bio 1.3.0 → 1.3.1

data/lib/bio/appl/gcg/msf.rb

@@ -5,7 +5,7 @@
 #               Naohisa Goto <ng@bioruby.org>
 # License::     The Ruby License
 #
-# $Id: msf.rb,v 1.2 2007/04/05 23:35:39 trevor Exp $
+# $Id:$
 #
 # = About Bio::GCG::Msf
 #
@@ -30,14 +30,13 @@ module Bio
       # Creates a new Msf object.
       def initialize(str)
         str = str.sub(/\A[\r\n]+/, '')
-        if /^\!\![A-Z]+\_MULTIPLE\_ALIGNMNENT/ =~ str[/.*/] then
-          @heading = str[/.*/] # '!!NA_MULTIPLE_ALIGNMENT 1.0' or like this
-          str.sub!(/.*/, '')
-        end
-        str.sub!(/.*\.\.$/m, '')
-        @description = $&.to_s.sub(/^.*\.\.$/, '').to_s
+        preamble, @data = str.split(/^\/\/$/, 2)
+        preamble.sub!(/\A\!\![A-Z]+\_MULTIPLE\_ALIGNMENT.*/, '')
+        @heading = $& # '!!NA_MULTIPLE_ALIGNMENT 1.0' or like this
+        preamble.sub!(/.*\.\.\s*$/m, '')
+        @description = $&.to_s.sub(/^.*\.\.\s*$/, '').to_s
         d = $&.to_s
-        if m = /(.+)\s+MSF\:\s+(\d+)\s+Type\:\s+(\w)\s+(.+)\s+(Comp)?Check\:\s+(\d+)/.match(d) then
+        if m = /^(?:(.+)\s+)?MSF\:\s+(\d+)\s+Type\:\s+(\w)\s+(.+)\s+(Comp)?Check\:\s+(\d+)/.match(d) then
           @entry_id = m[1].to_s.strip
           @length   = (m[2] ? m[2].to_i : nil)
           @seq_type = m[3]
@@ -45,10 +44,8 @@ module Bio
           @checksum = (m[6] ? m[6].to_i : nil)
         end
 
-        str.sub!(/.*\/\/$/m, '')
-        a = $&.to_s.split(/^/)
         @seq_info = []
-        a.each do |x|
+        preamble.each_line do |x|
           if /Name\: / =~ x then
             s = {}
             x.scan(/(\S+)\: +(\S*)/) { |y| s[$1] = $2 }
@@ -56,7 +53,6 @@ module Bio
           end
         end
 
-        @data = str
         @description.sub!(/\A(\r\n|\r|\n)/, '')
         @align = nil
       end
@@ -133,11 +129,12 @@ module Bio
       # parsing
       def do_parse
         return if @align
-        a = @data.strip.split(/\n\n/)
+        a = @data.split(/\r?\n\r?\n/)
         @seq_data = Array.new(@seq_info.size)
         @seq_data.collect! { |x| Array.new }
         a.each do |x|
-          b = x.split(/\n/)
+          next if x.strip.empty?
+          b = x.sub(/\A[\r\n]+/, '').split(/[\r\n]+/)
           nw = 0
           if b.size > @seq_info.size then
             if /^ +/ =~ b.shift.to_s

data/lib/bio/appl/pts1.rb

@@ -65,10 +65,6 @@ class PTS1
   # Output report.
   attr_reader :output
 
-  # Used function name (Integer). 
-  # function_name = Bio::PTS1::FUNCTION.find_all {|k,v| v == pts1.function }[0][0]
-  attr_reader :function
-
   # Short-cut for Bio::PTS1.new(Bio::PTS1::FUNCTION['METAZOA-specific'])
   def self.new_with_metazoa_function
     self.new('METAZOA-specific')

data/lib/bio/appl/sim4/report.rb

@@ -4,7 +4,7 @@
 # Copyright:: Copyright (C) 2004 GOTO Naohisa <ng@bioruby.org>
 # License::   The Ruby License
 #
-#  $Id: report.rb,v 1.9 2007/04/05 23:35:40 trevor Exp $
+#  $Id:$
 #
 # The sim4 report parser classes.
 #
@@ -43,7 +43,7 @@ module Bio
         @hits = []
         @all_hits = []
         overrun = ''
-        text.each("\n\nseq1 = ") do |str|
+        text.each_line("\n\nseq1 = ") do |str|
           str = str.sub(/\A\s+/, '')
           str.sub!(/\n(^seq1 \= .*)/m, "\n") # remove trailing hits for sure
           tmp = $1.to_s
@@ -161,7 +161,7 @@ module Bio
         attr_reader :percent_identity
 
         # Returns directions of mapping.
-        # Maybe one of "->", "<-" or "" or nil.
+        # Maybe one of "->", "<-", "==" or "" or nil.
         # This would be a Bio::Sim4 specific method.
         attr_reader :direction
 
@@ -170,7 +170,7 @@ module Bio
         # Bio::Sim4::Report::Hit class.
         # Users shall not use it directly.
         def self.parse(str, aln)
-          /^(\d+)\-(\d+)\s*\((\d+)\-(\d+)\)\s*([\d\.]+)\%\s*([\-\<\>]*)/ =~ str
+          /^(\d+)\-(\d+)\s*\((\d+)\-(\d+)\)\s*([\d\.]+)\%\s*([\-\<\>\=]*)/ =~ str
           self.new(Segment.new($1, $2, aln[0]),
                    Segment.new($3, $4, aln[2]),
                    aln[1], $5, $6)
@@ -198,6 +198,17 @@ module Bio
                    aln[1])
         end
 
+        # Parses part of sim4 result text and creates a new SegmentPair
+        # object for regions which can not be aligned correctly.
+        # It is designed to be called internally from
+        # Bio::Sim4::Report::Hit class.
+        # Users shall not use it directly.
+        def self.both_intron(prev_e, e, aln)
+          self.new(Segment.new(prev_e.seq1.to+1, e.seq1.from-1, aln[0]),
+                   Segment.new(prev_e.seq2.to+1, e.seq2.from-1, aln[2]),
+                   aln[1])
+        end
+
         #--
         # Bio::BLAST::*::Report::Hsp compatible methods
         #   Methods already defined: midline, percent_identity
@@ -318,7 +329,9 @@ module Bio
             exo << e
             if ai then
               # intron data in alignment
-              if ai[2].strip.empty? then
+              if ai[1].strip.empty? then
+                i = SegmentPair.both_intron(prev_e, e, ai)
+              elsif ai[2].strip.empty? then
                 i = SegmentPair.seq1_intron(prev_e, e, ai)
               else
                 i = SegmentPair.seq2_intron(prev_e, e, ai)
@@ -338,29 +351,49 @@ module Bio
         # Parses alignment.
         def parse_align
           s1 = []; ml = []; s2 = []
+          blocks = []
+          blocks.push [ s1, ml, s2 ]
           dat = @data[1..-1]
           return unless dat
           dat.each do |str|
             a = str.split(/\r?\n/)
-            a.shift
-            if /^(\s*\d+\s*)(.+)$/ =~ a[0] then
-              range = ($1.length)..($1.length + $2.strip.length - 1)
+            ruler = a.shift
+            # First line, for example,
+            # "     50     .    :    .    :    .    :    .    :    .    :"
+            # When the number is 0, forced to be a separated block
+            if /^\s*(\d+)/ =~ ruler and $1.to_i == 0 and !ml.empty? then
+              s1 = []; ml = []; s2 = []
+              blocks.push [ s1, ml, s2 ]
+            end
+            # For example,
+            # "    190 GAGTCATGCATGATACAA          CTTATATATGTACTTAGCGGCA"
+            # "        ||||||||||||||||||<<<...<<<-||-|||||||||||||||||||"
+            # "    400 GAGTCATGCATGATACAACTT...AGCGCT ATATATGTACTTAGCGGCA"
+            if /^(\s*\d+\s)(.+)$/ =~ a[0] then
+              range = ($1.length)..($1.length + $2.chomp.length - 1)
               a.collect! { |x| x[range] }
               s1 << a.shift
               ml << a.shift
               s2 << a.shift
             end
           end #each
-          alx  = ml.join('').split(/([\<\>]+\.+[\<\>]+)/)
-          seq1 = s1.join(''); seq2 = s2.join('')
-          i = 0
-          alx.collect! do |x|
-            len = x.length
-            y = [ seq1[i, len], x, seq2[i, len] ]
-            i += len
-            y
+          alx_all = []
+          blocks.each do |ary|
+            s1, ml, s2 = ary
+            alx  = ml.join('').split(/([\<\>]+\.+[\<\>]+)/)
+            seq1 = s1.join(''); seq2 = s2.join('')
+            i = 0
+            alx.collect! do |x|
+              len = x.length
+              y = [ seq1[i, len], x, seq2[i, len] ]
+              i += len
+              y
+            end
+            # adds virtual intron information if necessary
+            alx_all.push([ '', '', '' ]) unless alx_all.empty?
+            alx_all.concat alx
           end
-          @align = alx
+          @align = alx_all
         end
         private :parse_align
 

data/lib/bio/db/biosql/biosql_to_biosequence.rb

@@ -46,6 +46,9 @@ module Bio::Sequence::Adapter::BioSQL
   def_biosequence_adapter :date_created
 
   def_biosequence_adapter :date_modified
+  #do |bs|
+   #   Date.parse(bs.date_modified.to_s).strftime("%d-%b-%Y").upcase
+    # end
 
   def_biosequence_adapter :division
 
@@ -63,5 +66,12 @@ module Bio::Sequence::Adapter::BioSQL
   
   def_biosequence_adapter :comments
 
+  def_biosequence_adapter :other_seqids do |orig|
+    orig.identifier.split(',').collect do |dblink|
+      database, id = dblink.split(':')
+      Bio::Sequence::DBLink.new(database,id)
+    end
+  end
+
 end #module Bio::Sequence::Adapter::BioSQL
 

data/lib/bio/db/biosql/sequence.rb

@@ -1,8 +1,10 @@
 
 #TODO save on db reading from a genbank or embl object
 module Bio
-  class SQL 
-    
+  class SQL
+
+
+
     class Sequence
       private
       #      example
@@ -12,10 +14,10 @@ module Bio
       #molecule_type=value add a bioentry_qualifier value to the table
       #molecule_type_update(value, rank) update an entry of the table with an existing rank
       #the method inferr the qualifier term from the name of the first symbol, or you can specify a synonym to use
-      
+
       #creating an object with to_biosql is transaction safe.
-      
-      #TODO: implement setting for more than a qualifier-vale. 
+
+      #TODO: implement setting for more than a qualifier-vale.
       def self.bioentry_qualifier_anchor(sym, *args)
         options = args.first || Hash.new
         #options.assert_valid_keys(:rank,:synonym,:multi)
@@ -23,278 +25,305 @@ module Bio
         method_writer_operator = (sym.to_s+"=").to_sym
         method_writer_modder = (sym.to_s+"_update").to_sym
         synonym = options[:synonym].nil? ? sym.to_s : options[:synonym]
-        
-        #Bio::SQL::Term.create(:name=>synonym, :ontology=> Bio::SQL::Ontology.find_by_name('Annotation Tags')) unless Bio::SQL::Term.exists?(:name =>synonym)
+
+        #DELETE        #Bio::SQL::Term.create(:name=>synonym, :ontology=> Bio::SQL::Ontology.find_by_name('Annotation Tags')) unless Bio::SQL::Term.exists?(:name =>synonym)
         send :define_method, method_reader do
           #return an array of bioentry_qualifier_values
           begin
-	    ontology_annotation_tags = Ontology.find_or_create_by_name('Annotation Tags')
-            term  = Term.find_or_create_by_name(:name => synonym, :ontology=> ontology_annotation_tags)
-            bioentry_qualifier_values = @entry.bioentry_qualifier_values.find_all_by_term_id(term)          
-            bioentry_qualifier_values.map{|row| row.value} unless bioentry_qualifier_values.nil?
-          rescue Exception => e 
+            #DELETE            ontology_annotation_tags = Ontology.find_or_create({:name=>'Annotation Tags'})
+            term  = Term.first(:conditions=>["name = ?",synonym]) || Term.create({:name => synonym, :ontology=> Ontology.first(:conditions=>["name = ?",'Annotation Tags'])})
+            bioentry_qualifier_values = @entry.bioentry_qualifier_values.all(:conditions=>["term_id = ?",term.term_id])
+            data = bioentry_qualifier_values.map{|row| row.value} unless bioentry_qualifier_values.nil?
+            begin
+              # this block try to check if the data retrived is a
+              # Date or not and change it according to GenBank/EMBL format
+              # in that case return a string
+              # otherwise the []
+              Date.parse(data.to_s).strftime("%d-%b-%Y").upcase
+            rescue ArgumentError, TypeError, NoMethodError, NameError
+              data
+            end
+          rescue Exception => e
             puts "Reader Error: #{synonym} #{e.message}"
           end
         end
-        
+
         send :define_method, method_writer_operator do |value|
           begin
-	    ontology_annotation_tags = Ontology.find_or_create_by_name('Annotation Tags')
-            term  = Term.find_or_create_by_name(:name => synonym, :ontology=> ontology_annotation_tags)
-            datas = @entry.bioentry_qualifier_values.find_all_by_term_id(term.term_id)
+            #DELETE            ontology_annotation_tags = Ontology.find_or_create({:name=>'Annotation Tags'})
+            term  = Term.first(:conditions=>["name = ?",synonym]) || Term.create({:name => synonym, :ontology=> Ontology.first(:conditions=>["name = ?",'Annotation Tags'])})
+            datas = @entry.bioentry_qualifier_values.all(:conditions=>["term_id = ?",term.term_id])
             #add an element incrementing the rank or setting the first to 1
-            @entry.bioentry_qualifier_values.create(:term_id=>term.term_id, :rank=>datas.empty? ? 1 : datas.last.rank.succ, :value=>value)
-          rescue Exception => e 
+            be_qu_va=@entry.bioentry_qualifier_values.build({:term=>term, :rank=>(datas.empty? ? 1 : datas.last.rank.succ), :value=>value})
+            be_qu_va.save
+          rescue Exception => e
             puts "WriterOperator= Error: #{synonym} #{e.message}"
           end
         end
-        
+
         send :define_method, method_writer_modder do |value, rank|
           begin
-	    ontology_annotation_tags = Ontology.find_or_create_by_name('Annotation Tags')
-            term  = Term.find_or_create_by_name(:name => synonym, :ontology=> ontology_annotation_tags)
-            data = @entry.bioentry_qualifier_values.find_by_term_id_and_rank(term.term_id, rank)
+            #DELETE            ontology_annotation_tags = Ontology.find_or_create({:name=>'Annotation Tags'})
+            term  = Term.first(:conditions=>["name = ?",synonym]) || Term.create({:name => synonym, :ontology=> Ontology.first(:conditions=>["name = ?",'Annotation Tags'])})
+            data = @entry.bioentry_qualifier_values.all(:term_id=>term.term_id, :rank=>rank)
             if data.nil?
               send method_writer_operator, value
             else
               data.value=value
-              data.save!
+              data.save
             end
           rescue Exception => e
             puts "WriterModder Error: #{synonym} #{e.message}"
           end
         end
-        
+
       end
+
       public
       attr_reader :entry
-      
+
       def delete
+        #TODO: check is references connected to this bioentry are leaf or not.
+        #actually I think it should be more sofisticated, check if there are
+        #other bioentries connected to references; if not delete 'em
+        @entry.references.each { |ref| ref.delete if ref.bioentries.size==1}
         @entry.destroy
       end
-      
+
       def get_seqfeature(sf)
-        
+
         #in seqfeature BioSQL class
         locations_str = sf.locations.map{|loc| loc.to_s}.join(',')
         #pp sf.locations.inspect
-        locations_str = "join(#{locations_str})" if sf.locations.count>1 
+        locations_str = "join(#{locations_str})" if sf.locations.count>1
         Bio::Feature.new(sf.type_term.name, locations_str,sf.seqfeature_qualifier_values.collect{|sfqv| Bio::Feature::Qualifier.new(sfqv.term.name,sfqv.value)})
       end
-      
+
       def length=(len)
         @entry.biosequence.length=len
       end
-      
+
       def initialize(options={})
-        options.assert_valid_keys(:entry, :biodatabase_id,:biosequence)
+        #options.assert_valid_keys(:entry, :biodatabase,:biosequence)
         return @entry = options[:entry] unless options[:entry].nil?
-        
-        return to_biosql(options[:biosequence], options[:biodatabase_id]) unless options[:biosequence].nil? or options[:biodatabase_id].nil?
-        
+
+        return to_biosql(options[:biosequence], options[:biodatabase]) unless options[:biosequence].nil? or options[:biodatabase].nil?
+
       end
-      
-      def to_biosql(bs,biodatabase_id)
-        #Transcaction works greatly!!!
 
-        #
+      def to_biosql(bs,biodatabase)
+        #DELETE        #Transcaction works greatly!!!
         begin
-          Bioentry.transaction do           
-        
-            @entry = Bioentry.new(:biodatabase_id=>biodatabase_id, :name=>bs.entry_id)
-
-                        puts "primary" if $DEBUG
-            self.primary_accession = bs.primary_accession
-
-                        puts "def" if $DEBUG
-            self.definition = bs.definition unless bs.definition.nil?
-
-                        puts "seqver" if $DEBUG
-            self.sequence_version = bs.sequence_version || 0
-
-                        puts "divi" if $DEBUG
-            self.division = bs.division unless bs.division.nil?
-
-            @entry.save!
-                        puts "secacc" if $DEBUG
-            
-            bs.secondary_accessions.each do |sa|
-              #write as qualifier every secondary accession into the array
-              self.secondary_accessions = sa
-            end unless bs.secondary_accessions.nil?
-
-            
-            #to create the sequence entry needs to exists
-		puts "seq" if $DEBUG
-	            puts bs.seq if $DEBUG
-            self.seq = bs.seq unless bs.seq.nil?
-                       puts "mol" if $DEBUG
-            
-            self.molecule_type = bs.molecule_type unless bs.molecule_type.nil?
-                        puts "dc" if $DEBUG
-
-            self.data_class = bs.data_class unless bs.data_class.nil?
-                        puts "top" if $DEBUG
-            self.topology = bs.topology unless bs.topology.nil?
-                        puts "datec" if $DEBUG
-            self.date_created = bs.date_created unless bs.date_created.nil?
-                        puts "datemod" if $DEBUG
-            self.date_modified = bs.date_modified unless bs.date_modified.nil?
-                        puts "key" if $DEBUG
-            
-            bs.keywords.each do |kw|
-              #write as qualifier every secondary accessions into the array
-              self.keywords = kw
-            end unless bs.keywords.nil?
-            #FIX: problem settinf taxon_name: embl has "Arabidopsis thaliana (thale cress)" but in taxon_name table there isn't this name. I must check if there is a new version of the table
-            puts "spec" if $DEBUG
-            self.species = bs.species unless bs.species.nil?
-                        puts "Debug: #{bs.species}" if $DEBUG
-                        puts "Debug: feat..start" if $DEBUG
-            
-            bs.features.each do |feat|
-              self.feature=feat
-            end unless bs.features.nil?
-			puts "Debug: feat...end" if $DEBUG
-            
-            #TODO: add comments and references
-	    bs.references.each do |reference|
-		 #   puts reference.inspect
-              self.reference=reference
-	    end unless bs.references.nil?
-            
-            bs.comments.each do |comment|
-            	self.comment=comment
-            end unless bs.comments.nil?
-            
-          end #transaction
+          #DELETE          Bioentry.transaction do
+          @entry = biodatabase.bioentries.build({:name=>bs.entry_id})
+
+          puts "primary" if $DEBUG
+          self.primary_accession = bs.primary_accession
+
+          puts "def" if $DEBUG
+          self.definition = bs.definition unless bs.definition.nil?
+
+          puts "seqver" if $DEBUG
+          self.sequence_version = bs.sequence_version || 0
+
+          puts "divi" if $DEBUG
+          self.division = bs.division unless bs.division.nil?
+
+          puts "identifier" if $DEBUG
+          self.identifier = bs.other_seqids.collect{|dblink| "#{dblink.database}:#{dblink.id}"}.join(';') unless bs.other_seqids.nil?
+          @entry.save
+          puts "secacc" if $DEBUG
+
+          bs.secondary_accessions.each do |sa|
+            puts "#{sa}" if $DEBUG
+            #write as qualifier every secondary accession into the array
+            self.secondary_accessions = sa
+          end unless bs.secondary_accessions.nil?
+
+
+          #to create the sequence entry needs to exists
+          puts "seq" if $DEBUG
+          puts bs.seq if $DEBUG
+          self.seq = bs.seq unless bs.seq.nil?
+          puts "mol" if $DEBUG
+
+          self.molecule_type = bs.molecule_type unless bs.molecule_type.nil?
+          puts "dc" if $DEBUG
+
+          self.data_class = bs.data_class unless bs.data_class.nil?
+          puts "top" if $DEBUG
+          self.topology = bs.topology unless bs.topology.nil?
+          puts "datec" if $DEBUG
+          self.date_created = bs.date_created unless bs.date_created.nil?
+          puts "datemod" if $DEBUG
+          self.date_modified = bs.date_modified unless bs.date_modified.nil?
+          puts "key" if $DEBUG
+
+          bs.keywords.each do |kw|
+            #write as qualifier every secondary accessions into the array
+            self.keywords = kw
+          end unless bs.keywords.nil?
+
+          puts "spec" if $DEBUG
+          #self.species = bs.species unless bs.species.nil?
+          self.species = bs.species unless bs.species.empty?
+          puts "Debug: #{bs.species}" if $DEBUG
+          puts "Debug: feat..start" if $DEBUG
+
+          bs.features.each do |feat|
+            self.feature=feat
+          end unless bs.features.nil?
+
+          puts "Debug: feat...end" if $DEBUG
+          bs.references.each do |reference|
+            self.reference=reference
+          end unless bs.references.nil?
+
+          bs.comments.each do |comment|
+            self.comment=comment
+          end unless bs.comments.nil?
+
+          #DELETE          end #transaction
           return self
         rescue Exception => e
           puts "to_biosql exception: #{e}"
           puts $!
-	end #rescue
+        end #rescue
       end #to_biosql
-      
-      
-      def name 
+
+
+      def name
         @entry.name
       end
       alias entry_id name
-      
+
       def name=(value)
         @entry.name=value
       end
       alias entry_id= name=
-      
+
       def primary_accession
         @entry.accession
       end
-      
+
       def primary_accession=(value)
         @entry.accession=value
       end
-      
+
       #TODO      def secondary_accession
       #        @entry.bioentry_qualifier_values
       #      end
-      
+
       def organism
         @entry.taxon.nil? ? "" : "#{@entry.taxon.taxon_scientific_name.name}"+ (@entry.taxon.taxon_genbank_common_name ? "(#{@entry.taxon.taxon_genbank_common_name.name})" : '')
       end
       alias species organism
-      
+
       def organism=(value)
-        taxon_name=TaxonName.find_by_name_and_name_class(value.gsub(/\s+\(.+\)/,''),'scientific name')
+        #FIX there is a shortcut
+        taxon_name=TaxonName.first(:conditions=>["name = ? and name_class = ?",value.gsub(/\s+\(.+\)/,''),'scientific name'])
         if taxon_name.nil?
           puts "Error value doesn't exists in taxon_name table with scientific name constraint."
         else
           @entry.taxon_id=taxon_name.taxon_id
-          @entry.save!
+          @entry.save
         end
       end
       alias species= organism=
-      
+
       def database
         @entry.biodatabase.name
       end
-      
+
       def database_desc
         @entry.biodatabase.description
       end
-      
+
       def version
         @entry.version
       end
-      alias sequence_version version 
-      
+      alias sequence_version version
+
       def version=(value)
         @entry.version=value
       end
       alias sequence_version= version=
-      
+
       def division
         @entry.division
       end
+
       def division=(value)
         @entry.division=value
       end
-      
+
       def description
         @entry.description
       end
       alias definition description
-      
+
       def description=(value)
         @entry.description=value
       end
       alias definition= description=
-      
+
       def identifier
         @entry.identifier
       end
-      
+      alias other_seqids identifier
+
       def identifier=(value)
         @entry.identifier=value
       end
-      
+
       bioentry_qualifier_anchor :data_class
       bioentry_qualifier_anchor :molecule_type, :synonym=>'mol_type'
       bioentry_qualifier_anchor :topology
-      bioentry_qualifier_anchor :date_created      
+      bioentry_qualifier_anchor :date_created
       bioentry_qualifier_anchor :date_modified, :synonym=>'date_changed'
       bioentry_qualifier_anchor :keywords, :synonym=>'keyword'
       bioentry_qualifier_anchor :secondary_accessions, :synonym=>'secondary_accession'
-      
+
       def features
-        @entry.seqfeatures.collect {|sf|
-          self.get_seqfeature(sf)}
+        @entry.seqfeatures.collect do |sf|
+          self.get_seqfeature(sf)
+        end
       end
-      
+
       def feature=(feat)
-	      #ToDo: avoid Ontology find here, probably more efficient create class variables
-	type_term_ontology = Ontology.find_or_create_by_name('SeqFeature Keys')
-        type_term = Term.find_or_create_by_name(:name=>feat.feature, :ontology=>type_term_ontology)
-	source_term_ontology = Ontology.find_or_create_by_name('SeqFeature Sources')
-	source_term = Term.find_or_create_by_name(:name=>'EMBLGenBankSwit',:ontology=>source_term_ontology)
-        seqfeature = Seqfeature.create(:bioentry=>@entry, :source_term=>source_term, :type_term=>type_term, :rank=>@entry.seqfeatures.count.succ, :display_name=>'')
-        #seqfeature.save!       
+        #ToDo: avoid Ontology find here, probably more efficient create class variables
+        #DELETE        type_term_ontology = Ontology.find_or_create({:name=>'SeqFeature Keys'})
+        puts "feature:type_term = #{feat.feature}" if $DEBUG
+        type_term = Term.first(:conditions=>["name = ?", feat.feature]) || Term.create({:name=>feat.feature, :ontology=>Ontology.first(:conditions=>["name = ?",'SeqFeature Keys'])})
+        #DELETE        source_term_ontology = Ontology.find_or_create({:name=>'SeqFeature Sources'})
+        puts "feature:source_term" if $DEBUG
+        source_term = Term.first(:conditions=>["name = ?",'EMBLGenBankSwit'])
+        puts "feature:seqfeature" if $DEBUG
+        seqfeature = @entry.seqfeatures.build({:source_term=>source_term, :type_term=>type_term, :rank=>@entry.seqfeatures.count.succ, :display_name=>''})
+        seqfeature.save
+        puts "feature:location" if $DEBUG
         feat.locations.each do |loc|
-          location = Location.new(:seqfeature=>seqfeature, :start_pos=>loc.from, :end_pos=>loc.to, :strand=>loc.strand, :rank=>seqfeature.locations.count.succ)
-          location.save!
+          location = seqfeature.locations.build({:seqfeature=>seqfeature, :start_pos=>loc.from, :end_pos=>loc.to, :strand=>loc.strand, :rank=>seqfeature.locations.count.succ})
+          location.save
         end
-	qual_term_ontology = Ontology.find_or_create_by_name('Annotation Tags')
+
+        #DELETE        qual_term_ontology = Ontology.find_or_create({:name=>'Annotation Tags'})
+
+        puts "feature:qualifier" if $DEBUG
         feat.each do |qualifier|
-          qual_term = Term.find_or_create_by_name(:name=>qualifier.qualifier, :ontology=>qual_term_ontology)
-          qual = SeqfeatureQualifierValue.new(:seqfeature=>seqfeature, :term=>qual_term, :value=>qualifier.value.to_s, :rank=>seqfeature.seqfeature_qualifier_values.count.succ)
-          qual.save!          
+          #DELETE          qual_term = Term.find_or_create({:name=>qualifier.qualifier}, {:ontology=>qual_term_ontology})
+          qual_term = Term.first(:conditions=>["name = ?", qualifier.qualifier]) || Term.create({:name=>qualifier.qualifier, :ontology=>Ontology.first(:conditions=>["name = ?", 'Annotation Tags'])})
+          qual = seqfeature.seqfeature_qualifier_values.build({:seqfeature=>seqfeature, :term=>qual_term, :value=>qualifier.value.to_s, :rank=>seqfeature.seqfeature_qualifier_values.count.succ})
+          qual.save
+
         end
       end
-      
+
       #return the seqfeature mapped from BioSQL with a type_term like 'CDS'
       def cdsfeatures
         @entry.cdsfeatures
       end
-      
+
       # Returns the sequence.
       # Returns a Bio::Sequence::Generic object.
 
@@ -302,39 +331,41 @@ module Bio
         s = @entry.biosequence
         Bio::Sequence::Generic.new(s ? s.seq : '')
       end
-      
-      def seq=(value)
 
+      def seq=(value)
+        #TODO: revise this piece of code.
         #chk which type of alphabet is, NU/NA/nil
         if @entry.biosequence.nil?
-#          puts "intoseq1"
+          #DELETE          puts "intoseq1"
           @entry.biosequence = Biosequence.new(:seq=>value)
-	  @entry.biosequence.save!
-
+          #          biosequence = @entry.biosequence.build({:seq=>value})
+          @entry.biosequence.save
+          #          biosequence.save
         else
           @entry.biosequence.seq=value
         end
         self.length=value.length
-        #@entry.biosequence.length=value.length
-        #break
-        @entry.save!
+        #DELETE        #@entry.biosequence.length=value.length
+        #DELETE        #break
+        @entry.save
       end
-      
+
+      #report parents and exclude info with "no rank". Now I report rank == class but ... Question ? Have to be reported taxonomy with rank=="class"?
       def taxonomy
         tax = []
-        taxon = @entry.taxon
-        while taxon and taxon.taxon_id != taxon.parent_taxon_id
-          tax << taxon.taxon_scientific_name.name
+        taxon = Taxon.first(:conditions=>["taxon_id = ?",@entry.taxon.parent_taxon_id])
+        while taxon and taxon.taxon_id != taxon.parent_taxon_id and taxon.node_rank!='no rank'
+          tax << taxon.taxon_scientific_name.name if taxon.node_rank!='class'
           #Note: I don't like this call very much, correct with a relationship in the ref class.
-          taxon = Taxon.find(taxon.parent_taxon_id)
+          taxon = Taxon.first(:conditions=>["taxon_id = ?",taxon.parent_taxon_id])
         end
         tax.reverse
       end
-      
-      def length 
+
+      def length
         @entry.biosequence.length
       end
-      
+
       def references
         #return and array of hash, hash has these keys ["title", "dbxref_id", "reference_id", "authors", "crc", "location"]
         #probably would be better to d a class refrence to collect these informations
@@ -342,7 +373,7 @@ module Bio
           hash = Hash.new
           hash['authors'] = bio_ref.reference.authors.gsub(/\.\s/, "\.\s\|").split(/\|/)
 
-	  hash['sequence_position'] = "#{bio_ref.start_pos}-#{bio_ref.end_pos}" if (bio_ref.start_pos and bio_ref.end_pos)
+          hash['sequence_position'] = "#{bio_ref.start_pos}-#{bio_ref.end_pos}" if (bio_ref.start_pos and bio_ref.end_pos)
           hash['title'] = bio_ref.reference.title
           hash['embl_gb_record_number'] = bio_ref.rank
           #TODO: solve the problem with specific comment per reference.
@@ -350,159 +381,64 @@ module Bio
           #take a look when location is build up in def reference=(value)
 
           bio_ref.reference.location.split('|').each do |element|
-          	key,value=element.split('=')
-          	hash[key]=value
+            key,value=element.split('=')
+            hash[key]=value
           end unless bio_ref.reference.location.nil?
 
           hash['xrefs'] = bio_ref.reference.dbxref ? "#{bio_ref.reference.dbxref.dbname}; #{bio_ref.reference.dbxref.accession}." : ''
           Bio::Reference.new(hash)
-        end        
+        end
       end
-      
+
       def comments
         @entry.comments.map do |comment|
           comment.comment_text
         end
       end
 
-      
-      def reference=(value)
-      
-      		locations=Array.new
-      		locations << "journal=#{value.journal}" unless value.journal.empty?
-      		locations << "volume=#{value.volume}" unless value.volume.empty?
-      		locations << "issue=#{value.issue}" unless value.issue.empty?
-      		locations << "pages=#{value.pages}" unless value.pages.empty?
-      		locations << "year=#{value.year}" unless value.year.empty?
-      		locations << "pubmed=#{value.pubmed}" unless value.pubmed.empty?
-      		locations << "medline=#{value.medline}" unless value.medline.empty?
-      		locations << "doi=#{value.doi}" unless value.doi.nil?
-      		locations << "abstract=#{value.abstract}" unless value.abstract.empty?
-      		locations << "url=#{value.url}" unless value.url.nil?
-      		locations << "mesh=#{value.mesh}" unless value.mesh.empty?      		
-      		locations << "affiliations=#{value.affiliations}" unless value.affiliations.empty?
-      		locations << "comments=#{value.comments.join('~')}"unless value.comments.nil?
-      	      start_pos, end_pos = value.sequence_position ? value.sequence_position.gsub(/\s*/,'').split('-') : [nil,nil] 
-	      reference=Reference.find_or_create_by_title(:title=>value.title, :authors=>value.authors.join(' '), :location=>locations.join('|'))
-	      
-	      bio_reference=BioentryReference.new(:bioentry=>@entry,:reference=>reference,:rank=>value.embl_gb_record_number, :start_pos=>start_pos, :end_pos=>end_pos)
-	      bio_reference.save!
-      end 
-      
+      def reference=(value)       
+        locations=Array.new
+        locations << "journal=#{value.journal}" unless value.journal.empty?
+        locations << "volume=#{value.volume}" unless value.volume.empty?
+        locations << "issue=#{value.issue}" unless value.issue.empty?
+        locations << "pages=#{value.pages}" unless value.pages.empty?
+        locations << "year=#{value.year}" unless value.year.empty?
+        locations << "pubmed=#{value.pubmed}" unless value.pubmed.empty?
+        locations << "medline=#{value.medline}" unless value.medline.empty?
+        locations << "doi=#{value.doi}" unless value.doi.nil?
+        locations << "abstract=#{value.abstract}" unless value.abstract.empty?
+        locations << "url=#{value.url}" unless value.url.nil?
+        locations << "mesh=#{value.mesh}" unless value.mesh.empty?
+        locations << "affiliations=#{value.affiliations}" unless value.affiliations.empty?
+        locations << "comments=#{value.comments.join('~')}"unless value.comments.nil?
+        start_pos, end_pos = value.sequence_position ? value.sequence_position.gsub(/\s*/,'').split('-') : [nil,nil]
+        reference= Reference.first(:conditions=>["title = ?",value.title]) || Reference.create({:title=>value.title,:authors=>value.authors.join(' '), :location=>locations.join('|')})
+        bio_reference=@entry.bioentry_references.build({:reference=>reference,:rank=>value.embl_gb_record_number, :start_pos=>start_pos, :end_pos=>end_pos})
+        bio_reference.save
+      end
+
       def comment=(value)
-      		comment=Comment.new(:bioentry=>@entry, :comment_text=>value, :rank=>@entry.comments.count.succ)
-      		comment.save!
+        #DELETE        comment=Comment.new({:bioentry=>@entry, :comment_text=>value, :rank=>@entry.comments.count.succ})
+        comment = @entry.comments.build({:comment_text=>value, :rank=>@entry.comments.count.succ})
+        comment.save
       end
-      
+
       def save
         #I should add chks for SQL errors
-        @entry.biosequence.save!
-        @entry.save!
+        @entry.biosequence.save
+        @entry.save
       end
       def to_fasta
-        #prima erano 2 print in stdout, meglio ritornare una stringa in modo che poi ci si possa fare quello che si vuole
-        #print ">" + accession + "\n" 
-        #print seq.gsub(Regexp.new(".{1,#{60}}"), "\\0\n")
-				">" + accession + "\n" + seq.gsub(Regexp.new(".{1,#{60}}"), "\\0\n")
+        ">" + accession + "\n" + seq.gsub(Regexp.new(".{1,#{60}}"), "\\0\n")
       end
-      
+
       def to_fasta_reverse_complememt
-				">" + accession + "\n" + seq.reverse_complement.gsub(Regexp.new(".{1,#{60}}"), "\\0\n")
+        ">" + accession + "\n" + seq.reverse_complement.gsub(Regexp.new(".{1,#{60}}"), "\\0\n")
       end
-      
-      
-      
+
       def to_biosequence
-	 Bio::Sequence.adapter(self,Bio::Sequence::Adapter::BioSQL)
+        Bio::Sequence.adapter(self,Bio::Sequence::Adapter::BioSQL)
       end
     end #Sequence
-    
-    
   end #SQL
 end #Bio
-
-#TODO create tests for sequence object, roundtrip of informations
-
-if __FILE__ == $0
-  
-  require 'bio'
-  require 'bio/io/sql'
-  require 'pp'
-  
-  #  connection = Bio::SQL.establish_connection('bio/io/biosql/config/database.yml','development')
-  connection = Bio::SQL.establish_connection({'development'=>{'database'=>"biorails_development", 'adapter'=>"postgresql", 'username'=>"rails", 'password'=>nil}},'development') 
-  databases = Bio::SQL.list_databases
-  
-  #  parser = Bio::FlatFile.auto('/home/febo/Desktop/aj224122.embl')
-  parser = Bio::FlatFile.auto('/home/febo/Desktop/aj224122.gb')
-  #parser = Bio::FlatFile.auto('/home/febo/Desktop/aj224122.fasta')
-  
-  parser.each do |entry|
-    biosequence = entry.to_biosequence
-    result = Bio::SQL::Sequence.new(:biosequence=>biosequence,:biodatabase_id=>databases.first[:id]) unless Bio::SQL.exists_accession(biosequence.primary_accession) 
-    
-    if result.nil?
-      pp "The sequence is already present into the biosql database"
-    else
-      #      pp "Sequence"
-      puts result.to_biosequence.output(:genbank) #:embl
-      result.delete
-    end   
-  end
-  
-  if false
-    sqlseq = Bio::SQL.fetch_accession('AJ224122')
-    #only output tests.
-    pp "Connection"
-    pp connection
-    pp "Seq in dbs"
-    pp Bio::SQL.list_entries
-    #; NC_003098
-    
-    
-    #pp sqlseq
-    pp sqlseq.entry.inspect
-    pp "sequence"
-    #pp Bio::Sequence.auto(sqlseq.seq)
-    pp "entry_id"
-    pp sqlseq.entry_id
-    
-    pp "primary"
-    pp sqlseq.accession
-    pp "secondary_accessions"
-    pp sqlseq.secondary_accessions
-    pp "molecule type"
-    pp sqlseq.molecule_type
-    pp "data_class"
-    pp sqlseq.data_class
-    pp "division"
-    pp sqlseq.division
-    # NOTE : Topology is not represented in biosql?
-    pp "topology"
-    #TODO:  CIRCULAR   this at present maps to bioentry_qualifier_value, though there are plans to make it a column in table biosequence.
-    pp sqlseq.topology
-    pp "version"
-    pp sqlseq.version
-    #sequence.date_created = nil #????
-    pp "date modified"
-    pp sqlseq.date_modified
-    pp "definition"
-    pp sqlseq.definition
-    pp "keywords"
-    pp sqlseq.keywords
-    pp "species"
-    pp sqlseq.organism
-    #sequence.classification = self.taxonomy.to_s.sub(/\.\z/, '').split(/\s*\;\s*/)"
-    pp "classification"
-    pp sqlseq.taxonomy
-    #sequence.organnella = nil # not used
-    pp "comments"
-    pp sqlseq.comments
-    pp "references"
-    pp sqlseq.references
-    pp "features"
-    pp sqlseq.features
-    puts sqlseq.to_biosequence.output(:embl)
-  end
-  ##
-end