ms-ident 0.0.2

data/lib/ms/ident/pepxml/msms_pipeline_analysis.rb

@@ -0,0 +1,70 @@
+require 'merge'
+
+require 'ms/ident/pepxml/msms_run_summary'
+
+module Ms ; end
+module Ms::Ident ; end
+class Ms::Ident::Pepxml; end
+
+class Ms::Ident::Pepxml::MsmsPipelineAnalysis
+  include Merge
+  XMLNS = "http://regis-web.systemsbiology.net/pepXML"
+  XMLNS_XSI = "http://www.w3.org/2001/XMLSchema-instance"
+  # (this doesn't actually exist), also, the space is supposed to be there
+  XSI_SCHEMA_LOCATION_BASE = "http://regis-web.systemsbiology.net/pepXML /tools/bin/TPP/tpp/schema/pepXML_v"
+  # the only additions concerning a writer are from v18 are to the 'spectrum': retention_time_sec and activationMethodType
+  PEPXML_VERSION = 115
+
+  #include SpecIDXML
+  # Version 1.2.3
+  #attr_writer :date
+  #attr_writer :xmlns, :xmlns_xsi, :xsi_schemaLocation
+  #attr_accessor :summary_xml 
+  
+  attr_accessor :xmlns
+  attr_accessor :xmlns_xsi
+  attr_accessor :xsi_schema_location
+  # an Integer
+  attr_accessor :pepxml_version
+  # self referential path to the outputfile
+  attr_accessor :summary_xml
+  attr_accessor :msms_run_summary
+  attr_writer :date
+
+  def block_arg 
+    @msms_run_summary = Ms::Ident::Pepxml::MsmsRunSummary.new
+  end
+
+  # if block given, yields a new msms_run_summary to return value of block
+  def initialize(hash={}, &block)
+    @xmlns = XMLNS
+    @xmlns_xsi = XMLNS_XSI
+    @xsi_schema_location = xsi_schema_location
+    @pepxml_version = PEPXML_VERSION
+    merge!(hash, &block)
+  end
+
+  # returns the location based on the pepxml version number
+  def xsi_schema_location
+    XSI_SCHEMA_LOCATION_BASE + pepxml_version.to_s + '.xsd'
+  end
+
+  # if no date string given, then it will set to Time.now
+  def date
+    return @date if @date
+    tarr = Time.now.to_a 
+    tarr[3..5].reverse.join('-') + "T#{tarr[0..2].reverse.join(':')}"
+  end
+
+  # uses the filename as summary_xml (if it is nil) attribute and builds a complete, valid xml document,
+  # writing it to the filename
+  def to_xml(builder)
+    xmlb = builder || Nokogiri::XML::Builder.new
+    xmlb.msms_pipeline_analysis(:date => date, :xmlns => xmlns, 'xsi:schemaLocation'.to_sym => xsi_schema_location, :summary_xml => summary_xml) do |xmlb|
+      msms_run_summary.to_xml(xmlb) if msms_run_summary
+    end
+    builder || xmlb.doc.root.to_xml
+  end
+end
+
+

data/lib/ms/ident/pepxml/msms_run_summary.rb

@@ -0,0 +1,81 @@
+require 'merge'
+require 'nokogiri'
+
+require 'ms/ident/pepxml/sample_enzyme'
+require 'ms/ident/pepxml/search_summary'
+
+module Ms ; end
+module Ms::Ident ; end
+class Ms::Ident::Pepxml; end
+
+class Ms::Ident::Pepxml::MsmsRunSummary
+  include Merge
+  # The name of the pep xml file without any extension
+  attr_accessor :base_name
+  # The name of the mass spec manufacturer 
+  attr_accessor :ms_manufacturer
+  attr_accessor :ms_model
+  attr_accessor :ms_mass_analyzer
+  attr_accessor :ms_detector
+  attr_accessor :raw_data_type
+  attr_accessor :raw_data
+  attr_accessor :ms_ionization
+  attr_accessor :pepxml_version
+
+  # A SampleEnzyme object (responds to: name, cut, no_cut, sense)
+  attr_accessor :sample_enzyme
+  # A SearchSummary object
+  attr_accessor :search_summary
+  # An array of spectrum_queries
+  attr_accessor :spectrum_queries
+
+  def block_arg
+    [@sample_enzyme = Ms::Ident::Pepxml::SampleEnzyme.new,    
+      @search_summary = Ms::Ident::Pepxml::SearchSummary.new,
+      @spectrum_queries ]
+  end
+
+  # takes a hash of name, value pairs
+  # if block given, yields a SampleEnzyme object, a SearchSummary and an array
+  # for SpectrumQueries
+  def initialize(hash={}, &block)
+    @spectrum_queries = []
+    merge!(hash, &block)
+    block.call(block_arg) if block
+  end
+
+  # optionally takes an xml builder object and returns the builder, or the xml
+  # string if no builder was given
+  # sets the index attribute of each spectrum query if it is not already set
+  def to_xml(builder=nil)
+    xmlb = builder || Nokogiri::XML::Builder.new
+    hash = {:base_name => base_name, :msManufacturer => ms_manufacturer, :msModel => ms_model, :msIonization => ms_ionization, :msMassAnalyzer => ms_mass_analyzer, :msDetector => ms_detector, :raw_data_type => raw_data_type, :raw_data => raw_data}
+    hash.each {|k,v| hash.delete(k) unless v }
+    xmlb.msms_run_summary(hash) do |xmlb|
+      sample_enzyme.to_xml(xmlb) if sample_enzyme
+      search_summary.to_xml(xmlb) if search_summary
+      spectrum_queries.each_with_index do |sq,i| 
+        sq.index = i+1 unless sq.index
+        sq.to_xml(xmlb)
+      end
+    end
+    builder || xmlb.doc.root.to_xml
+  end
+
+  def self.from_pepxml_node(node)
+    self.new.from_pepxml_node(node)
+  end
+
+  # peps correspond to search_results
+  def from_pepxml_node(node)
+    @base_name = node['base_name']
+    @ms_manufacturer = node['msManufacturer']
+    @ms_model = node['msModel']
+    @ms_manufacturer = node['msIonization']
+    @ms_mass_analyzer = node['msMassAnalyzer']
+    @ms_detector = node['msDetector']
+    @raw_data_type = node['raw_data_type']
+    @raw_data = node['raw_data']
+    self
+  end
+end

data/lib/ms/ident/pepxml/parameters.rb

@@ -0,0 +1,14 @@
+module Ms
+  module Ident
+    class Pepxml
+      class Parameters < Hash
+        def to_xml(builder)
+          self.each do |k,v|
+            builder.parameter(:name => k, :value => v)
+          end
+        end
+      end
+    end
+  end
+end
+

data/lib/ms/ident/pepxml/pep_summary.rb

@@ -0,0 +1,104 @@
+
+require 'arrayclass'
+require 'spec_id/sequest/pepxml'
+require 'spec_id/parser/proph'
+
+module Sequest ; end
+class Sequest::PepXML ; end
+class Sequest::PepXML::MSMSRunSummary ; end
+class Sequest::PepXML::SearchHit ; end
+
+module SpecID ; end
+module SpecID::Prot ; end
+module SpecID::Pep ; end
+
+
+
+module Proph
+
+  class PepSummary
+    include SpecID
+
+    Filetype_and_version_re_new = /version="PeptideProphet v([\d\.]+) /
+
+    # inherits prots and peps
+
+    # the protein groups
+    # currently these are just xml nodes returned!
+    attr_accessor :peptideprophet_summary
+    attr_accessor :msms_run_summaries
+    attr_accessor :version
+
+    def hi_prob_best ; true end
+
+    def get_version(file)
+      answer = nil
+      File.open(file) do |fh|
+        8.times do
+          line = fh.gets
+          answer = 
+            if line =~ Filetype_and_version_re_new
+              $1.dup
+            end
+          break if answer
+        end
+      end
+      raise(ArgumentError, "couldn't detect version in #{file}") unless answer
+      answer
+    end
+
+    def search_hit_class
+      PepSummary::Pep
+    end
+
+    def initialize(file=nil)
+      if file
+        @version = get_version(file)
+        spec_id = SpecID::Parser::PepProph.new(:spec_id).parse(file, :spec_id => self)
+      end
+    end
+  end
+
+  # this is a SpecID::Pep (by interface: not including stuff yet)
+  class PepSummary::Pep < Sequest::PepXML::SearchHit
+    # aaseq is defined in SearchHit
+    
+    %w(probability fval ntt nmc massd prots).each do |guy|
+      self.add_member(guy)
+    end
+
+    # returns self
+    def from_pepxml_node(node)
+      super(node)
+
+      an_res = node.find_first('child::analysis_result')
+      pp_n = an_res.find_first('child::peptideprophet_result')
+      self.probability = pp_n['probability'].to_f
+      pp_n.find('descendant::parameter').each do |par_n|
+        case par_n['name']
+        when 'fval'
+          self.fval = par_n['value'].to_f
+        when 'ntt'
+          self.ntt = par_n['value'].to_i
+        when 'nmc'
+          self.nmc = par_n['value'].to_i
+        when 'massd'
+          self.massd = par_n['value'].to_f
+        end
+      end
+      self
+    end
+  end
+
+  ::Proph::PepSummary::Prot = Arrayclass.new(%w(name protein_descr peps))
+
+  class PepSummary::Prot
+    def first_entry ; self[0] end ## name
+    def reference ; self[0] + ' ' + self[1] end
+  end
+
+end
+
+
+
+