protk 1.4.1 → 1.4.2

data/bin/spectrast_filter.rb

@@ -0,0 +1,108 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 30/4/2015
+#
+# A wrapper for SpectraST commands that manipulate splib files 
+#
+#
+
+require 'protk/constants'
+require 'protk/command_runner'
+require 'protk/tool'
+require 'protk/galaxy_util'
+
+for_galaxy = GalaxyUtil.for_galaxy?
+
+genv=Constants.instance
+
+# Setup specific command-line options for this tool. Other options are inherited from ProphetTool
+#
+spectrast_tool=Tool.new([:explicit_output])
+spectrast_tool.option_parser.banner = "Manipulate splib files.\n\nUsage: spectrast_filter.rb [options] file1.splib file1.splib ..."
+spectrast_tool.add_boolean_option(:binary_output,false,['-B','--binary-output','Produce spectral libraries in binary format rather than ASCII'])
+spectrast_tool.add_value_option(:filter_predicate,nil,['--predicate pred','Keep only spectra satifying predicate pred. Should be a C-style predicate'])
+spectrast_tool.add_value_option(:merge_operation,"U",['--merge method',
+								'How to combine multiple splib files (if provided). Options are U,S,H
+				     U: Union. Include all the peptide ions in all the files.
+				     S: Subtraction. Only include peptide ions in the first file 
+				     	that are not present in any of the other files.
+				     H: Subtraction of homologs. Only include peptide ions in the
+				     	first file that do not have any homologs with 
+				     	same charge and similar m/z in any of the other files.
+				     A: Appending. Each peptide ion is added from only one library: 
+				     	the first file in the argument list that contains that peptide ion.
+				     	Useful for keeping existing consensus spectra unchanged while adding
+				     	only previously unseen peptide ions.'])
+spectrast_tool.add_value_option(:spectrum_operation,"None",['--replicates method',
+								'How to derive a single spectrum from replicates. Options are None, C,B
+				     C: Consensus. Create the consensus spectrum of all replicate spectra of each peptide ion.
+				     B: Best replicate. Pick the best replicate of each peptide ion.'])
+
+exit unless spectrast_tool.check_options(true)
+
+spectrast_bin = %x[which spectrast].chomp
+
+        # LIBRARY MANIPULATION OPTIONS (Applicable with .splib files)
+        #  -cf<pred>    Filter library. Keep only those entries satisfying the predicate <pred>. 
+        #                    <pred> should be a C-style predicate in quotes. 
+        #  -cJU         Union. Include all the peptide ions in all the files. 
+        #  -cJI         Intersection. Only include peptide ions that are present in all the files. 
+        #  -cJS         Subtraction. Only include peptide ions in the first file that are not present in any of the other files.
+        #  -cJH         Subtraction of homologs. Only include peptide ions in the first file 
+        #                    that do not have any homologs with same charge and similar m/z in any of the other files.
+        #  -cJA         Appending. Each peptide ion is added from only one library: the first file in the argument list that contains that peptide ion.
+        #                    Useful for keeping existing consensus spectra unchanged while adding only previously unseen peptide ions.
+        #  -cAB         Best replicate. Pick the best replicate of each peptide ion. 
+        #  -cAC         Consensus. Create the consensus spectrum of all replicate spectra of each peptide ion. 
+        #  -cAQ         Quality filter. Apply quality filters to library.
+        #                    IMPORTANT: Quality filter can only be applied on a SINGLE .splib file with no peptide ion represented by more than one spectrum.
+        #  -cAD         Create artificial decoy spectra. 
+        #  -cAN         Sort library entries by descending number of replicates used (tie-breaking by probability). 
+        #  -cAM         Create semi-empirical spectra based on allowable modifications specified by -cx option. 
+        #  -cQ<num>     Produce reduced spectra of at most <num> peaks. Inactive with -cAQ and -cAD.
+        #  -cD<file>    Refresh protein mappings of each library entry against the protein database <file> (Must be in .fasta format).
+        #  -cu          Delete entries whose peptide sequences do not map to any protein during refreshing with -cD option.
+        #                    When off, unmapped entries will be marked with Protein=0/UNMAPPED but retained in library. (Turn off with -cu!).
+        #  -cd          Delete entries whose peptide sequences map to multiple proteins during refreshing with -cD option. (Turn off with -cd!).
+
+input_stagers=[]
+inputs=ARGV.collect { |file_name| file_name.chomp}
+if for_galaxy
+  input_stagers = inputs.collect {|ip| GalaxyStager.new(ip,{:extension=>".splib"}) }
+  inputs=input_stagers.collect { |sg| sg.staged_path }
+end
+
+
+cmd="#{spectrast_bin} "
+
+unless spectrast_tool.binary_output
+	cmd << " -c_BIN!"	
+end
+
+if spectrast_tool.filter_predicate
+	cmd << "  -cf'#{spectrast_tool.filter_predicate}'"	
+end
+
+if inputs.length > 1
+	cmd << " -cJ#{spectrast_tool.merge_operation}"
+end
+
+if spectrast_tool.spectrum_operation!="None"
+	cmd << " -cA#{spectrast_tool.spectrum_operation}"
+end
+
+if spectrast_tool.explicit_output==nil
+    output_file_name=Tool.default_output_path(inputs,"","","")
+else
+    output_file_name=spectrast_tool.explicit_output
+end
+
+cmd << " -cN#{output_file_name}"
+
+inputs.each { |ip| cmd << " #{ip}" }
+
+# code = spectrast_tool.run(cmd,genv)
+# throw "Command failed with exit code #{code}" unless code==0
+
+%x[#{cmd}]

data/lib/protk/command_runner.rb

@@ -30,7 +30,7 @@ class CommandRunner
   def run_local(command_string)
     @env.log("Command: #{command_string} started",:info)
     status = Open4::popen4("#{command_string} ") do |pid, stdin, stdout, stderr|
-      puts "PID #{pid}" 
+      @env.log "PID #{pid}" , :info
       
       stdout.each { |line| @env.log(line.chomp,:info) }
 

data/lib/protk/data/template_pep.xml

@@ -0,0 +1,34 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<?xml-stylesheet type="text/xsl" href="/Users/icooke/Sources/protk/spec/data/mr176-BSA100fmole_BA3_01_8167.d_tandem_pproph.pep.xsl"?>
+<msms_pipeline_analysis date="2014-06-22T15:28:36" summary_xml="/Users/icooke/Sources/protk/spec/data/mr176-BSA100fmole_BA3_01_8167.d_tandem_pproph.pep.xml" xmlns="http://regis-web.systemsbiology.net/pepXML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://regis-web.systemsbiology.net/pepXML /Users/icooke/bin/tpp/schema/pepXML_v117.xsd">
+<analysis_summary analysis="peptideprophet" time="2014-06-22T15:28:36">
+</analysis_summary>
+<analysis_summary analysis="database_refresh" time="2014-06-22T15:28:36"/>
+<analysis_summary analysis="interact" time="2014-06-22T15:28:36">
+<interact_summary filename="/Users/icooke/Sources/protk/spec/data/mr176-BSA100fmole_BA3_01_8167.d_tandem_pproph.pep.xml" directory="">
+<inputfile name="mr176-BSA100fmole_BA3_01_8167.d_tandem.pep.xml" directory="/Users/icooke/Sources/protk/spec/data"/>
+</interact_summary>
+</analysis_summary>
+<dataset_derivation generation_no="0"/>
+<msms_run_summary base_name="/Users/icooke/Sources/protk/spec/data/mr176-BSA100fmole_BA3_01_8167.d_tandem.tandem" search_engine="X! Tandem" raw_data_type="raw" raw_data=".?">
+<sample_enzyme name="trypsin">
+<specificity cut="KR" no_cut="P" sense="C"/>
+</sample_enzyme>
+<search_summary base_name="mr176-BSA100fmole_BA3_01_8167.d_tandem.tandem" search_engine="X! Tandem" precursor_mass_type="monoisotopic" fragment_mass_type="monoisotopic" search_id="1">
+<search_database local_path="/Users/icooke/Sources/protk/spec/data/AASequences.fasta" type="AA"/>
+<enzymatic_search_constraint enzyme="trypsin" max_num_internal_cleavages="2" min_number_termini="1"/>
+<aminoacid_modification aminoacid="E" massdiff="-18.0106" mass="111.0320" variable="Y" symbol="^"/>
+<!--X! Tandem n-terminal AA variable modification-->
+<aminoacid_modification aminoacid="M" massdiff="15.9949" mass="147.0354" variable="Y"/>
+<aminoacid_modification aminoacid="Q" massdiff="-17.0265" mass="111.0321" variable="Y" symbol="^"/>
+<!--X! Tandem n-terminal AA variable modification-->
+<terminal_modification terminus="n" massdiff="42.0106" mass="43.0184" protein_terminus="N" variable="Y" symbol="^"/>
+
+</search_summary>
+<analysis_timestamp analysis="peptideprophet" time="2014-06-22T15:28:36" id="1"/>
+<analysis_timestamp analysis="database_refresh" time="2014-06-22T15:28:36" id="1">
+<database_refresh_timestamp database="/Users/icooke/Sources/protk/spec/data/AASequences.fasta" min_num_enz_term="1"/>
+</analysis_timestamp>
+
+</msms_run_summary>
+</msms_pipeline_analysis>

data/lib/protk/data/template_prot.xml

@@ -0,0 +1,39 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<protein_summary xmlns="http://regis-web.systemsbiology.net/protXML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://sashimi.sourceforge.net/schema_revision/protXML/protXML_v6.xsd" summary_xml="">
+<protein_summary_header reference_database="FULLPATH_TO_REFERENCE_DB" residue_substitution_list="I -&gt; L" source_files="FULLPATH_TO_SOURCE_PEPXML" source_files_alt="FULLPATH_TO_SOURCE_PEPXML" min_peptide_probability="" min_peptide_weight="" num_predicted_correct_prots="" num_input_1_spectra="" num_input_2_spectra="" num_input_3_spectra="" num_input_4_spectra="" num_input_5_spectra="" initial_min_peptide_prob="" total_no_spectrum_ids="" sample_enzyme="trypsin">
+<program_details analysis="proteinprophet" time="2014-01-20T14:17:37" version=" Insilicos_LabKey_C++ (TPP v0.0 Development trunk rev 0, Build 201307090846 (linux))">
+<proteinprophet_details occam_flag="Y" groups_flag="Y" degen_flag="Y" nsp_flag="Y" initial_peptide_wt_iters="2" nsp_distribution_iters="2" final_peptide_wt_iters="3">
+      <nsp_information neighboring_bin_smoothing="Y">
+         <nsp_distribution bin_no="0" nsp_lower_bound_incl="0.00" nsp_upper_bound_incl="0.00" pos_freq="0.057" neg_freq="0.625" pos_to_neg_ratio="0.09"/>
+         <nsp_distribution bin_no="1" nsp_lower_bound_excl="0.00" nsp_upper_bound_incl="0.31" pos_freq="0.037" neg_freq="0.152" pos_to_neg_ratio="0.24"/>
+         <nsp_distribution bin_no="2" nsp_lower_bound_excl="0.31" nsp_upper_bound_incl="1.00" pos_freq="0.077" neg_freq="0.032" pos_to_neg_ratio="2.42"/>
+         <nsp_distribution bin_no="3" nsp_lower_bound_excl="1.00" nsp_upper_bound_incl="2.50" pos_freq="0.113" neg_freq="0.033" pos_to_neg_ratio="3.39"/>
+         <nsp_distribution bin_no="4" nsp_lower_bound_excl="2.50" nsp_upper_bound_incl="4.63" pos_freq="0.123" neg_freq="0.032" pos_to_neg_ratio="3.91"/>
+         <nsp_distribution bin_no="5" nsp_lower_bound_excl="4.63" nsp_upper_bound_incl="7.90" pos_freq="0.143" neg_freq="0.032" pos_to_neg_ratio="4.50"/>
+         <nsp_distribution bin_no="6" nsp_lower_bound_excl="7.90" nsp_upper_bound_incl="14.92" pos_freq="0.196" neg_freq="0.041" pos_to_neg_ratio="4.78"/>
+         <nsp_distribution bin_no="7" nsp_lower_bound_excl="14.92" nsp_upper_bound_excl="inf" pos_freq="0.254" neg_freq="0.054" pos_to_neg_ratio="4.72" alt_pos_to_neg_ratio="4.78"/>
+      </nsp_information>
+      <ni_information>
+      </ni_information>
+      <protein_summary_data_filter min_probability="0.00" sensitivity="1.000" false_positive_error_rate="0.835" predicted_num_correct="1787" predicted_num_incorrect="9044"/>
+      <protein_summary_data_filter min_probability="0.10" sensitivity="1.000" false_positive_error_rate="0.235" predicted_num_correct="1787" predicted_num_incorrect="548"/>
+      <protein_summary_data_filter min_probability="0.20" sensitivity="1.000" false_positive_error_rate="0.235" predicted_num_correct="1787" predicted_num_incorrect="548"/>
+      <protein_summary_data_filter min_probability="0.30" sensitivity="0.956" false_positive_error_rate="0.151" predicted_num_correct="1709" predicted_num_incorrect="305"/>
+      <protein_summary_data_filter min_probability="0.40" sensitivity="0.916" false_positive_error_rate="0.095" predicted_num_correct="1638" predicted_num_incorrect="171"/>
+      <protein_summary_data_filter min_probability="0.50" sensitivity="0.887" false_positive_error_rate="0.063" predicted_num_correct="1585" predicted_num_incorrect="106"/>
+      <protein_summary_data_filter min_probability="0.60" sensitivity="0.853" false_positive_error_rate="0.036" predicted_num_correct="1525" predicted_num_incorrect="58"/>
+      <protein_summary_data_filter min_probability="0.70" sensitivity="0.826" false_positive_error_rate="0.020" predicted_num_correct="1477" predicted_num_incorrect="31"/>
+      <protein_summary_data_filter min_probability="0.80" sensitivity="0.805" false_positive_error_rate="0.012" predicted_num_correct="1438" predicted_num_incorrect="18"/>
+      <protein_summary_data_filter min_probability="0.90" sensitivity="0.773" false_positive_error_rate="0.006" predicted_num_correct="1381" predicted_num_incorrect="8"/>
+      <protein_summary_data_filter min_probability="0.95" sensitivity="0.749" false_positive_error_rate="0.004" predicted_num_correct="1339" predicted_num_incorrect="5"/>
+      <protein_summary_data_filter min_probability="0.96" sensitivity="0.738" false_positive_error_rate="0.003" predicted_num_correct="1318" predicted_num_incorrect="4"/>
+      <protein_summary_data_filter min_probability="0.97" sensitivity="0.728" false_positive_error_rate="0.002" predicted_num_correct="1302" predicted_num_incorrect="3"/>
+      <protein_summary_data_filter min_probability="0.98" sensitivity="0.711" false_positive_error_rate="0.002" predicted_num_correct="1272" predicted_num_incorrect="2"/>
+      <protein_summary_data_filter min_probability="0.99" sensitivity="0.609" false_positive_error_rate="0.000" predicted_num_correct="1088" predicted_num_incorrect="0"/>
+      <protein_summary_data_filter min_probability="1.00" sensitivity="0.164" false_positive_error_rate="0.000" predicted_num_correct="294" predicted_num_incorrect="0"/>
+</proteinprophet_details>
+</program_details>
+</protein_summary_header>
+<dataset_derivation generation_no="0">
+</dataset_derivation>
+</protein_summary>

data/lib/protk/mzidentml_doc.rb

@@ -0,0 +1,140 @@
+require 'libxml'
+
+include LibXML
+
+class MzIdentMLDoc < Object
+
+	MZID_NS_PREFIX="mzidentml"
+	MZID_NS='http://psidev.info/psi/pi/mzIdentML/1.1'
+
+	def initialize(path)
+		parser=XML::Parser.file(path)
+		@document=parser.parse
+	end
+
+
+	def spectrum_queries
+		@document.find("//#{MZID_NS_PREFIX}:SpectrumIdentificationResult","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+	def peptide_evidence
+		@document.find("//#{MZID_NS_PREFIX}:PeptideEvidence","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+	def psms
+		@document.find("//#{MZID_NS_PREFIX}:SpectrumIdentificationItem","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end	
+
+	def protein_groups
+		@document.find("//#{MZID_NS_PREFIX}:ProteinAmbiguityGroup","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+
+	def proteins
+		@document.find("//#{MZID_NS_PREFIX}:ProteinDetectionHypothesis","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+	# Peptides are referenced in many ways in mzidentml. 
+	# We define a "Peptide" as a peptide supporting a particular protein
+	# Such peptides may encompass several PSM's
+	#
+	def peptides
+		@document.find("//#{MZID_NS_PREFIX}:PeptideHypothesis","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+
+
+	# -----------------------------------------------------------
+	#
+	# Class Level Utility methods for searching from a given node
+	#
+	# -----------------------------------------------------------
+
+	def self.find(node,expression,root=false)
+		pp = root ? "//" : "./"
+		node.find("#{pp}#{MZID_NS_PREFIX}:#{expression}","#{MZID_NS_PREFIX}:#{MZID_NS}")
+	end
+
+
+	def self.get_cvParam(mzidnode,accession)
+		self.find(mzidnode,"cvParam[@accession=\'#{accession}\']")[0]
+	end
+
+	def self.get_dbsequence(mzidnode,accession)
+		self.find(mzidnode,"DBSequence[@accession=\'#{accession}\']",true)[0]
+	end
+
+	# As per PeptideShaker. Assume group probability used for protein if it is group rep otherwise 0
+	def self.get_protein_probability(protein_node)
+
+		#MS:1002403
+		is_group_representative=(self.get_cvParam(protein_node,"MS:1002403")!=nil)
+		if is_group_representative
+			return 	self.get_cvParam(protein_node.parent,"MS:1002470").attributes['value'].to_f*0.01
+		else
+			return 0
+		end
+	end
+
+	def self.get_proteins_for_group(group_node)
+		self.find(group_node,"ProteinDetectionHypothesis")
+	end
+
+	# def self.get_sister_proteins(protein_node)
+	# 	self.find(protein_node.parent,"ProteinDetectionHypothesis")
+	# end
+
+	def self.get_peptides_for_protein(protein_node)
+		self.find(protein_node,"PeptideHypothesis")
+	end
+
+	# <PeptideHypothesis peptideEvidence_ref="PepEv_1">
+	# 	<SpectrumIdentificationItemRef spectrumIdentificationItem_ref="SII_1_1"/>
+	# </PeptideHypothesis>
+	def self.get_best_psm_for_peptide(peptide_node)
+
+		best_score=-1
+		best_psm=nil
+		self.find(peptide_node,"SpectrumIdentificationItemRef").each do |id_ref_node|  
+			id_ref = id_ref_node.attributes['spectrumIdentificationItem_ref']
+			psm_node = self.find(peptide_node,"SpectrumIdentificationItem[@id=\'#{id_ref}\']",true)[0]
+			score = self.get_cvParam(psm_node,"MS:1002466")['value'].to_f
+			if score>best_score
+				best_psm=psm_node
+				best_score=score
+			end
+		end
+		best_psm
+	end
+
+	def self.get_sequence_for_peptide(peptide_node)
+		evidence_ref = peptide_node.attributes['peptideEvidence_ref']
+		pep_ref = peptide_node.find("//#{MZID_NS_PREFIX}:PeptideEvidence[@id=\'#{evidence_ref}\']","#{MZID_NS_PREFIX}:#{MZID_NS}")[0].attributes['peptide_ref']
+		peptide=peptide_node.find("//#{MZID_NS_PREFIX}:Peptide[@id=\'#{pep_ref}\']","#{MZID_NS_PREFIX}:#{MZID_NS}")[0]
+		# require 'byebug';byebug
+		peptide.find("./#{MZID_NS_PREFIX}:PeptideSequence","#{MZID_NS_PREFIX}:#{MZID_NS}")[0].content
+	end
+
+	def self.get_sequence_for_psm(psm_node)
+		pep_ref = psm_node.attributes['peptide_ref']
+		peptide=psm_node.find("//#{MZID_NS_PREFIX}:Peptide[@id=\'#{pep_ref}\']","#{MZID_NS_PREFIX}:#{MZID_NS}")[0]
+		peptide.find("./#{MZID_NS_PREFIX}:PeptideSequence","#{MZID_NS_PREFIX}:#{MZID_NS}")[0].content
+	end
+
+	def self.get_peptide_evidence_from_psm(psm_node)
+		pe_nodes = []
+		self.find(psm_node,"PeptideEvidenceRef").each do |pe_node|
+			ev_id=pe_node.attributes['peptideEvidence_ref']   
+			pe_nodes << self.find(pe_node,"PeptideEvidence[@id=\'#{ev_id}\']",true)[0]
+		end
+		pe_nodes
+	end
+
+
+
+
+
+
+
+
+end

data/lib/protk/mzml_parser.rb

@@ -14,6 +14,15 @@ class MzMLParser < Object
 		@file_reader=XML::Reader.document(doc)
 	end
 
+	def next_runid()
+		until @file_reader.name=="run" 
+			if !@file_reader.read()
+				return nil
+			end
+		end
+		return @file_reader.get_attribute('id')
+	end
+
 	def next_spectrum()
 
 		until @file_reader.name=="spectrum" 

data/lib/protk/peptide.rb

@@ -1,6 +1,7 @@
 require 'libxml'
 require 'bio'
 require 'protk/bio_gff3_extensions'
+require 'protk/mzidentml_doc'
 require 'protk/error'
 
 include LibXML
@@ -10,22 +11,55 @@ end
 
 class Peptide
 
+	# Stripped sequence (no modifications)
 	attr_accessor :sequence
 	attr_accessor :protein_name
 	attr_accessor :charge
-	attr_accessor :nsp_adjusted_probability
-
-
+	attr_accessor :probability
+	attr_accessor :theoretical_neutral_mass
+
+	def as_protxml
+		node = XML::Node.new('peptide')
+		node['peptide_sequence']=self.sequence.to_s
+		node['charge']=self.charge.to_s
+		node['nsp_adjusted_probability']=self.probability.to_s
+		node['calc_neutral_pep_mass']=self.theoretical_neutral_mass.to_s
+		node
+	end
 
 	class << self
 		def from_protxml(xmlnode)
 			pep=new()
 			pep.sequence=xmlnode['peptide_sequence']
-			pep.nsp_adjusted_probability=xmlnode['nsp_adjusted_probability'].to_f
+			pep.probability=xmlnode['nsp_adjusted_probability'].to_f
 			pep.charge=xmlnode['charge'].to_i
 			pep
 		end
 
+		# <ProteinDetectionHypothesis id="PAG_0_1" dBSequence_ref="JEMP01000193.1_rev_g3500.t1 280755" passThreshold="false">
+		# 	<PeptideHypothesis peptideEvidence_ref="PepEv_1">
+		# 		<SpectrumIdentificationItemRef spectrumIdentificationItem_ref="SII_1_1"/>
+		# 	</PeptideHypothesis>
+		# 	<cvParam cvRef="PSI-MS" accession="MS:1002403" name="group representative"/>
+		# 	<cvParam cvRef="PSI-MS" accession="MS:1002401" name="leading protein"/>
+		# 	<cvParam cvRef="PSI-MS" accession="MS:1001093" name="sequence coverage" value="0.0"/>
+		# </ProteinDetectionHypothesis>
+
+		def from_mzid(xmlnode)
+			pep=new()
+			pep.sequence=MzIdentMLDoc.get_sequence_for_peptide(xmlnode)
+			best_psm = MzIdentMLDoc.get_best_psm_for_peptide(xmlnode)
+			# require 'byebug';byebug
+			pep.probability = MzIdentMLDoc.get_cvParam(best_psm,"MS:1002466")['value'].to_f
+			pep.theoretical_neutral_mass = MzIdentMLDoc.get_cvParam(best_psm,"MS:1001117")['value'].to_f
+			pep.charge = best_psm.attributes['chargeState'].to_i
+			pep.protein_name = MzIdentMLDoc.get_dbsequence(xmlnode.parent,xmlnode.parent.attributes['dBSequence_ref']).attributes['accession']
+
+			# pep.charge = MzIdentMLDoc.get_charge_for_psm(best_psm)
+
+			pep
+		end
+
 		def from_sequence(seq,charge=nil)
 			pep=new()
 			pep.sequence=seq
@@ -146,7 +180,7 @@ class Peptide
 		cds_id = parent_record.id
 		this_id = "#{cds_id}.#{self.sequence}"
 		this_id << ".#{self.charge}" unless self.charge.nil?
-		score = self.nsp_adjusted_probability.nil? ? "." : self.nsp_adjusted_probability.to_s
+		score = self.probability.nil? ? "." : self.probability.to_s
 		gff_string = "#{parent_record.seqid}\tMSMS\tpolypeptide\t#{start_i}\t#{end_i}\t#{score}\t#{parent_record.strand}\t0\tID=#{this_id};Parent=#{cds_id}"
 		Bio::GFF::GFF3::Record.new(gff_string)
 	end

data/lib/protk/pepxml_writer.rb

@@ -0,0 +1,24 @@
+include LibXML
+
+class PepXMLWriter < Object
+
+	PEPXML_NS_PREFIX="pepxml"
+	PEPXML_NS="http://regis-web.systemsbiology.net/pepXML"
+
+	attr :template_doc
+
+	def initialize
+		template_path="#{File.dirname(__FILE__)}/data/template_pep.xml"
+		template_parser=XML::Parser.file(template_path)
+		@template_doc=template_parser.parse
+	end
+
+	def append_spectrum_query(query_node)
+		@template_doc.root << query_node
+	end
+
+	def save(file_path)
+		@template_doc.save(file_path,:indent=>true,:encoding => XML::Encoding::UTF_8)
+	end
+
+end