bio-sambamba 0.0.3 → 0.0.4

data/Gemfile

@@ -1,7 +1,7 @@
 source "http://rubygems.org"
 
 gem "bio", "~> 1.4.2"
-gem "oj", "~> 1.3.4"
+gem "msgpack", "~> 0.4.7"
 
 group :development do
   gem "bundler", "~> 1.1.4"

data/VERSION

@@ -1 +1 @@
-0.0.3
+0.0.4

data/lib/bio-sambamba.rb

@@ -1,5 +1,6 @@
 require 'bio/command'
-require 'oj'
+require 'json'
+require 'msgpack'
 
 require 'bio-sambamba/exception.rb'
 require 'bio-sambamba/samheader.rb'

data/lib/bio-sambamba/alignment.rb

@@ -4,95 +4,101 @@ module Bio
     # Class representing an alignment record
     class Alignment
       
-      # Creates a new object from JSON output of sambamba tool
-      def initialize(json)
-        @json = json
+      # Creates a new object from MessagePack record
+      def initialize(obj, reference_sequence_names)
+        @obj = obj
+        @reference = reference_sequence_names[ref_id]
+        @mate_reference = reference_sequence_names[mate_ref_id]
       end
 
       # Access a record tag
       def [](tag)
         raise 'tag length must be two' unless tag.length == 2
-        @json['tags'][tag]
+        tags[tag]
       end
 
       def ==(read)
-        read.json == json
+        read.obj == obj
       end
 
       # Hash of record tags
-      attr_reader :tags if false
+      def tags
+        obj[12]
+      end
 
-      # Name of reference sequence
-      attr_reader :reference if false
+      # ID of reference sequence
+      def ref_id
+        obj[2]
+      end
 
       # Query template name
-      attr_reader :read_name if false
+      def read_name
+        obj[0]
+      end
 
       # 1-based leftmost mapping position
-      attr_reader :position if false
+      def position
+        obj[3]
+      end
 
       # Mapping quality
-      attr_reader :mapping_quality if false
+      def mapping_quality
+        obj[4]
+      end
+
+      # CIGAR: pairs of operations and lengths, 
+      # or nil if information is not available
+      def cigar_operations
+        return nil if obj[5].nil?
+        obj[6].chars.zip obj[5]
+      end
 
       # CIGAR string
-      attr_reader :cigar if false
+      def cigar
+        return '*' if cigar_operations.nil?
+        cigar_operations.reduce(''){|s, op_len| s + op_len[0] + op_len[1].to_s}
+      end
 
       # Observed template length
-      attr_reader :template_length if false
+      def template_length
+        obj[9]
+      end
 
       # Bitwise flag
-      attr_reader :flag if false
+      def flag
+        obj[1]
+      end
 
       # Phred-scaled base quality, an integer array
       # of the same length as the sequence
-      attr_reader :quality if false
+      def quality
+        obj[11].bytes.to_a
+      end
 
       # Segment sequence
-      attr_reader :sequence if false
+      def sequence
+        obj[10]
+      end
 
       # Reference sequence name of the mate/next segment
-      attr_reader :mate_reference if false
+      def mate_ref_id
+        obj[7]
+      end
 
       # 1-based leftmost position of the mate/next segment
-      attr_reader :mate_position if false
+      def mate_position
+        obj[8]
+      end
 
       # The number of reference bases covered
       def bases_covered
-        return 0 if cigar == '*'
-        cigar.split(/([MIDNSHP=X])/).each_slice(2).reduce(0) {|res, op| 
-          res += op[0].to_i unless ('M=XDN'.index op[1]).nil?
+        return 0 if cigar_operations.nil?
+        cigar_operations.reduce(0) {|res, op| 
+          res += op[1] unless ('M=XDN'.index op[0]).nil?
           res
         }
       end
-
-      {'position' => 'pos',
-       'mapping_quality' => 'mapq',
-       'template_length' => 'tlen',
-       'flag' => 'flag',
-       'mate_position' => 'pnext'
-      }.each do |k, v|
-        eval <<-DEFINE_READER
-          def #{k}
-            @json['#{v}'].to_i
-          end
-        DEFINE_READER
-      end
-
-      {'tags' => 'tags',
-       'reference' => 'rname',
-       'read_name' => 'qname',
-       'cigar' => 'cigar',
-       'quality' => 'qual',
-       'sequence' => 'seq',
-       'mate_reference' => 'rnext'
-       }.each do |k, v|
-        eval <<-DEFINE_READER
-          def #{k}
-            @json['#{v}']
-          end
-        DEFINE_READER
-      end
-
+      
       # Template having multiple segments in sequencing
       def is_paired                
         (flag & 0x1) != 0
@@ -148,8 +154,14 @@ module Bio
         (flag & 0x400) != 0
       end
 
+      # Reference sequence name
+      attr_reader :reference
+
+      # Mate reference sequence name
+      attr_reader :mate_reference
+
       private
-      attr_accessor :json
+      attr_accessor :obj
 
     end
 

data/lib/bio-sambamba/alignmentiterator.rb

@@ -8,8 +8,10 @@ module Bio
 
       # Creates a new AlignmentIterator object which will
       # parse JSON outputted by a specified command.
-      def initialize(command)
+      # Names of reference sequences must be provided as well.
+      def initialize(command, references)
         @command = command
+        @references = references
       end
 
       # Iterate only through valid alignments
@@ -17,12 +19,14 @@ module Bio
 
         return enum_for(:each_valid) if not block_given?
 
-        command = @command
+        command = get_command
         if command.index('--valid').nil?
           command.push '--valid'
         end
 
-        AlignmentIterator.new(command).each do |read|
+        iter = self.clone
+        iter.command = command
+        iter.each do |read|
           yield read
         end
       end
@@ -55,10 +59,21 @@ module Bio
         command = get_command
 
         Bio::Command.call_command_open3(command) do |pin, pout, perr|
-          pout.each do |line|
-            json = Oj.load(line)
-            yield Bio::Bam::Alignment.new(json)
+
+          counter = 0 # for triggering garbage collection manually
+          unpacker = MessagePack::Unpacker.new pout
+
+          begin
+            unpacker.each do |obj|
+              counter += 1
+              yield Bio::Bam::Alignment.new(obj, @references)
+              if (counter & 0xFFF) == 0 then
+                ObjectSpace.garbage_collect
+              end
+            end
+          rescue EOFError
           end
+
           raise_exception_if_stderr_is_not_empty(perr)
         end
       end
@@ -101,7 +116,7 @@ module Bio
       end
 
       def clone
-        iter = AlignmentIterator.new @command
+        iter = AlignmentIterator.new @command, @references
         iter.chromosome = chromosome
         iter.region = region
         iter

data/lib/bio-sambamba/bamfile.rb

@@ -29,7 +29,8 @@ module Bio
 
       # Returns an AlignmentIterator object for iterating over all alignments in the file
       def alignments
-        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format', 'json', @filename]
+        cmdline = ['sambamba', 'view', '--format', 'msgpack', @filename]
+        Bio::Bam::AlignmentIterator.new(cmdline, reference_sequence_names)
       end
 
       # True if index file was found 
@@ -47,16 +48,25 @@ module Bio
       # * _chr_: reference sequence
       # * _region_: a Range representing an interval. Coordinates are 1-based.
       def fetch(chr, region)
-        iter = Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format=json', 
-                                               @filename]
+        cmdline = ['sambamba', 'view', '--format=msgpack', @filename]
+        iter = Bio::Bam::AlignmentIterator.new(cmdline, reference_sequence_names)
         iter.chromosome = chr
         iter.region = region
         iter
       end
 
+      def reference_sequences
+        @reference_sequences ||= JSON.parse(Bio::Command.query_command ['sambamba', 'view', '-I', @filename])
+      end
+
       def [](chr)
         fetch(chr, nil)
       end
+
+      private
+      def reference_sequence_names
+        @reference_sequence_names ||= reference_sequences.map {|ref| ref['name']}
+      end
     end # class File
     
   end # module Bam

data/lib/bio-sambamba/samfile.rb

@@ -20,7 +20,17 @@ module Bio
 
       # Returns an AlignmentIterator object for iterating over all alignments in the file
       def alignments
-        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format', 'json', '-S', @filename]
+        cmdline = ['sambamba', 'view', '--format', 'msgpack', '-S', @filename],
+        Bio::Bam::AlignmentIterator.new(cmdline, reference_sequences)
+      end
+
+      def reference_sequences
+        @reference_sequences ||= Oj.load(Bio::Command.query_command ['sambamba', 'view', '-I', '-S', @filename])
+      end
+
+      private
+      def reference_sequence_names
+        @reference_sequence_names ||= reference_sequences.map {|ref| ref['name']}
       end
     end
 

data/lib/bio-sambamba/samheader.rb

@@ -26,172 +26,185 @@ module Bio
 
       # Format version
       def version
-        @json ||= get_json
-        @json['format_version']
+        obj[0]
       end
 
       # Sorting order
       def sorting_order
-        @json ||= get_json
-        @json['sorting_order']
+        obj[1]
       end
 
       # An array of SQLine objects
       def sq_lines
-        @json ||= get_json
-        @sq_lines ||= @json['sq_lines'].map{|json| SQLine.new(json)}
+        @sq_lines ||= obj[2].map{|rec| SQLine.new(rec)}
       end
 
       # An array of RGLine objects
       def rg_lines
-        @json ||= get_json
-        @sq_lines ||= @json['rg_lines'].map{|json| RGLine.new(json)}
+        @rg_lines ||= obj[3].map{|rec| RGLine.new(rec)}
       end
 
-      # @return [PGLine] array of @PG lines
+      # An array of PGLine objects
       def pg_lines
-        @json ||= get_json
-        @sq_lines ||= @json['pg_lines'].map{|json| PGLine.new(json)}
+        @pg_lines ||= obj[4].map{|rec| PGLine.new(rec)}
       end
 
       private
-      # Calls sambamba to get underlying JSON object
-      def get_json
-        cmd = ['sambamba', 'view', '-H', '--format=json', @filename] + @opts
+      def obj
+        return @obj unless @obj.nil?
+        cmd = ['sambamba', 'view', '-H', '--format', 'msgpack', @filename] + @opts
         line = ''
         Bio::Command.call_command_open3(cmd) do |pin, pout, perr|
-          line = pout.read
+          @obj = MessagePack.unpack(pout.read)
           raise_exception_if_stderr_is_not_empty(perr)
         end
-        @json = Oj.load(line)
+        @obj
       end
     end
 
     # Represents a @SQ line from SAM header
     class SQLine
 
-      # Wrap JSON object from sambamba output
-      def initialize(json)
-        @json = json
+      # Wrap MessagePack record from sambamba output
+      def initialize(obj)
+        @obj = obj
       end
 
       # Reference sequence name
-      attr_reader :sequence_name if false
+      def sequence_name
+        @obj['SN']
+      end
 
       # Reference sequence length
-      attr_reader :sequence_length if false
+      def sequence_length
+        @obj['LN']
+      end
 
       # Genome assembly identifier
-      attr_reader :assembly if false
+      def assembly
+        @obj['AS']
+      end
 
       # MD5 checksum of the sequence in uppercase, with gaps and spaces removed
-      attr_reader :md5 if false
+      def md5
+        @obj['M5']
+      end
 
       # Species
-      attr_reader :species if false
+      def species
+        @obj['SP']
+      end
 
       # URI of the sequence
-      attr_reader :uri if false
-
-      ['sequence_name', 'sequence_length', 
-       'assembly', 'md5', 'species', 'uri'].each do |sq_line_field|
-        eval <<-DEFINE_READER
-          def #{sq_line_field}
-            @json['#{sq_line_field}']
-          end
-        DEFINE_READER
+      def uri
+        @obj['UR']
       end
+
     end
 
     # Represents @RG line from SAM header, i.e. a read group
     class RGLine
 
-      # Wrap JSON object from sambamba output
-      def initialize(json)
-        @json = json
+      # Wrap MessagePack record from sambamba output
+      def initialize(obj)
+        @obj = obj
       end
 
       # Unique read group identifier
-      attr_reader :identifier if false
+      def identifier
+        @obj['ID']
+      end
 
       # Name of sequencing center
-      attr_reader :sequencing_center if false
+      def sequencing_center
+        @obj['CN']
+      end
 
       # Description
-      attr_reader :description if false
+      def description
+        @obj['DS']
+      end
 
       # Date the run was produced (ISO8601 date or date/time)
-      attr_reader :date if false
+      def date
+        @obj['DT']
+      end
 
       # Flow order. The array of nucleotide bases that correspond to the 
       # nucleotides used for each flow of each read. Multi-base flows are 
       # encoded in IUPAC format, and non-nucleotide flows by various other
       # characters.
-      attr_reader :flow_order if false
+      def flow_order
+        @obj['FO']
+      end
 
       # The array of nucleotide bases that correspond to the key sequence of each read
-      attr_reader :key_sequence if false
+      def key_sequence
+        @obj['KS']
+      end
 
       # Library
-      attr_reader :library if false
+      def library
+        @obj['LB']
+      end
 
       # Programs used for processing the read group
-      attr_reader :programs if false
+      def programs
+        @obj['PG']
+      end
 
       # Predicted median insert size
-      attr_reader :predicted_insert_size if false
+      def predicted_insert_size
+        @obj['PI']
+      end
 
       # Platform/technology used to produce the reads
-      attr_reader :platform if false
+      def platform
+        @obj['PL']
+      end
 
-      # Platform unit (e.g. flowcell-barcode.lane for Illumina or slide for SOLiD). Unique identifier.
-      attr_reader :platform_unit if false
+      # Platform unit (e.g. flowcell-barcode lane for Illumina or slide for SOLiD). Unique identifier.
+      def platform_unit
+        @obj['PU']
+      end
 
       # Sample
-      attr_reader :sample if false
-
-      ['identifier', 'sequencing_center', 'description', 'date',
-       'flow_order', 'key_sequence', 'library', 'programs',
-       'predicted_insert_size', 'platform', 
-       'platform_unit', 'sample'].each do |rg_line_field|
-        eval <<-DEFINE_READER
-          def #{rg_line_field}
-            @json['#{rg_line_field}']
-          end
-        DEFINE_READER
+      def sample
+        @obj['SM']
       end
     end
 
     # Represents @PG line from SAM header (program record)
     class PGLine
 
-      # Wrap JSON object from sambamba output
-      def initialize(json)
-        @json = json
+      # Wrap MessagePack record from sambamba output
+      def initialize(obj)
+        @obj = obj
       end
 
       # Unique program record identifier
-      attr_reader :identifier if false
+      def identifier
+        @obj['ID']
+      end
 
       # Program name
-      attr_reader :program_name if false
+      def program_name
+        @obj['PN']
+      end
 
       # Command line
-      attr_reader :command_line if false
+      def command_line
+        @obj['CL']
+      end
 
       # Identifier of previous program in chain
-      attr_reader :previous_program if false
+      def previous_program
+        @obj['PP']
+      end
 
       # Program version
-      attr_reader :program_version if false
-
-      ['identifier', 'program_name', 'command_line',
-       'previous_program', 'program_version'].each do |rg_line_field|
-        eval <<-DEFINE_READER
-          def #{rg_line_field}
-            @json['#{rg_line_field}']
-          end
-        DEFINE_READER
+      def program_version
+        @obj['VN']
       end
     end
 

metadata

@@ -1,13 +1,8 @@
 --- !ruby/object:Gem::Specification 
 name: bio-sambamba
 version: !ruby/object:Gem::Version 
-  hash: 1889055196096400351
   prerelease: 
-  segments: 
-  - 0
-  - 0
-  - 3
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors: 
 - Artem Tarasov
@@ -15,104 +10,74 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2012-08-15 00:00:00 Z
+date: 2012-08-20 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
-  type: :runtime
-  prerelease: false
   name: bio
-  version_requirements: &id001 !ruby/object:Gem::Requirement 
+  requirement: &id001 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 3725784833518553922
-        segments: 
-        - 1
-        - 4
-        - 2
         version: 1.4.2
-  requirement: *id001
-- !ruby/object:Gem::Dependency 
   type: :runtime
   prerelease: false
-  name: oj
-  version_requirements: &id002 !ruby/object:Gem::Requirement 
+  version_requirements: *id001
+- !ruby/object:Gem::Dependency 
+  name: msgpack
+  requirement: &id002 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 3635686038185777981
-        segments: 
-        - 1
-        - 3
-        - 4
-        version: 1.3.4
-  requirement: *id002
-- !ruby/object:Gem::Dependency 
-  type: :development
+        version: 0.4.7
+  type: :runtime
   prerelease: false
+  version_requirements: *id002
+- !ruby/object:Gem::Dependency 
   name: bundler
-  version_requirements: &id003 !ruby/object:Gem::Requirement 
+  requirement: &id003 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 1287940794363108929
-        segments: 
-        - 1
-        - 1
-        - 4
         version: 1.1.4
-  requirement: *id003
-- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
+  version_requirements: *id003
+- !ruby/object:Gem::Dependency 
   name: jeweler
-  version_requirements: &id004 !ruby/object:Gem::Requirement 
+  requirement: &id004 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 4544990976642784663
-        segments: 
-        - 1
-        - 8
-        - 3
         version: 1.8.3
-  requirement: *id004
-- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
+  version_requirements: *id004
+- !ruby/object:Gem::Dependency 
   name: rspec
-  version_requirements: &id005 !ruby/object:Gem::Requirement 
+  requirement: &id005 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 3576421176010082098
-        segments: 
-        - 2
-        - 7
-        - 0
         version: 2.7.0
-  requirement: *id005
-- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
+  version_requirements: *id005
+- !ruby/object:Gem::Dependency 
   name: cucumber
-  version_requirements: &id006 !ruby/object:Gem::Requirement 
+  requirement: &id006 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 242951009964279202
-        segments: 
-        - 1
-        - 2
-        - 0
         version: 1.2.0
-  requirement: *id006
+  type: :development
+  prerelease: false
+  version_requirements: *id006
 description: New Sambamba library comes with a command-line tool for working with SAM/BAM files. This gem brings some of its functionality to Ruby.
 email: lomereiter@gmail.com
 executables: []
@@ -179,7 +144,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
-      hash: 2002549777813010636
+      hash: -3697473229949428622
       segments: 
       - 0
       version: "0"
@@ -188,14 +153,11 @@ required_rubygems_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
-      hash: 2002549777813010636
-      segments: 
-      - 0
       version: "0"
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.8.12
+rubygems_version: 1.8.24
 signing_key: 
 specification_version: 3
 summary: Ruby wrapper for Sambamba tool