bio 0.7.0

data/lib/bio/appl/psort/report.rb

@@ -0,0 +1,473 @@
+#
+# = bio/appl/psort/report.rb - PSORT systems report classes
+#
+# Copyright::   Copyright (C) 2003 Mitsuteru C. Nakao <n@bioruby.org>
+# License::     LGPL
+#
+#  $Id: report.rb,v 1.12 2005/11/03 10:50:58 nakao Exp $
+#
+# == A Report classes for PSORT Systems
+# 
+#--
+#
+#  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/sequence'
+require 'bio/appl/psort'
+
+
+module Bio
+
+  class PSORT
+
+    class PSORT1
+
+      # = Bio::PSORT::PSORT1::Report
+      # Parser class for PSORT1 output report.
+      # 
+      # == Example
+      class Report
+
+        # Returns aBio::PSORT::PSORT1::Report.
+        def self.parser(output_report)
+          self.default_parser(output_report)
+        end
+
+        # Returns aBio::PSORT::PSORT1::Report.
+        def self.default_parser(output_report)
+          rpt = self.new
+          rpt.raw = output_report
+          query_info = output_report.scan(/^Query Information\n\n(.+?)\n\n/m)[0][0].split(/\n/)
+          result_info = output_report.scan(/^Result Information\n\n(.+?)\n\n\*/m)[0][0]
+          step1 = output_report.scan(/^\*\*\* Reasoning Step: 1\n\n(.+?)\n\n/m)[0][0]
+          step2 = output_report.scan(/^\*\*\* Reasoning Step: 2\n\n(.+?)\n\n/m)[0][0]
+          final_result = output_report.scan(/\n\n----- Final Results -----\n\n(.+?)\n\n\n/m)[0][0]
+
+          rpt.entry_id = query_info[2].scan(/^>(\S+) */).to_s
+          rpt.origin   = query_info[0].scan(/ORIGIN (\w+)/).to_s
+          rpt.sequence = Bio::Sequence::AA.new(query_info[3..query_info.size].to_s)
+          # rpt.reasoning
+
+          rpt.final_result = final_result.split(/\n/).map {|x|
+            x = x.strip.split(/---/).map {|y| y.strip }
+            { 'prediction' => x[0], 
+              'certainty'  => x[1].scan(/Certainty= (\d\.\d{3})/).to_s,
+              'comment'    => x[1].scan(/\((\w+)\)/).to_s
+            }
+          }
+          return rpt
+        end
+
+        attr_accessor :entry_id
+        attr_accessor :origin
+        attr_accessor :title
+        attr_accessor :sequence
+        attr_accessor :result_info
+        attr_accessor :reasoning
+        attr_accessor :final_result
+        attr_accessor :raw
+
+
+
+        # Constructs aBio::PSORT::PSORT1::Report object.
+        def initialize(entry_id = '', origin = '', title = '', sequence = '',
+                       result_info = '', reasoning = {}, final_result = [])
+          @entry_id = entry_id
+          @origin = origin
+          @title = title
+          @sequence = sequence
+          @result_info = result_info
+          @reasoning = reasoning
+          @final_result = final_result
+          @raw = ''
+        end
+
+
+      end # class Report
+
+    end # class PSORT1
+
+
+
+    class PSORT2
+
+      # Subcellular localization name codes used by PSORT2 
+      SclNames = { 
+        'csk' => 'cytoskeletal',
+        'cyt' => 'cytoplasmic',
+        'nuc' => 'nuclear',
+        'mit' => 'mitochondrial',
+        'ves' => 'vesicles of secretory system',
+        'end' => 'endoplasmic reticulum',
+        'gol' => 'Golgi',
+        'vac' => 'vacuolar',
+        'pla' => 'plasma membrane',
+        'pox' => 'peroxisomal',
+        'exc' => 'extracellular, including cell wall',
+        '---' => 'other'
+      }
+    
+      # Feature name codes
+      Features = [
+        'psg',  # PSG: PSG score
+        'gvh',  # GvH: GvH score
+        'alm',  # ALOM: $xmax
+        'tms',  # ALOM: $count
+        'top',  # MTOP: Charge difference: $mtopscr
+        'mit',  # MITDISC: Score: $score
+        'mip',  # Gavel: motif at $isite
+        'nuc',  # NUCDISC: NLS Score: $score
+        'erl',  # KDEL: ($seg|none)
+        'erm',  # ER Membrane Retention Signals: ($cseg|none) $scr
+        'pox',  # SKL: ($pat|none) $scr
+        'px2',  # PTS2: (found|none)              ($#match < 0) ? 0 : ($#match+1);
+        'vac',  # VAC: (found|none)               ($#match < 0) ? 0 : ($#match+1);
+        'rnp',  # RNA-binding motif: (found|none) ($#match < 0) ? 0 : ($#match+1);
+        'act',  # Actinin-type actin-binding motif: (found|none)  $hit
+        'caa',  # Prenylation motif: (2|1|0) CaaX,CXC,CC,nil
+        'yqr',  # memYQRL: (found|none) $scr
+        'tyr',  # Tyrosines in the tail: (none|\S+[,])  
+                # 10 * scalar(@ylist) / ($end - $start + 1);
+        'leu',  # Dileucine motif in the tail: (none|found) $scr
+        'gpi',  # >>> Seem to be GPI anchored
+        'myr',  # NMYR: (none|\w) $scr
+        'dna',  # checking 63 PROSITE DNA binding motifs:              $hit
+        'rib',  # checking 71 PROSITE ribosomal protein motifs:        $hit
+        'bac',  # checking 33 PROSITE prokaryotic DNA binding motifs:  $hit
+        'm1a',  # $mtype eq '1a'  
+        'm1b',  # $mtype eq '1b'
+        'm2',   # $mtype eq '2 '
+        'mNt',  # $mtype eq 'Nt'
+        'm3a',  # $mtype eq '3a' 
+        'm3b',  # $mtype eq '3b'  
+        'm_',   # $mtype eq '__'  tms == 0
+        'ncn',  # NNCN: ($NetOutput[1] > $NetOutput[0]) ? $output : (-$output);
+        'lps',  # COIL: $count
+        'len'   # $leng
+      ]
+      
+      # Feature name codes (long version).
+      FeaturesLong = {
+        'psg' => 'PSG',  
+        'gvh' => 'GvH',
+        'tms' => 'ALOM',
+        'alm' => 'ALOM',
+        'top' => 'MTOP',
+        'mit' => 'MITDISC',
+        'mip' => 'Gavel',
+        'nuc' => 'NUCDISC',
+        'erl' => 'KDEL',
+        'erm' => 'ER Membrane Retention Signals',
+        'pox' => 'SKL',
+        'px2' => 'PTS2', 
+        'vac' => 'VAC', 
+        'rnp' => 'RNA-binding motif',
+        'act' => 'Actinin-type actin-binding motif',
+        'caa' => 'Prenylation motif',
+        'yqr' => 'memYQRL', 
+        'tyr' => 'Tyrosines in the tail',
+        'leu' => 'Dileucine motif in the tail',
+        'gpi' => '>>> Seems to be GPI anchored',
+        'myr' => 'NMYR', 
+        'dna' => 'checking 63 PROSITE DNA binding motifs',
+        'rib' => 'checking 71 PROSITE ribosomal protein motifs',
+        'bac' => 'ochecking 33 PROSITE prokaryotic DNA binding motifs:', 
+        'm1a' => '', 
+        'm1b' => '', 
+        'm2'  => '', 
+        'mNt' => '', 
+        'm3a' => '', 
+        'm3b' => '', 
+        'm_'  => '', 
+        'ncn' => 'NNCN', 
+        'lps' => 'COIL',
+        'len' => 'AA'       # length of input sequence
+      }
+
+      # = Bio::PSORT::PSORT2::Report
+      # Report parser classe for PSORT II(PSORT2).
+      # == Example
+      class Report
+
+        # Report boundary string.
+        BOUNDARY  = '-' * 75
+
+
+        # Report delimiter.
+        RS = DELIMITER = "\)\n\n#{BOUNDARY}"
+
+        # entry_id of query sequence. 
+        attr_accessor :entry_id 
+
+        # Given subcellular localization (three letters code).
+        attr_accessor :scl
+
+        # Definition of query sequence.
+        attr_accessor :definition
+        
+        # Sequence of query sequence.
+        attr_accessor :seq
+
+        # k parameter of k-nearest neighbors classifier.
+        attr_accessor :k
+
+        # Feature vector used the kNN prediction.
+        attr_accessor :features
+
+        # Probability vector of kNN prediction.
+        attr_accessor :prob
+
+        # Predicted subcellular localization (three letters code).
+        attr_accessor :pred
+        
+        # Raw text of output report.
+        attr_accessor :raw
+
+
+        # Constructs aBio::PSORT::PSORT2::Report object.
+        def initialize(raw = '', entry_id = nil, scl = nil, definition = nil, 
+                       seq = nil, k = nil, features = {}, prob = {}, pred = nil)
+          @entry_id   = entry_id
+          @scl        = scl
+          @definition = definition
+          @seq        = seq
+          @features   = features
+          @prob       = prob
+          @pred       = pred
+          @k          = k
+          @raw        = raw
+        end
+
+        
+        # Parses output report with output format detection automatically.
+        def self.parser(str, entry_id)
+          case str
+          when /^ psg:/   # default report
+            self.default_parser(str, entry_id)
+          when /^PSG:/    # -v report
+            self.v_parser(str, entry_id)
+          when /: too short length /
+            self.too_short_parser(str, entry_id)
+          when /PSORT II server/
+            tmp = self.new(ent, entry_id)
+          else
+            raise ArgumentError, "invalid format\n[#{str}]"
+          end
+        end
+
+        # Parser for ``too short length'' report.
+        #
+        #  $id: too short length ($leng), skipped\n";
+        def self.too_short_parser(ent, entry_id = nil)
+          report = self.new(ent)
+          report.entry_id = entry_id
+          if ent =~ /^(.+)?: too short length/
+            report.entry_id = $1 unless report.entry_id
+            report.scl = '---'
+          end
+          report
+        end
+
+
+        # Parser for the default report format.
+        # ``psort report'' output.
+        def self.default_parser(ent, entry_id = nil)
+          report = self.new(ent, entry_id)
+          ent = ent.split(/\n\n/).map {|e| e.chomp }
+
+          report.set_header_line(ent[0])
+
+          # feature matrix
+          ent[1].gsub(/\n/,' ').strip.split(/  /).map {|fe|
+            pair = fe.split(/: /)
+            report.features[pair[0].strip] = pair[1].strip.to_f
+          }
+
+          report.prob = self.set_kNN_prob(ent[2])
+          report.set_prediction(ent[3])         
+
+          return report
+        end
+
+        # Returns header information.
+        def set_header_line(str)
+          str.sub!(/^-+\n/,'')
+          tmp = str.split(/\t| /)
+          @entry_id = tmp.shift.sub(/^-+/,'').strip unless @entry_id
+
+          case tmp.join(' ').chomp
+          when /\(\d+ aa\) (.+)$/
+            @definition = $1
+          else
+            @definition = tmp.join(' ').chomp
+          end
+          scl = @definition.split(' ')[0]
+
+          @scl = scl if SclNames.keys.index(scl)
+        end
+
+        # Returns @prob value.
+        def self.set_kNN_prob(str)
+          prob = Hash.new
+          Bio::PSORT::PSORT2::SclNames.keys.each {|a| 
+            prob.update( {a => 0.0} )
+          }
+          str.gsub(/\t/,'').split(/\n/).each {|a|
+            val,scl = a.strip.split(/ %: /)
+            key = Bio::PSORT::PSORT2::SclNames.index(scl)
+            prob[key] = val.to_f
+          }
+          return prob
+        end
+
+        # Returns @prob and @k values.
+        def set_prediction(str)
+          case str
+          when /prediction for (\S+?) is (\w{3}) \(k=(\d+)\)/
+            @entry_id ||= $1 unless @entry_id
+            @pred = $2
+            @k    = $3
+          else
+            raise ArgumentError, 
+              "Invalid format at(#{self.entry_id}):\n[#{str}]\n"
+          end
+        end
+
+
+        # Parser for the verbose output report format.
+        # ``psort -v report'' and WWW server output.
+        def self.v_parser(ent, entry_id = nil)
+          report = Bio::PSORT::PSORT2::Report.new(ent, entry_id)
+
+          ent = ent.split(/\n\n/).map {|e| e.chomp }
+          ent.each_with_index {|e, i|
+            unless /^(\w|-|\>|\t)/ =~ e
+              j = self.__send__(:search_j, i, ent)
+              ent[i - j] += e
+              ent[i] = nil
+            end
+            if /^none/ =~ e    # psort output bug
+              j = self.__send__(:search_j, i, ent)
+              ent[i - j] += e
+              ent[i] = nil
+            end
+          }
+          ent.compact!
+
+          if /^ PSORT II server/ =~ ent[0] # for WWW version
+            ent.shift 
+            delline = ''
+            ent.each {|e| delline = e if /^Results of Subprograms/ =~ e }
+            i = ent.index(delline)
+            ent.delete(delline)
+            ent.delete_at(i - 1)
+          end
+
+          report.set_header_line(ent.shift)  
+          report.seq = Bio::Sequence::AA.new(ent.shift)
+
+          fent, pent = self.divent(ent)
+          report.set_features(fent)          
+          report.prob = self.set_kNN_prob(pent[0].strip)  
+          report.set_prediction(pent[1].strip)
+
+          return report
+        end
+
+
+        # 
+        def self.search_j(i, ent)
+          j = 1
+          1.upto(ent.size) {|x|
+            if ent[i - x]
+              j = x
+              break
+            end
+          }
+          return j
+        end
+        private_class_method :search_j
+
+
+        # Divides entry body
+        def self.divent(entry)
+          boundary = entry.index(BOUNDARY)
+          return entry[0..(boundary - 1)], entry[(boundary + 2)..(entry.length)]
+        end
+
+        # Sets @features values.
+        def set_features(features_ary)
+          features_ary.each {|fent|
+            key = fent.split(/\:( |\n)/)[0].strip
+            self.features[key] = fent # unless /^\>/ =~ key
+          }
+          self.features['AA'] = self.seq.length
+        end
+        
+      end # class Report
+ 
+    end # class PSORT2      
+
+  end # class PSORT
+
+end # module Bio
+
+
+
+
+
+# testing code
+
+if __FILE__ == $0
+
+
+  while entry = $<.gets(Bio::PSORT::PSORT2::Report::DELIMITER)
+
+    puts "\n ==> a = Bio::PSORT::PSORT2::Report.parser(entry)"
+    a = Bio::PSORT::PSORT2::Report.parser(entry)
+
+    puts "\n ==> a.entry_id "
+    p a.entry_id
+    puts "\n ==> a.scl "
+    p a.scl
+    puts "\n ==> a.pred "
+    p a.pred
+    puts "\n ==> a.prob "
+    p a.prob
+    p a.prob.keys.sort.map {|k| k.rjust(4)}.inspect.gsub('"','')
+    p a.prob.keys.sort.map {|k| a.prob[k].to_s.rjust(4) }.inspect.gsub('"','')
+
+    puts "\n ==> a.k "
+    p a.k
+    puts "\n ==> a.definition"
+    p a.definition
+    puts "\n ==> a.seq"
+    p a.seq
+
+    puts "\n ==> a.features.keys.sort "
+    p a.features.keys.sort
+
+    a.features.keys.sort.each do |key|
+      puts "\n ==> a.features['#{key}'] "
+      puts a.features[key]
+    end
+
+    
+  end
+
+end