LENGTH}."
+  #
+  # === Examples
+  #   obj.id_line  #=> {"ENTRY_NAME"=>"P53_HUMAN", "DATA_CLASS"=>"Reviewed", 
+  #                     "SEQUENCE_LENGTH"=>393, "MOLECULE_TYPE"=>nil}
+  #
+  #   obj.id_line('ENTRY_NAME') #=> "P53_HUMAN"
+  #
+  # 
+  # === ID Line (older style)
+  #   ID   P53_HUMAN      STANDARD;      PRT;   393 AA.
+  #   #"ID  #{ENTRY_NAME} #{DATA_CLASS}; #{MOLECULE_TYPE}; #{SEQUENCE_LENGTH}."
+  #
+  # === Examples
+  #   obj.id_line  #=> {"ENTRY_NAME"=>"P53_HUMAN", "DATA_CLASS"=>"STANDARD", 
+  #                     "SEQUENCE_LENGTH"=>393, "MOLECULE_TYPE"=>"PRT"}
+  #
+  #   obj.id_line('ENTRY_NAME') #=> "P53_HUMAN"
+  #
+  def id_line(key = nil)
+    return id_line[key] if key
+    return @data['ID'] if @data['ID']
+
+    part = @orig['ID'].split(/ +/)         
+    if part[4].to_s.chomp == 'AA.' then
+      # after UniProtKB release 9.0 of 31-Oct-2006
+      # (http://www.uniprot.org/docs/sp_news.htm)
+      molecule_type   = nil
+      sequence_length = part[3].to_i
+    else
+      molecule_type   = part[3].sub(/;/,'')
+      sequence_length = part[4].to_i
+    end
+    @data['ID'] = {
+      'ENTRY_NAME'      => part[1],
+      'DATA_CLASS'      => part[2].sub(/;/,''),
+      'MOLECULE_TYPE'   => molecule_type,
+      'SEQUENCE_LENGTH' => sequence_length
+    }
+  end
+
+
+  # returns a ENTRY_NAME in the ID line. 
+  #
+  def entry_id
+    id_line('ENTRY_NAME')
+  end
+  alias entry_name entry_id
+  alias entry entry_id
+
+
+  # returns a MOLECULE_TYPE in the ID line.
+  #
+  # A short-cut for Bio::UniProtKB#id_line('MOLECULE_TYPE').
+  def molecule
+    id_line('MOLECULE_TYPE')
+  end
+  alias molecule_type molecule
+
+
+  # returns a SEQUENCE_LENGTH in the ID line.
+  # 
+  # A short-cut for Bio::UniProtKB#id_line('SEQUENCE_LENGHT').
+  def sequence_length
+    id_line('SEQUENCE_LENGTH')
+  end
+  alias aalen sequence_length
+
+
+  # Bio::EMBLDB::Common#ac  -> ary
+  #                  #accessions  -> ary
+  #                  #accession  -> String (accessions.first)
+  @@ac_regrexp = /[OPQ][0-9][A-Z0-9]{3}[0-9]/ 
+
+
+
+  # returns a Hash of information in the DT lines.
+  #  hash keys: 
+  #    ['created', 'sequence', 'annotation']
+  #--
+  #  also Symbols acceptable (ASAP):
+  #    [:created, :sequence, :annotation]
+  #++
+  #
+  # Since UniProtKB release 7.0 of 07-Feb-2006, the DT line format is
+  # changed, and the word "annotation" is no longer used in DT lines.
+  # Despite the change, the word "annotation" is still used for keeping
+  # compatibility.
+  #
+  # returns a String of information in the DT lines by a given key.
+  #
+  # === DT Line; date (3/entry)
+  #   DT DD-MMM-YYY (integrated into UniProtKB/XXXXX.)
+  #   DT DD-MMM-YYY (sequence version NN)
+  #   DT DD-MMM-YYY (entry version NN)
+  #
+  # The format have been changed in UniProtKB release 7.0 of 07-Feb-2006.
+  # Below is the older format.
+  #
+  # === Old format of DT Line; date (3/entry)
+  #   DT DD-MMM-YYY (rel. NN, Created)
+  #   DT DD-MMM-YYY (rel. NN, Last sequence update)
+  #   DT DD-MMM-YYY (rel. NN, Last annotation update)
+  def dt(key = nil)
+    return dt[key] if key
+    return @data['DT'] if @data['DT']
+
+    part = self.get('DT').split(/\n/)
+    @data['DT'] = {
+      'created'    => part[0].sub(/\w{2}   /,'').strip,
+      'sequence'   => part[1].sub(/\w{2}   /,'').strip,
+      'annotation' => part[2].sub(/\w{2}   /,'').strip
+    }
+  end
+
+
+  # (private) parses DE line (description lines)
+  # since UniProtKB release 14.0 of 22-Jul-2008
+  #
+  # Return array containing array.
+  #
+  # http://www.uniprot.org/docs/sp_news.htm
+  def parse_DE_line_rel14(str)
+    # Retruns if it is not the new format since Rel.14
+    return nil unless /^DE   (RecName|AltName|SubName)\: / =~ str
+    ret = []
+    cur = nil
+    str.each_line do |line|
+      case line
+      when /^DE   (Includes|Contains)\: *$/
+        cur = [ $1 ]
+        ret.push cur
+        cur = nil
+        #subcat_and_desc = nil
+        next
+      when /^DE   *(RecName|AltName|SubName)\: +(.*)/
+        category = $1
+        subcat_and_desc = $2
+        cur = [ category ]
+        ret.push cur
+      when /^DE   *(Flags)\: +(.*)/
+        category = $1
+        desc = $2
+        flags = desc.strip.split(/\s*\;\s*/) || []
+        cur = [ category, flags ]
+        ret.push cur
+        cur = nil
+        #subcat_and_desc = nil
+        next
+      when /^DE   *(.*)/
+        subcat_and_desc = $1
+      else
+        warn "Warning: skipped DE line in unknown format: #{line.inspect}"
+        #subcat_and_desc = nil
+        next
+      end
+      case subcat_and_desc
+      when nil
+        # does nothing
+      when /\A([^\=]+)\=(.*)/
+        subcat = $1
+        desc = $2
+        desc.sub!(/\;\s*\z/, '')
+        unless cur
+          warn "Warning: unknown category in DE line: #{line.inspect}"
+          cur = [ '' ]
+          ret.push cur
+        end
+        cur.push [ subcat, desc ]
+      else
+        warn "Warning: skipped DE line description in unknown format: #{line.inspect}"
+      end
+    end
+    ret
+  end
+  private :parse_DE_line_rel14
+
+  # returns the proposed official name of the protein.
+  # Returns a String.
+  #
+  # Since UniProtKB release 14.0 of 22-Jul-2008, the DE line format have
+  # been changed. The method returns the full name which is taken from
+  # "RecName: Full=" or "SubName: Full=" line normally in the beginning of
+  # the DE lines. 
+  # Unlike parser for old format, no special treatments for fragment or
+  # precursor.
+  #
+  # For old format, the method parses the DE lines and returns the protein
+  # name as a String.
+  # 
+  # === DE Line; description (>=1)
+  #  "DE #{OFFICIAL_NAME} (#{SYNONYM})"
+  #  "DE #{OFFICIAL_NAME} (#{SYNONYM}) [CONTEINS: #1; #2]."
+  #  OFFICIAL_NAME  1/entry
+  #  SYNONYM        >=0
+  #  CONTEINS       >=0
+  def protein_name
+    @data['DE'] ||= parse_DE_line_rel14(get('DE'))
+    parsed_de_line = @data['DE']
+    if parsed_de_line then
+      # since UniProtKB release 14.0 of 22-Jul-2008
+      name = nil
+      parsed_de_line.each do |a|
+        case a[0]
+        when 'RecName', 'SubName'
+          if name_pair = a[1..-1].find { |b| b[0] == 'Full' } then
+            name = name_pair[1]
+            break
+          end
+        end
+      end
+      name = name.to_s
+    else
+      # old format (before Rel. 13.x)
+      name = ""
+      if de_line = fetch('DE') then
+        str = de_line[/^[^\[]*/] # everything preceding the first [ (the "contains" part)
+        name = str[/^[^(]*/].strip
+        name << ' (Fragment)' if str =~ /fragment/i
+      end
+    end
+    return name
+  end
+
+
+  # returns synonyms (unofficial and/or alternative names).
+  # Returns an Array containing String objects.
+  #
+  # Since UniProtKB release 14.0 of 22-Jul-2008, the DE line format have
+  # been changed. The method returns the full or short names which are
+  # taken from "RecName: Short=", "RecName: EC=", and AltName lines,
+  # except after "Contains:" or "Includes:".
+  # For keeping compatibility with old format parser, "RecName: EC=N.N.N.N"
+  # is reported as "EC N.N.N.N".
+  # In addition, to prevent confusion, "Allergen=" and "CD_antigen=" 
+  # prefixes are added for the corresponding fields.
+  #
+  # For old format, the method parses the DE lines and returns synonyms.
+  # synonyms are each placed in () following the official name on the DE line.
+  def synonyms
+    ary = Array.new
+    @data['DE'] ||= parse_DE_line_rel14(get('DE'))
+    parsed_de_line = @data['DE']
+    if parsed_de_line then
+      # since UniProtKB release 14.0 of 22-Jul-2008
+      parsed_de_line.each do |a|
+        case a[0]
+        when 'Includes', 'Contains'
+          break #the each loop
+        when 'RecName', 'SubName', 'AltName'
+          a[1..-1].each do |b|
+            if name = b[1] and b[1] != self.protein_name then
+              case b[0]
+              when 'EC'
+                name = "EC " + b[1]
+              when 'Allergen', 'CD_antigen'
+                name = b[0] + '=' + b[1]
+              else
+                name = b[1]
+              end
+              ary.push name
+            end
+          end
+        end #case a[0]
+      end #parsed_de_line.each
+    else
+      # old format (before Rel. 13.x)
+      if de_line = fetch('DE') then
+        line = de_line.sub(/\[.*\]/,'') # ignore stuff between [ and ].  That's the "contains" part
+      line.scan(/\([^)]+/) do |synonym| 
+        unless synonym =~ /fragment/i then 
+          ary << synonym[1..-1].strip # index to remove the leading (  
+        end
+        end
+      end
+    end
+    return ary
+  end
+
+
+  # returns gene names in the GN line.
+  #
+  # New UniProt/SwissProt format:
+  # * Bio::UniProtKB#gn -> [ <gene record>* ]
+  # where <gene record> is:
+  #                    { :name => '...', 
+  #                      :synonyms => [ 's1', 's2', ... ],
+  #                      :loci   => [ 'l1', 'l2', ... ],
+  #                      :orfs     => [ 'o1', 'o2', ... ] 
+  #                    }
+  #
+  # Old format:
+  # * Bio::UniProtKB#gn -> Array      # AND 
+  # * Bio::UniProtKB#gn[0] -> Array   # OR
+  #
+  # === GN Line: Gene name(s) (>=0, optional)
+  def gn
+    unless @data['GN']
+      case fetch('GN')
+      when /Name=/,/ORFNames=/,/OrderedLocusNames=/,/Synonyms=/
+        @data['GN'] = gn_uniprot_parser
+      else
+        @data['GN'] = gn_old_parser
+      end
+    end
+    @data['GN']
+  end
+
+
+  # returns contents in the old style GN line.
+  # === GN Line: Gene name(s) (>=0, optional)
+  #  GN   HNS OR DRDX OR OSMZ OR BGLY.
+  #  GN   CECA1 AND CECA2.
+  #  GN   CECA1 AND (HOGE OR FUGA).
+  #
+  #  GN NAME1 [(AND|OR) NAME]+.
+  #
+  # Bio::UniProtKB#gn -> Array      # AND 
+  #          #gn[0] -> Array   # OR
+  #          #gene_names -> Array
+  def gn_old_parser
+    names = Array.new
+    if get('GN').size > 0
+      names = fetch('GN').sub(/\.$/,'').split(/ AND /)
+      names.map! { |synonyms|
+        synonyms = synonyms.gsub(/\(|\)/,'').split(/ OR /).map { |e|
+          e.strip 
+        }
+      }
+    end
+    @data['GN'] = names
+  end
+  private :gn_old_parser
+
+  # returns contents in the structured GN line.
+  # The new format of the GN line is:
+  #  GN   Name=; Synonyms=[, ...]; OrderedLocusNames=[, ...];
+  #  GN   ORFNames=[, ...];
+  #
+  # * Bio::UniProtKB#gn -> [ <gene record>* ]
+  # where <gene record> is:
+  #                    { :name => '...', 
+  #                      :synonyms => [ 's1', 's2', ... ],
+  #                      :loci   => [ 'l1', 'l2', ... ],
+  #                      :orfs     => [ 'o1', 'o2', ... ] 
+  #                    }
+  def gn_uniprot_parser
+    @data['GN'] = Array.new
+    gn_line = fetch('GN').strip
+    records = gn_line.split(/\s*and\s*/)
+    records.each do |record|
+      gene_hash = {:name => '', :synonyms => [], :loci => [], :orfs => []}
+      record.each_line(';') do |element|
+        case element
+        when /Name=/ then
+          gene_hash[:name] = $'[0..-2]
+        when /Synonyms=/ then
+          gene_hash[:synonyms] = $'[0..-2].split(/\s*,\s*/)
+        when /OrderedLocusNames=/ then
+          gene_hash[:loci] = $'[0..-2].split(/\s*,\s*/)
+        when /ORFNames=/ then
+          gene_hash[:orfs] = $'[0..-2].split(/\s*,\s*/)
+        end
+      end
+      @data['GN'] << gene_hash
+    end
+    return @data['GN']
+  end
+  private :gn_uniprot_parser
+
+
+  # returns a Array of gene names in the GN line.
+  def gene_names
+    gn # set @data['GN'] if it hasn't been already done
+    if @data['GN'].first.class == Hash then
+      @data['GN'].collect { |element| element[:name] }
+    else
+      @data['GN'].first
+    end
+  end
+
+
+  # returns a String of the first gene name in the GN line.
+  def gene_name
+    (x = self.gene_names) ? x.first : nil
+  end
+
+
+  # returns a Array of Hashs or a String of the OS line when a key given.
+  # * Bio::EMBLDB#os  -> Array
+  #  [{'name' => '(Human)', 'os' => 'Homo sapiens'}, 
+  #   {'name' => '(Rat)', 'os' => 'Rattus norveticus'}]
+  # * Bio::EPTR#os[0] -> Hash 
+  #  {'name' => "(Human)", 'os' => 'Homo sapiens'}
+  # * Bio::UniProtKB#os[0]['name'] -> "(Human)"
+  # * Bio::EPTR#os(0) -> "Homo sapiens (Human)"
+  # 
+  # === OS Line; organism species (>=1)
+  #  OS   Genus species (name).
+  #  OS   Genus species (name0) (name1).
+  #  OS   Genus species (name0) (name1).
+  #  OS   Genus species (name0), G s0 (name0), and G s (name0) (name1).
+  #  OS   Homo sapiens (Human), and Rarrus norveticus (Rat)
+  #  OS   Hippotis sp. Clark and Watts 825.
+  #  OS   unknown cyperaceous sp.
+  def os(num = nil)
+    unless @data['OS']
+      os = Array.new
+      fetch('OS').split(/, and|, /).each do |tmp|
+        if tmp =~ /(\w+ *[\w \:\'\+\-\.]+[\w\.])/
+          org = $1
+          tmp =~ /(\(.+\))/ 
+          os.push({'name' => $1, 'os' => org})
+        else
+          raise "Error: OS Line. #{$!}\n#{fetch('OS')}\n"
+        end
+      end
+      @data['OS'] = os
+    end
+
+    if num
+      # EX. "Trifolium repens (white clover)"
+      return "#{@data['OS'][num]['os']} #{@data['OS'][num]['name']}"
+    else
+      return @data['OS']
+    end
+  end
+  
+
+  # Bio::EMBLDB::Common#og -> Array
+  # OG Line; organella (0 or 1/entry)
+  # ["MITOCHONDRION", "CHLOROPLAST", "Cyanelle", "Plasmid"]
+  #  or a plasmid name (e.g. "Plasmid pBR322").  
+
+
+  # Bio::EMBLDB::Common#oc -> Array
+  # OC Line; organism classification (>=1)
+  # "OC   Eukaryota; Alveolata; Apicomplexa; Piroplasmida; Theileriidae;"
+  # "OC   Theileria."
+
+
+
+  # returns a Hash of oraganism taxonomy cross-references.
+  # * Bio::UniProtKB#ox -> Hash
+  #    {'NCBI_TaxID' => ['1234','2345','3456','4567'], ...}
+  #
+  # === OX Line; organism taxonomy cross-reference (>=1 per entry)
+  #  OX   NCBI_TaxID=1234;
+  #  OX   NCBI_TaxID=1234, 2345, 3456, 4567;
+  def ox
+    unless @data['OX']
+      tmp = fetch('OX').sub(/\.$/,'').split(/;/).map { |e| e.strip }
+      hsh = Hash.new
+      tmp.each do |e|
+        db,refs = e.split(/=/)
+        hsh[db] = refs.split(/, */)
+      end
+      @data['OX'] = hsh
+    end
+    return @data['OX']
+  end
+
+  # === The OH Line;  
+  #
+  # OH   NCBI_TaxID=TaxID; HostName.
+  # http://br.expasy.org/sprot/userman.html#OH_line
+  def oh
+    unless @data['OH']
+      @data['OH'] = fetch('OH').split("\. ").map {|x|
+        if x =~ /NCBI_TaxID=(\d+);/
+          taxid = $1
+        else
+          raise ArgumentError, ["Error: Invalid OH line format (#{self.entry_id}):",
+                                $!, "\n", get('OH'), "\n"].join
+          
+        end
+        if x =~ /NCBI_TaxID=\d+; (.+)/ 
+          host_name = $1
+          host_name.sub!(/\.$/, '')
+        else
+          host_name = nil
+        end
+        {'NCBI_TaxID' => taxid, 'HostName' => host_name}
+      }
+    end
+    @data['OH']
+  end
+
+
+  
+  # Bio::EMBLDB::Common#ref -> Array
+  # R Lines
+  # RN RC RP RX RA RT RL
+
+  # returns contents in the R lines.
+  # * Bio::EMBLDB::Common#ref -> [ <refernece information Hash>* ]
+  # where <reference information Hash> is:
+  #  {'RN' => '', 'RC' => '', 'RP' => '', 'RX' => '', 
+  #   'RA' => '', 'RT' => '', 'RL' => '', 'RG' => ''}
+  # 
+  # R Lines
+  # * RN RC RP RX RA RT RL RG
+  def ref
+    unless @data['R']
+      @data['R'] = [get('R').split(/\nRN   /)].flatten.map { |str|
+        hash = {'RN' => '', 'RC' => '', 'RP' => '', 'RX' => '', 
+               'RA' => '', 'RT' => '', 'RL' => '', 'RG' => ''}
+        str = 'RN   ' + str unless /^RN   / =~ str
+
+        str.split("\n").each do |line|
+          if /^(R[NPXARLCTG])   (.+)/ =~ line
+            hash[$1] += $2 + ' '
+          else
+            raise "Invalid format in R lines, \n[#{line}]\n"
+          end
+        end
+
+        hash['RN'] = set_RN(hash['RN'])
+        hash['RC'] = set_RC(hash['RC'])
+        hash['RP'] = set_RP(hash['RP'])
+        hash['RX'] = set_RX(hash['RX'])
+        hash['RA'] = set_RA(hash['RA'])
+        hash['RT'] = set_RT(hash['RT'])
+        hash['RL'] = set_RL(hash['RL'])
+        hash['RG'] = set_RG(hash['RG'])
+
+        hash
+      }
+
+    end
+    @data['R']
+  end
+
+  def set_RN(data)
+    data.strip
+  end
+
+  def set_RC(data)
+    data.scan(/([STP]\w+)=(.+);/).map { |comment|
+      [comment[1].split(/, and |, /)].flatten.map { |text|
+        {'Token' => comment[0], 'Text' => text}
+      }
+    }.flatten
+  end
+  private :set_RC
+
+  def set_RP(data)
+    data = data.strip
+    data = data.sub(/\.$/, '')
+    data.split(/, AND |, /i).map {|x| 
+      x = x.strip
+      x = x.gsub('  ', ' ')
+    }
+  end
+  private :set_RP
+
+  def set_RX(data)
+    rx = {'MEDLINE' => nil, 'PubMed' => nil, 'DOI' => nil}
+    if data =~ /MEDLINE=(.+?);/
+      rx['MEDLINE'] = $1
+    end
+    if data =~ /PubMed=(.+?);/
+      rx['PubMed'] = $1
+    end
+    if data =~ /DOI=(.+?);/
+      rx['DOI'] = $1
+    end
+    rx
+  end
+  private :set_RX
+
+  def set_RA(data)
+    data = data.sub(/; *$/, '')
+  end
+  private :set_RA
+
+  def set_RT(data)
+    data = data.sub(/; *$/, '')
+    data = data.gsub(/(^"|"$)/, '')
+  end
+  private :set_RT
+
+  def set_RL(data)
+    data = data.strip
+  end
+  private :set_RL
+
+  def set_RG(data)
+    data = data.split('; ')
+  end
+  private :set_RG
+
+
+
+  # returns Bio::Reference object from Bio::EMBLDB::Common#ref.
+  # * Bio::EMBLDB::Common#ref -> Bio::References
+  def references
+    unless @data['references']
+      ary = self.ref.map {|ent|
+        hash = Hash.new('')
+        ent.each {|key, value|
+          case key
+          when 'RA'
+            hash['authors'] = value.split(/, /)
+          when 'RT'
+            hash['title'] = value
+          when 'RL'
+            if value =~ /(.*) (\d+) \((\d+)\), (\d+-\d+) \((\d+)\)$/
+              hash['journal'] = $1
+              hash['volume']  = $2
+              hash['issue']   = $3
+              hash['pages']   = $4
+              hash['year']    = $5
+            else
+              hash['journal'] = value
+            end
+          when 'RX'  # PUBMED, MEDLINE, DOI
+            value.each do |tag, xref|
+              hash[ tag.downcase ]  = xref
+            end
+          end
+        }
+        Reference.new(hash)
+      }
+      @data['references'] = References.new(ary)
+    end
+    @data['references']
+  end
+
+
+
+
+
+
+  # === The HI line
+  # Bio::UniProtKB#hi #=> hash
+  def hi
+    unless @data['HI']
+      @data['HI'] = []
+      fetch('HI').split(/\. /).each do |hlist|
+        hash = {'Category' => '',  'Keywords' => [], 'Keyword' => ''}
+        hash['Category'], hash['Keywords'] = hlist.split(': ')
+        hash['Keywords'] = hash['Keywords'].split('; ')
+        hash['Keyword'] = hash['Keywords'].pop
+        hash['Keyword'].sub!(/\.$/, '')
+        @data['HI'] << hash
+      end
+    end
+    @data['HI']
+  end
+
+
+  @@cc_topics = ['PHARMACEUTICAL',
+                 'BIOTECHNOLOGY',
+                 'TOXIC DOSE', 
+                 'ALLERGEN',   
+                 'RNA EDITING',
+                 'POLYMORPHISM',
+                 'BIOPHYSICOCHEMICAL PROPERTIES',
+                 'MASS SPECTROMETRY',
+                 'WEB RESOURCE', 
+                 'ENZYME REGULATION',
+                 'DISEASE',
+                 'INTERACTION',
+                 'DEVELOPMENTAL STAGE',
+                 'INDUCTION',
+                 'CAUTION',
+                 'ALTERNATIVE PRODUCTS',
+                 'DOMAIN',
+                 'PTM',
+                 'MISCELLANEOUS',
+                 'TISSUE SPECIFICITY',
+                 'COFACTOR',
+                 'PATHWAY',
+                 'SUBUNIT',
+                 'CATALYTIC ACTIVITY',
+                 'SUBCELLULAR LOCATION',
+                 'FUNCTION',
+                 'SIMILARITY']
+  # returns contents in the CC lines.
+  # * Bio::UniProtKB#cc -> Hash
+  #
+  # returns an object of contents in the TOPIC.
+  # * Bio::UniProtKB#cc(TOPIC) -> Array w/in Hash, Hash
+  #
+  # returns contents of the "ALTERNATIVE PRODUCTS".
+  # * Bio::UniProtKB#cc('ALTERNATIVE PRODUCTS') -> Hash
+  #    {'Event' => str, 
+  #     'Named isoforms' => int,  
+  #     'Comment' => str,
+  #     'Variants'=>[{'Name' => str, 'Synonyms' => str, 'IsoId' => str, 'Sequence' => []}]}
+  # 
+  #    CC   -!- ALTERNATIVE PRODUCTS:
+  #    CC       Event=Alternative splicing; Named isoforms=15;
+  #    ...
+  #    CC         placentae isoforms. All tissues differentially splice exon 13;
+  #    CC       Name=A; Synonyms=no del;
+  #    CC         IsoId=P15529-1; Sequence=Displayed;
+  #
+  # returns contents of the "DATABASE".
+  # * Bio::UniProtKB#cc('DATABASE') -> Array
+  #    [{'NAME'=>str,'NOTE'=>str, 'WWW'=>URI,'FTP'=>URI}, ...]
+  #
+  #    CC   -!- DATABASE: NAME=Text[; NOTE=Text][; WWW="Address"][; FTP="Address"].
+  #
+  # returns contents of the "MASS SPECTROMETRY".
+  # * Bio::UniProtKB#cc('MASS SPECTROMETRY') -> Array
+  #    [{'MW"=>float,'MW_ERR'=>float, 'METHOD'=>str,'RANGE'=>str}, ...]
+  #
+  #    CC   -!- MASS SPECTROMETRY: MW=XXX[; MW_ERR=XX][; METHOD=XX][;RANGE=XX-XX].
+  #
+  # === CC lines (>=0, optional)
+  #   CC   -!- TISSUE SPECIFICITY: HIGHEST LEVELS FOUND IN TESTIS. ALSO PRESENT
+  #   CC       IN LIVER, KIDNEY, LUNG AND BRAIN.
+  # 
+  #   CC   -!- TOPIC: FIRST LINE OF A COMMENT BLOCK;
+  #   CC       SECOND AND SUBSEQUENT LINES OF A COMMENT BLOCK.
+  #
+  # See also http://www.expasy.org/sprot/userman.html#CC_line
+  #
+  def cc(topic = nil)
+    unless @data['CC']
+      cc  = Hash.new
+      comment_border= '-' * (77 - 4 + 1)
+      dlm = /-!- /
+
+      # 12KD_MYCSM has no CC lines.
+      return cc if get('CC').size == 0
+      
+      cc_raw = fetch('CC')
+
+      # Removing the copyright statement.
+      cc_raw.sub!(/ *---.+---/m, '')
+
+      # Not any CC Lines without the copyright statement.
+      return cc if cc_raw == ''
+
+      begin
+        cc_raw, copyright = cc_raw.split(/#{comment_border}/)[0]
+        _ = copyright #dummy for suppress "assigned but unused variable"
+        cc_raw = cc_raw.sub(dlm,'')
+        cc_raw.split(dlm).each do |tmp|
+          tmp = tmp.strip
+
+          if /(^[A-Z ]+[A-Z]): (.+)/ =~ tmp
+            key  = $1
+            body = $2
+            body.gsub!(/- (?!AND)/,'-')
+            body.strip!
+            unless cc[key]
+              cc[key] = [body]
+            else
+              cc[key].push(body)
+            end
+          else
+            raise ["Error: [#{entry_id}]: CC Lines", '"', tmp, '"',
+                   '', get('CC'),''].join("\n")
+          end
+        end
+      rescue NameError
+        if fetch('CC') == ''
+          return {}
+        else
+          raise ["Error: Invalid CC Lines: [#{entry_id}]: ",
+                 "\n'#{self.get('CC')}'\n", "(#{$!})"].join
+        end
+      rescue NoMethodError
+      end
+      
+      @data['CC'] = cc
+    end
+
+
+    case topic
+    when 'ALLERGEN'
+      return @data['CC'][topic]
+    when 'ALTERNATIVE PRODUCTS'
+      return cc_alternative_products(@data['CC'][topic])
+    when 'BIOPHYSICOCHEMICAL PROPERTIES'
+      return cc_biophysiochemical_properties(@data['CC'][topic])
+    when 'BIOTECHNOLOGY'
+      return @data['CC'][topic]
+    when 'CATALITIC ACTIVITY'
+      return cc_catalytic_activity(@data['CC'][topic])
+    when 'CAUTION'
+      return cc_caution(@data['CC'][topic])
+    when 'COFACTOR'
+      return @data['CC'][topic]
+    when 'DEVELOPMENTAL STAGE'
+      return @data['CC'][topic].join('')
+    when 'DISEASE'
+      return @data['CC'][topic].join('')
+    when 'DOMAIN'
+      return @data['CC'][topic]
+    when 'ENZYME REGULATION'
+      return @data['CC'][topic].join('')
+    when 'FUNCTION'
+      return @data['CC'][topic].join('')
+    when 'INDUCTION'
+      return @data['CC'][topic].join('')
+    when 'INTERACTION'
+      return cc_interaction(@data['CC'][topic])
+    when 'MASS SPECTROMETRY'
+      return cc_mass_spectrometry(@data['CC'][topic])
+    when 'MISCELLANEOUS'
+      return @data['CC'][topic]
+    when 'PATHWAY'
+      return cc_pathway(@data['CC'][topic])
+    when 'PHARMACEUTICAL'
+      return @data['CC'][topic]
+    when 'POLYMORPHISM'
+      return @data['CC'][topic]
+    when 'PTM'
+      return @data['CC'][topic]
+    when 'RNA EDITING'
+      return cc_rna_editing(@data['CC'][topic])
+    when 'SIMILARITY'
+      return @data['CC'][topic]
+    when 'SUBCELLULAR LOCATION'
+      return cc_subcellular_location(@data['CC'][topic])
+    when 'SUBUNIT'
+      return @data['CC'][topic]
+    when 'TISSUE SPECIFICITY'
+      return @data['CC'][topic]
+    when 'TOXIC DOSE'
+      return @data['CC'][topic]
+    when 'WEB RESOURCE'
+      return cc_web_resource(@data['CC'][topic])
+    when 'DATABASE'
+      # DATABASE: NAME=Text[; NOTE=Text][; WWW="Address"][; FTP="Address"].
+      tmp = Array.new
+      db = @data['CC']['DATABASE']
+      return db unless db
+
+      db.each do |e|
+        db = {'NAME' => nil, 'NOTE' => nil, 'WWW' => nil, 'FTP' => nil}
+        e.sub(/.$/,'').split(/;/).each do |line|
+          case line
+          when /NAME=(.+)/
+            db['NAME'] = $1
+          when /NOTE=(.+)/
+            db['NOTE'] = $1
+          when /WWW="(.+)"/
+            db['WWW'] = $1
+          when /FTP="(.+)"/
+            db['FTP'] = $1
+          end 
+        end
+        tmp.push(db)
+      end
+      return tmp
+    when nil
+      return @data['CC']
+    else
+      return @data['CC'][topic]
+    end
+  end
+
+
+  def cc_alternative_products(data)
+    ap = data.join('')
+    return ap unless ap
+
+    # Event, Named isoforms, Comment, [Name, Synonyms, IsoId, Sequnce]+
+    tmp = {'Event' => "", 'Named isoforms' => "", 'Comment' => "", 
+           'Variants'  => []}
+    if /Event=(.+?);/ =~ ap
+      tmp['Event'] = $1
+      tmp['Event'] = tmp['Event'].sub(/;/,'').split(/, /)
+    end
+    if /Named isoforms=(\S+?);/ =~ ap
+      tmp['Named isoforms'] = $1
+    end
+    if /Comment=(.+?);/m =~ ap
+      tmp['Comment'] = $1
+    end
+    ap.scan(/Name=.+?Sequence=.+?;/).each do |ent|
+      tmp['Variants'] << cc_alternative_products_variants(ent)
+    end
+    return tmp
+  end
+  private :cc_alternative_products
+
+  def cc_alternative_products_variants(data)
+    variant = {'Name' => '', 'Synonyms' => [], 'IsoId' => [], 'Sequence' => []}
+    data.split(/; /).map {|x| x.split(/=/) }.each do |e|
+      case e[0]
+      when 'Sequence', 'Synonyms', 'IsoId'
+        e[1] = e[1].sub(/;/,'').split(/, /)
+      end
+      variant[e[0]] = e[1]
+    end
+    variant
+  end
+  private :cc_alternative_products_variants
+
+
+  def cc_biophysiochemical_properties(data)
+    data = data[0]
+
+    hash = {'Absorption' => {}, 
+            'Kinetic parameters' => {},
+            'pH dependence' => "",
+            'Redox potential' => "",
+            'Temperature dependence' => ""}
+    if data =~ /Absorption: Abs\(max\)=(.+?);/
+      hash['Absorption']['Abs(max)'] = $1
+    end
+    if data =~ /Absorption: Abs\(max\)=.+; Note=(.+?);/
+      hash['Absorption']['Note'] = $1
+    end
+    if data =~ /Kinetic parameters: KM=(.+?); Vmax=(.+?);/
+      hash['Kinetic parameters']['KM'] = $1
+      hash['Kinetic parameters']['Vmax'] = $2
+    end
+    if data =~ /Kinetic parameters: KM=.+; Vmax=.+; Note=(.+?);/
+      hash['Kinetic parameters']['Note'] = $1
+    end
+    if data =~ /pH dependence: (.+?);/
+      hash['pH dependence'] = $1
+    end
+    if data =~ /Redox potential: (.+?);/
+      hash['Redox potential'] = $1
+    end
+    if data =~ /Temperature dependence: (.+?);/
+      hash['Temperature dependence'] = $1
+    end
+    hash
+  end
+  private :cc_biophysiochemical_properties
+
+
+  def cc_caution(data)
+    data.join('')
+  end
+  private :cc_caution
+
+
+  # returns conteins in a line of the CC INTERACTION section.
+  #
+  #   CC       P46527:CDKN1B; NbExp=1; IntAct=EBI-359815, EBI-519280;
+  def cc_interaction(data)
+    str = data.join('')
+    it = str.scan(/(.+?); NbExp=(.+?); IntAct=(.+?);/)
+    it.map {|ent|
+      ent.map! {|x| x.strip }
+      if ent[0] =~ /^(.+):(.+)/
+        spac = $1
+        spid = $2.split(' ')[0]
+        optid = nil
+      elsif ent[0] =~ /Self/
+        spac = self.entry_id
+        spid = self.entry_id
+        optid = nil
+      end
+      if ent[0] =~ /^.+:.+ (.+)/
+        optid = $1
+      end
+
+      {'SP_Ac' => spac,
+       'identifier' => spid,
+       'NbExp' => ent[1],
+       'IntAct' => ent[2].split(', '),
+       'optional_identifier' => optid}
+    }
+  end
+  private :cc_interaction
+
+
+  def cc_mass_spectrometry(data)
+    # MASS SPECTROMETRY: MW=XXX[; MW_ERR=XX][; METHOD=XX][;RANGE=XX-XX].
+    return data unless data
+
+    data.map { |m|
+      mass = {'MW' => nil, 'MW_ERR' => nil, 'METHOD' => nil, 'RANGE' => nil,
+              'NOTE' => nil}
+      m.sub(/.$/,'').split(/;/).each do |line|
+        case line
+        when /MW=(.+)/
+          mass['MW'] = $1
+        when /MW_ERR=(.+)/
+          mass['MW_ERR'] = $1
+        when /METHOD=(.+)/
+          mass['METHOD'] = $1
+        when /RANGE=(\d+-\d+)/ 
+          mass['RANGE'] = $1          # RANGE class ? 
+        when /NOTE=(.+)/
+          mass['NOTE'] = $1
+        end 
+      end
+      mass
+    }
+  end
+  private :cc_mass_spectrometry
+
+
+  def cc_pathway(data)
+    data.map {|x| x.sub(/\.$/, '') }.map {|x|
+      x.split(/; | and |: /)
+    }[0]
+  end
+  private :cc_pathway
+
+
+  def cc_rna_editing(data)
+    data = data.join('')
+    entry = {'Modified_positions' => [], 'Note' => ""}
+    if data =~ /Modified_positions=(.+?)(\.|;)/
+      entry['Modified_positions'] = $1.sub(/\.$/, '').split(', ')
+    else
+      raise ArgumentError, "Invarid CC RNA Editing lines (#{self.entry_id}):#{$!}\n#{get('CC')}"
+    end
+    if data =~ /Note=(.+)/
+      entry['Note'] = $1
+    end
+    entry
+  end
+  private :cc_rna_editing
+
+
+  def cc_subcellular_location(data)
+    data.map {|x| 
+      x.split('. ').map {|y| 
+        y.split('; ').map {|z| 
+          z.sub(/\.$/, '') 
+        } 
+      } 
+    }[0]
+  end
+  private :cc_subcellular_location
+
+
+  #--
+  # Since UniProtKB release 12.2 of 11-Sep-2007:
+  # CC   -!- WEB RESOURCE: Name=ResourceName[; Note=FreeText][; URL=WWWAddress].  # Old format:
+  # CC   -!- WEB RESOURCE: NAME=ResourceName[; NOTE=FreeText][; URL=WWWAddress].
+  #++
+
+  def cc_web_resource(data)
+    data.map {|x|
+      entry = {'Name' => nil, 'Note' => nil, 'URL' => nil}
+      x.split(';').each do |y|
+        case y
+        when /(Name|Note)\=(.+)/
+          key = $1
+          val = $2.strip
+          entry[key] = val
+        when /(NAME|NOTE)\=(.+)/
+          key = $1.downcase.capitalize
+          val = $2.strip
+          entry[key] = val
+        when /URL\=\"(.+)\"/
+          entry['URL'] = $1.strip
+        end
+      end
+      entry
+    }
+  end
+  private :cc_web_resource
+
+  # returns databases cross-references in the DR lines.
+  # * Bio::UniProtKB#dr  -> Hash w/in Array
+  #
+  # === DR Line; defabases cross-reference (>=0)
+  #    DR  database_identifier; primary_identifier; secondary_identifier.
+  #  a cross_ref pre one line
+  @@dr_database_identifier = ['EMBL','CARBBANK','DICTYDB','ECO2DBASE',
+    'ECOGENE',
+    'FLYBASE','GCRDB','HIV','HSC-2DPAGE','HSSP','INTERPRO','MAIZEDB',
+    'MAIZE-2DPAGE','MENDEL','MGD''MIM','PDB','PFAM','PIR','PRINTS',
+    'PROSITE','REBASE','AARHUS/GHENT-2DPAGE','SGD','STYGENE','SUBTILIST',
+    'SWISS-2DPAGE','TIGR','TRANSFAC','TUBERCULIST','WORMPEP','YEPD','ZFIN']
+
+  # Backup Bio::EMBLDB#dr as embl_dr
+  alias :embl_dr :dr 
+
+  # Bio::UniProtKB#dr
+  def dr(key = nil)
+    unless key
+      embl_dr
+    else
+      (embl_dr[key] or []).map {|x|
+        {'Accession' => x[0],
+         'Version' => x[1],
+         ' ' => x[2],
+         'Molecular Type' => x[3]}
+      }
+    end
+  end
+
+
+  # Bio::EMBLDB::Common#kw - Array
+  #                    #keywords  -> Array
+  #
+  # KW Line; keyword (>=1)
+  # KW   [Keyword;]+
+
+
+  # returns contents in the feature table.
+  #
+  # == Examples
+  #
+  #  sp = Bio::UniProtKB.new(entry)
+  #  ft = sp.ft
+  #  ft.class #=> Hash
+  #  ft.keys.each do |feature_key|
+  #    ft[feature_key].each do |feature|
+  #      feature['From'] #=> '1'
+  #      feature['To']   #=> '21'
+  #      feature['Description'] #=> ''
+  #      feature['FTId'] #=> ''
+  #      feature['diff'] #=> []
+  #      feature['original'] #=> [feature_key, '1', '21', '', '']
+  #    end
+  #  end
+  #
+  # * Bio::UniProtKB#ft -> Hash
+  #    {FEATURE_KEY => [{'From' => int, 'To' => int, 
+  #                      'Description' => aStr, 'FTId' => aStr,
+  #                      'diff' => [original_residues, changed_residues],
+  #                      'original' => aAry }],...}
+  #
+  # returns an Array of the information about the feature_name in the feature table.
+  # * Bio::UniProtKB#ft(feature_name) -> Array of Hash
+  #    [{'From' => str, 'To' => str, 'Description' => str, 'FTId' => str},...]
+  #
+  # == FT Line; feature table data (>=0, optional)
+  #
+  #   Col     Data item
+  #   -----   -----------------
+  #    1- 2   FT
+  #    6-13   Feature name 
+  #   15-20   `FROM' endpoint
+  #   22-27   `TO' endpoint
+  #   35-75   Description (>=0 per key)
+  #   -----   -----------------
+  #
+  # Note: 'FROM' and 'TO' endopoints are allowed to use non-numerial charactors 
+  # including '<', '>' or '?'. (c.f. '<1', '?42')
+  #
+  # See also http://www.expasy.org/sprot/userman.html#FT_line
+  #
+  def ft(feature_key = nil)
+    return ft[feature_key] if feature_key
+    return @data['FT'] if @data['FT']
+
+    table = []
+    begin
+      get('FT').split("\n").each do |line|
+        if line =~ /^FT   \w/
+          feature = line.chomp.ljust(74)
+          table << [feature[ 5..12].strip,   # Feature Name
+                    feature[14..19].strip,   # From
+                    feature[21..26].strip,   # To
+                    feature[34..74].strip ]  # Description
+        else
+          table.last << line.chomp.sub!(/^FT +/, '')
+        end
+      end
+
+      # Joining Description lines
+      table = table.map { |feature| 
+        ftid = feature.pop if feature.last =~ /FTId=/
+        if feature.size > 4
+          feature = [feature[0], 
+                     feature[1], 
+                     feature[2], 
+                     feature[3, feature.size - 3].join(" ")]
+        end
+        feature << if ftid then ftid else '' end
+      }
+
+      hash = {}
+      table.each do |feature|
+        hash[feature[0]] = [] unless hash[feature[0]]
+        hash[feature[0]] << {
+          # Removing '<', '>' or '?' in FROM/TO endopoint.
+          'From' => feature[1].sub(/\D/, '').to_i,  
+          'To'   => feature[2].sub(/\D/, '').to_i, 
+          'Description' => feature[3], 
+          'FTId' => feature[4].to_s.sub(/\/FTId=/, '').sub(/\.$/, ''),
+          'diff' => [],
+          'original' => feature
+        }
+
+        case feature[0]
+        when 'VARSPLIC', 'VARIANT', 'VAR_SEQ', 'CONFLICT'
+          case hash[feature[0]].last['Description']
+          when /(\w[\w ]*\w*) - ?> (\w[\w ]*\w*)/
+            original_res = $1
+            changed_res = $2
+            original_res = original_res.gsub(/ /,'').strip
+            chenged_res = changed_res.gsub(/ /,'').strip
+          when /Missing/i
+            original_res = seq.subseq(hash[feature[0]].last['From'],
+                                      hash[feature[0]].last['To'])
+            changed_res = ''
+          end
+          hash[feature[0]].last['diff'] = [original_res, chenged_res]
+        end
+      end
+    rescue
+      raise "Invalid FT Lines(#{$!}) in #{entry_id}:, \n'#{self.get('FT')}'\n"
+    end
+
+    @data['FT'] = hash
+  end
+
+
+
+  # returns a Hash of conteins in the SQ lines.
+  # * Bio::UniProtKBL#sq  -> hsh
+  #
+  # returns a value of a key given in the SQ lines.
+  # * Bio::UniProtKBL#sq(key)  -> int or str
+  # * Keys: ['MW', 'mw', 'molecular', 'weight', 'aalen', 'len', 'length', 
+  #          'CRC64']
+  #
+  # === SQ Line; sequence header (1/entry)
+  #    SQ   SEQUENCE   233 AA;  25630 MW;  146A1B48A1475C86 CRC64;
+  #    SQ   SEQUENCE  \d+ AA; \d+ MW;  [0-9A-Z]+ CRC64;
+  #
+  # MW, Dalton unit.
+  # CRC64 (64-bit Cyclic Redundancy Check, ISO 3309).
+  def sq(key = nil)
+    unless @data['SQ']
+      if fetch('SQ') =~ /(\d+) AA\; (\d+) MW; (.+) CRC64;/
+        @data['SQ'] = { 'aalen' => $1.to_i, 'MW' => $2.to_i, 'CRC64' => $3 }
+      else
+        raise "Invalid SQ Line: \n'#{fetch('SQ')}'"
+      end
+    end
+
+    if key
+      case key
+      when /mw/, /molecular/, /weight/
+        @data['SQ']['MW']
+      when /len/, /length/, /AA/
+        @data['SQ']['aalen']
+      else
+        @data['SQ'][key]
+      end
+    else 
+      @data['SQ']
+    end
+  end
+
+
+  # returns a Bio::Sequence::AA of the amino acid sequence.
+  # * Bio::UniProtKB#seq -> Bio::Sequence::AA
+  #
+  # blank Line; sequence data (>=1)
+  def seq
+    unless @data['']
+      @data[''] = Sequence::AA.new( fetch('').gsub(/ |\d+/,'') )
+    end
+    return @data['']
+  end
+  alias aaseq seq
+
+end # class UniProtKB
+
+end # module Bio
+
+
+
+=begin
+
+= Bio::UniProtKB < Bio::DB
+
+Class for a entry in the SWISS-PROT/TrEMBL database.
+
+  * ((<URL:http://www.ebi.ac.uk/swissprot/>))
+  * ((<URL:http://www.ebi.ac.uk/trembl/>))
+  * ((<URL:http://www.ebi.ac.uk/sprot/userman.html>))
+  
+
+--- Bio::UniProtKB.new(a_sp_entry)
+
+=== ID line (Identification)
+
+--- Bio::UniProtKB#id_line -> {'ENTRY_NAME' => str, 'DATA_CLASS' => str,
+                          'MOLECULE_TYPE' => str, 'SEQUENCE_LENGTH' => int }  
+--- Bio::UniProtKB#id_line(key) -> str
+
+       key = (ENTRY_NAME|MOLECULE_TYPE|DATA_CLASS|SEQUENCE_LENGTH)
+
+--- Bio::UniProtKB#entry_id -> str
+--- Bio::UniProtKB#molecule -> str
+--- Bio::UniProtKB#sequence_length -> int
+    
+
+=== AC lines (Accession number)
+
+--- Bio::UniProtKB#ac -> ary
+--- Bio::UniProtKB#accessions -> ary
+--- Bio::UniProtKB#accession -> accessions.first
+
+ 
+=== GN line (Gene name(s))
+
+--- Bio::UniProtKB#gn -> [ary, ...] or [{:name => str, :synonyms => [], :loci => [], :orfs => []}]
+--- Bio::UniProtKB#gene_name -> str
+--- Bio::UniProtKB#gene_names -> [str] or [str]
+
+
+=== DT lines (Date) 
+
+--- Bio::UniProtKB#dt -> {'created' => str, 'sequence' => str, 'annotation' => str}
+--- Bio::UniProtKB#dt(key) -> str
+
+      key := (created|annotation|sequence)
+
+
+=== DE lines (Description)
+
+--- Bio::UniProtKB#de -> str
+             #definition -> str
+
+--- Bio::UniProtKB#protein_name
+
+      Returns the proposed official name of the protein
+
+
+--- Bio::UniProtKB#synonyms
+
+      Returns an array of synonyms (unofficial names)
+
+=== KW lines (Keyword)
+
+--- Bio::UniProtKB#kw -> ary
+
+=== OS lines (Organism species)
+
+--- Bio::UniProtKB#os -> [{'name' => str, 'os' => str}, ...]
+
+=== OC lines (organism classification)
+
+--- Bio::UniProtKB#oc -> ary
+
+=== OG line (Organella)
+
+--- Bio::UniProtKB#og -> ary
+
+=== OX line (Organism taxonomy cross-reference)
+
+--- Bio::UniProtKB#ox -> {'NCBI_TaxID' => [], ...}
+
+=== RN RC RP RX RA RT RL RG lines (Reference)  
+
+--- Bio::UniProtKB#ref -> [{'RN' => int, 'RP' => str, 'RC' => str, 'RX' => str, ''RT' => str, 'RL' => str, 'RA' => str, 'RC' => str, 'RG' => str},...]
+
+=== DR lines (Database cross-reference)
+
+--- Bio::UniProtKB#dr -> {'EMBL' => ary, ...}
+
+=== FT lines (Feature table data)
+
+--- Bio::UniProtKB#ft -> hsh
+
+=== SQ lines (Sequence header and data)
+
+--- Bio::UniProtKB#sq -> {'CRC64' => str, 'MW' => int, 'aalen' => int}
+--- Bio::UniProtKB#sq(key) -> int or str
+
+          key := (aalen|MW|CRC64)
+
+--- Bio::UniProtKB#seq -> Bio::Sequece::AA
+                  #aaseq -> Bio::Sequece::AA
+
+=end
+
+  #      Content                      Occurrence in an entry
+  # ---- ---------------------------  --------------------------------
+  # ID - identification               (begins each entry; 1 per entry)
+  # AC - accession number(s)          (>=1 per entry)
+  # DT - date                         (3 per entry)
+  # DE - description                  (>=1 per entry)
+  # GN - gene name(s)                 (>=0 per entry; optional)
+  # OS - organism species             (>=1 per entry)
+  # OG - organelle                    (0 or 1 per entry; optional)
+  # OC - organism classification      (>=1 per entry)
+  # OX - organism taxonomy x-ref      (>=1 per entry)
+  # OH - Organism Host
+  # RN - reference number             (>=1 per entry)
+  # RP - reference positions          (>=1 per entry)
+  # RC - reference comment(s)         (>=0 per entry; optional)
+  # RX - reference cross-reference(s) (>=0 per entry; optional)
+  # RA - reference author(s)          (>=1 per entry)
+  # RT - reference title              (>=0 per entry; optional)
+  # RL - reference location           (>=1 per entry)
+  # RG - reference group(s)
+  # CC - comments or notes            (>=0 per entry; optional)
+  # DR - database cross-references    (>=0 per entry; optional)
+  # KW - keywords                     (>=1 per entry)
+  # FT - feature table data           (>=0 per entry; optional)
+  # SQ - sequence header              (1 per entry)
+  #    - (blanks) The sequence data   (>=1 per entry)
+  # // - termination line             (ends each entry; 1 per entry)
+  # ---- ---------------------------  --------------------------------
+
+