bio 0.7.0

data/lib/bio/db/aaindex.rb

@@ -0,0 +1,280 @@
+#
+# = bio/db/aaindex.rb - AAindex database class
+#
+# Copyright::  Copyright (C) 2001 KAWASHIMA Shuichi <s@bioruby.org>
+# License::    LGPL
+#
+# $Id: aaindex.rb,v 1.16 2005/12/18 15:58:41 k Exp $
+#
+# == Description
+# Classes for Amino Acid Index Database (AAindex and AAindex2).
+# * AAindex Manual: http://www.genome.jp/dbget-bin/show_man?aaindex
+#
+# == Examples
+#  aax1 = Bio::AAindex1.new("PRAM900102.aaindex1")
+#  aax1.entry_id
+#  aax1.index
+#
+#  aax2 = Bio::AAindex2.new("HENS920102.aaindex2")
+#  aax2.entry_id
+#  aax2.matrix
+#  aax2.matrix[2,2]
+#
+# == References
+# * http://www.genome.jp/aaindex/
+#
+#--
+#
+#  This library is free software; you can redistribute it and/or
+#  modify it under the terms of the GNU Lesser General Public
+#  License as published by the Free Software Foundation; either
+#  version 2 of the License, or (at your option) any later version.
+#
+#  This library is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+#  Lesser General Public License for more details.
+#
+#  You should have received a copy of the GNU Lesser General Public
+#  License along with this library; if not, write to the Free Software
+#  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307  USA
+#
+#++
+#
+
+require "bio/db"
+require "matrix"
+
+module Bio
+
+  class AAindex < KEGGDB
+
+    # Delimiter
+    DELIMITER ="\n//\n"
+
+    # Delimiter
+    RS = DELIMITER
+
+    # Bio::DB API
+    TAGSIZE = 2
+
+
+    def initialize(entry)
+      super(entry, TAGSIZE)
+    end
+
+    # Returns
+    def entry_id
+      field_fetch('H')
+    end
+
+    # Returns
+    def definition
+      field_fetch('D')
+    end
+
+    # Returns
+    def dblinks
+      field_fetch('R')
+    end
+
+    # Returns
+    def author
+      field_fetch('A')
+    end
+
+    # Returns
+    def title
+      field_fetch('T')
+    end
+
+    # Returns
+    def journal
+      field_fetch('J')
+    end
+
+    # Returns
+    def comment
+      get('*')
+    end
+
+  end
+
+
+  class AAindex1 < AAindex
+
+
+    def initialize(entry)
+      super(entry)
+    end
+
+    # Returns
+    def correlation_coefficient
+      field_fetch('C')
+    end
+
+    # Returns
+    def index(type = :float)
+      aa = %w( A R N D C Q E G H I L K M F P S T W Y V )
+      values = field_fetch('I', 1).split(' ')
+
+      if values.size != 20
+        raise "Invalid format in #{entry_id} : #{values.inspect}"
+      end
+
+      if type == :zscore and values.size > 0
+        sum = 0.0
+        values.each do |a|
+          sum += a.to_f
+        end
+        mean = sum / values.size # / 20
+        var = 0.0
+        values.each do |a|
+          var += (a.to_f - mean) ** 2
+        end
+        sd = Math.sqrt(var)
+      end
+
+      if type == :integer
+        figure = 0
+        values.each do |a|
+          figure = [ figure, a[/\..*/].length - 1 ].max
+        end
+      end
+
+      hash = {}
+
+      aa.each_with_index do |a, i|
+        case type
+        when :string
+          hash[a] = values[i]
+        when :float
+          hash[a] = values[i].to_f
+        when :zscore
+          hash[a] = (values[i].to_f - mean) / sd
+        when :integer
+          hash[a] = (values[i].to_f * 10 ** figure).to_i
+        end
+      end
+      return hash
+    end
+
+  end
+
+
+  class AAindex2 < AAindex
+
+
+    def initialize(entry)
+      super(entry)
+    end
+
+    # Returns
+    def rows
+      label_data
+      @rows
+    end
+
+    # Returns
+    def cols
+      label_data
+      @cols
+    end
+
+    # Returns
+    def matrix
+      ma = Array.new
+
+      data = label_data
+      data.each_line do |line|
+        list = line.strip.split(/\s+/).map{|x| x.to_f}
+        ma.push(list)
+      end
+
+      Matrix[*ma]
+    end
+
+    # Returns
+    def old_matrix # for AAindex <= ver 5.0
+
+      @aa = {} # used to determine row/column of the aa
+      attr_reader :aa
+
+      field = field_fetch('I')
+
+      case field
+      when / (ARNDCQEGHILKMFPSTWYV)\s+(.*)/ # 20x19/2 matrix
+        aalist = $1
+        values = $2.split(/\s+/)
+
+        0.upto(aalist.length - 1) do |i|
+          @aa[aalist[i].chr] = i
+        end
+
+        ma = Array.new
+        20.times do
+          ma.push(Array.new(20)) # 2D array of 20x(20)
+        end
+
+        for i in 0 .. 19 do
+          for j in i .. 19 do
+            ma[i][j] = values[i + j*(j+1)/2].to_f
+            ma[j][i] = ma[i][j]
+          end
+        end
+        Matrix[*ma]
+
+      when / -ARNDCQEGHILKMFPSTWYV / # 21x20/2 matrix (with gap)
+        raise NotImplementedError
+      when / ACDEFGHIKLMNPQRSTVWYJ- / # 21x21 matrix (with gap)
+        raise NotImplementedError
+      end
+    end
+
+    private
+
+    def label_data
+      label, data = get('M').split("\n", 2)
+      if /M rows = (\S+), cols = (\S+)/.match(label)
+        rows, cols = $1, $2
+        @rows = rows.split('')
+        @cols = cols.split('')
+      end
+      return data
+    end
+
+  end
+
+end
+
+
+if __FILE__ == $0
+  require 'bio/io/fetch'
+
+  puts "### AAindex1 (PRAM900102)"
+  aax1 = Bio::AAindex1.new(Bio::Fetch.query('aaindex', 'PRAM900102', 'raw'))
+  p aax1.entry_id
+  p aax1.definition
+  p aax1.dblinks
+  p aax1.author
+  p aax1.title
+  p aax1.journal
+  p aax1.correlation_coefficient
+  p aax1.index
+  puts "### AAindex2 (HENS920102)"
+  aax2 = Bio::AAindex2.new(Bio::Fetch.query('aaindex', 'HENS920102', 'raw'))
+  p aax2.entry_id
+  p aax2.definition
+  p aax2.dblinks
+  p aax2.author
+  p aax2.title
+  p aax2.journal
+  p aax2.rows
+  p aax2.cols
+  p aax2.matrix
+  p aax2.matrix[2,2]
+  p aax2.matrix.determinant
+  p aax2.matrix.rank
+  p aax2.matrix.transpose
+end
+

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

@@ -0,0 +1,332 @@
+#
+# = bio/db/embl.rb - Common methods for EMBL style database classes
+#
+# Copyright::   Copyright (C) 2001-2005 Mitsuteru C. Nakao <n@bioruby.org>
+# License::     LGPL
+#
+# $Id: common.rb,v 1.8 2005/11/02 07:30:14 nakao Exp $
+#
+# == EMBL style databases class
+#
+# This module defines a common framework among EMBL, SWISS-PROT, TrEMBL.
+# For more details, see the documentations in each embl/*.rb libraries.
+#
+# EMBL style format:
+#     ID - identification             (begins each entry; 1 per entry)
+#     AC - accession number           (>=1 per entry)
+#     SV - sequence version           (1 per entry)
+#     DT - date                       (2 per entry)
+#     DE - description                (>=1 per entry)
+#     KW - keyword                    (>=1 per entry)
+#     OS - organism species           (>=1 per entry)
+#     OC - organism classification    (>=1 per entry)
+#     OG - organelle                  (0 or 1 per entry)
+#     RN - reference number           (>=1 per entry)
+#     RC - reference comment          (>=0 per entry)
+#     RP - reference positions        (>=1 per entry)
+#     RX - reference cross-reference  (>=0 per entry)
+#     RA - reference author(s)        (>=1 per entry)
+#     RG - reference group            (>=0 per entry)
+#     RT - reference title            (>=1 per entry)
+#     RL - reference location         (>=1 per entry)
+#     DR - database cross-reference   (>=0 per entry)
+#     FH - feature table header       (0 or 2 per entry)
+#     FT - feature table data         (>=0 per entry)
+#     CC - comments or notes          (>=0 per entry)
+#     XX - spacer line                (many per entry)
+#     SQ - sequence header            (1 per entry)
+#     bb - (blanks) sequence data     (>=1 per entry)
+#     // - termination line           (ends each entry; 1 per entry)
+#
+#
+# == Example
+# 
+#  require 'bio/db/embl/common'
+#  module Bio
+#    class NEWDB < EMBLDB
+#      include Bio::EMBLDB::Common
+#    end
+#  end
+#
+#--
+#
+#  This library is free software; you can redistribute it and/or
+#  modify it under the terms of the GNU Lesser General Public
+#  License as published by the Free Software Foundation; either
+#  version 2 of the License, or (at your option) any later version.
+#
+#  This library is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+#  Lesser General Public License for more details.
+#
+#  You should have received a copy of the GNU Lesser General Public
+#  License along with this library; if not, write to the Free Software
+#  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307  USA
+#
+#++
+#
+
+require 'bio/db'
+require 'bio/reference'
+
+module Bio
+class EMBLDB
+module Common
+
+  DELIMITER = "\n//\n"
+  RS = DELIMITER
+  TAGSIZE = 5
+
+  def initialize(entry)
+    super(entry, TAGSIZE)
+  end
+
+  # returns a Array of accession numbers in the AC lines.
+  #
+  # AC Line
+  #   "AC   A12345; B23456;"
+  #   AC [AC1;]+
+  #
+  # Accession numbers format:
+  #   1       2     3          4          5          6
+  #   [O,P,Q] [0-9] [A-Z, 0-9] [A-Z, 0-9] [A-Z, 0-9] [0-9]
+  def ac
+    unless @data['AC']
+      tmp = Array.new
+      field_fetch('AC').split(/ /).each do |e|
+        tmp.push(e.sub(/;/,''))
+      end
+      @data['AC'] = tmp
+    end
+    @data['AC']
+  end
+  alias accessions ac
+
+
+  # returns the first accession number in the AC lines
+  def accession
+    ac[0]
+  end
+
+
+  # returns a String int the DE line.
+  #
+  # DE Line
+  def de
+    unless @data['DE']
+      @data['DE'] = fetch('DE')
+    end
+    @data['DE']
+  end
+  alias description de
+  alias definition de   # API
+  
+
+
+  # returns contents in the OS line.
+  # * Bio::EMBLDB#os  -> Array of <OS Hash>
+  # where <OS Hash> is:
+  #  [{'name'=>'Human', 'os'=>'Homo sapiens'}, 
+  #   {'name'=>'Rat', 'os'=>'Rattus norveticus'}]
+  # * Bio::SPTR#os[0]['name'] => "Human"
+  # * Bio::SPTR#os[0] => {'name'=>"Human", 'os'=>'Homo sapiens'}
+  # * Bio::STPR#os(0) => "Homo sapiens (Human)"
+  #
+  # OS Line; organism species (>=1)
+  #   "OS   Trifolium repens (white clover)"
+  #
+  #   OS   Genus species (name).
+  #   OS   Genus species (name0) (name1).
+  #   OS   Genus species (name0) (name1).
+  #   OS   Genus species (name0), G s0 (name0), and G s (name1).
+  def os(num = nil)
+    unless @data['OS']
+      os = Array.new
+      fetch('OS').split(/, and|, /).each do |tmp|
+        if tmp =~ /([A-Z][a-z]* *[\w\d \:\'\+\-]+[\w\d])/
+          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)"
+      "#{@data['OS'][num]['os']} {#data['OS'][num]['name']"
+    end
+    @data['OS']
+  end
+
+
+  # returns contents in the OG line.
+  # * Bio::EMBLDB::Common#og  -> [ <ogranella String>* ]
+  #
+  # OG Line; organella (0 or 1/entry)
+  #  OG   Plastid; Chloroplast.
+  #  OG   Mitochondrion.
+  #  OG   Plasmid sym pNGR234a.
+  #  OG   Plastid; Cyanelle.
+  #  OG   Plasmid pSymA (megaplasmid 1).
+  #  OG   Plasmid pNRC100, Plasmid pNRC200, and Plasmid pHH1.
+  def og
+    unless @data['OG']
+      og = Array.new
+      if get('OG').size > 0
+        ogstr = fetch('OG')
+        ogstr.sub!(/\.$/,'')
+        ogstr.sub!(/ and/,'')
+        ogstr.sub!(/;/, ',')
+        ogstr.split(',').each do |tmp|
+          og.push(tmp.strip)
+        end
+      end
+      @data['OG'] = og
+    end
+    @data['OG']
+  end
+  
+
+  # returns contents in the OC line.
+  # * Bio::EMBLDB::Common#oc  -> [ <organism class String>* ]
+  # OC Line; organism classification (>=1)
+  #  OC   Eukaryota; Alveolata; Apicomplexa; Piroplasmida; Theileriidae;
+  #  OC   Theileria.
+  def oc
+    unless @data['OC']
+      begin
+        @data['OC'] = fetch('OC').sub(/.$/,'').split(/;/).map {|e|
+          e.strip 
+        }
+      rescue NameError
+        nil
+      end
+    end
+    @data['OC']
+  end
+
+  # returns keywords in the KW line.
+  # * Bio::EMBLDB::Common#kw  -> [ <keyword>* ]
+  # KW Line; keyword (>=1)
+  #  KW   [Keyword;]+
+  def kw
+    unless @data['KW']
+      if get('KW').size > 0
+        tmp = fetch('KW').sub(/.$/,'')
+        @data['KW'] = tmp.split(/;/).map {|e| e.strip }
+      else
+        @data['KW'] = []
+      end
+    end
+    @data['KW']
+  end
+  alias keywords kw
+
+
+  # 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']
+      ary = Array.new
+      get('R').split(/\nRN   /).each do |str|
+        raw = {'RN' => '', 'RC' => '', 'RP' => '', 'RX' => '', 
+               'RA' => '', 'RT' => '', 'RL' => '', 'RG' => ''}
+        str = 'RN   ' + str unless /^RN   / =~ str
+        str.split("\n").each do |line|
+          if /^(R[NPXARLCTG])   (.+)/ =~ line
+            raw[$1] += $2 + ' '
+          else
+            raise "Invalid format in R lines, \n[#{line}]\n"
+          end
+        end
+        raw.each_value {|v| 
+          v.strip! 
+          v.sub!(/^"/,'')
+          v.sub!(/;$/,'')
+          v.sub!(/"$/,'')
+        }
+        ary.push(raw)
+      end
+      @data['R'] = ary
+    end
+    @data['R']
+  end
+
+  # 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']    = $5
+            else
+              hash['journal'] = value
+            end
+          when 'RX'  # PUBMED, MEDLINE
+            value.split('.').each {|item|
+              tag, xref = item.split(/; /).map {|i| i.strip }
+              hash[ tag.downcase ]  = xref
+            }
+          end
+        }
+        Reference.new(hash)
+      }
+      @data['references'] = References.new(ary)
+    end
+    @data['references']
+  end
+
+
+  # returns contents in the DR line.
+  # * Bio::EMBLDB::Common#dr  -> [ <Database cross-reference Hash>* ]
+  # where <Database cross-reference Hash> is:
+  # * Bio::EMBLDB::Common#dr {|k,v| } 
+  # 
+  # DR Line; defabases cross-reference (>=0)
+  # a cross_ref pre one line
+  #  "DR  database_identifier; primary_identifier; secondary_identifier."
+  def dr
+    unless @data['DR']
+      tmp = Hash.new
+      self.get('DR').split(/\n/).each do |db|
+        a = db.sub(/^DR   /,'').sub(/.$/,'').strip.split(/;[ ]/)
+        dbname = a.shift
+        tmp[dbname] = Array.new unless tmp[dbname]
+        tmp[dbname].push(a)
+      end
+      @data['DR'] = tmp
+    end
+    if block_given?
+      @data['DR'].each do |k,v|
+        yield(k, v)
+      end
+    else
+      @data['DR']
+    end
+  end
+
+end # module Common
+end # class EMBLDB
+end # module Bio
+