mspire 0.3.1 → 0.3.9

data/Rakefile

@@ -238,8 +238,8 @@ spec = Gem::Specification.new do |s|
   s.rdoc_options = rdoc_options
   s.extra_rdoc_files = rdoc_extra_includes
   s.executables = FL["bin/*"].map {|file| File.basename(file) }
-  s.add_dependency('libjtp', '~> 0.2.12')
-  s.add_dependency('axml')
+  s.add_dependency('libjtp', '~> 0.2.13')
+  s.add_dependency('axml', '~> 0.0.0')
   s.requirements << '"libxml" is the prefered xml parser right now.  libxml, xmlparser, REXML and regular expressions are used as fallback in some routines.'
   s.requirements << 'some plotting functions will not be available without the "gnuplot" gem (and underlying gnuplot binary)'
   s.requirements << 'the "t2x" binary (in archive) or readw.exe is required to convert .RAW files to mzXML in some applications'

data/bin/bioworks_to_pepxml.rb

@@ -43,14 +43,26 @@ opt_obj = OptionParser.new do |op|
   op.separator "Options:"
   op.on('-h', '--help', "display this and more notes and exit") {|v| opt.help = v }
   op.on('-o', '--outdir path', "output directory     d: '#{DEFAULT_OUTDIR}'") {|v| opt.outdir = v }
+  op.on('--sample_enzyme <type>', "For digested samples run with no enzymatic",
+                                  "search constraint, the enzyme used for",
+                                  "digestion, options: 'Trypsin_KR_P'") {|v| 
+    case v
+    when 'Trypsin_KR_P'
+      opt.sample_enzyme = SampleEnzyme.new("trypsin")
+    else
+      raise ArgumentError, "Don't recognize enzyme: #{v}"
+    end
+  } 
+  op.on('-a', '--all_hits', "includes all hits, not just top xcorr") {|v| opt.all_hits = v }
+  op.on('--deltacn_orig', "top hit deltacn = 0.0, (no deltacnstar att)") {|v| opt.deltacn_orig = v }
+  op.on('-m', '--mspath path', "path to MS files     d: '#{DEFAULT_MZ_PATH}'") {|v| opt.mspath = v }
+  op.on('--copy_mzxml', "copies mzXML files to outdir path"){|v| opt.copy_mzxml = v }
 
   op.separator ""
   op.separator "bioworks.xml files may require additional options:"
   op.separator ""
   op.on('-p', '--params file', "sequest params file  d: '#{DEFAULT_PARAMS_FILE}'") {|v| opt.params = v }
   op.on('-d', '--dbpath path', "path to databases    d: '#{DEFAULT_DATABASE_PATH}'") {|v| opt.dbpath = v }
-  op.on('-m', '--mspath path', "path to MS files     d: '#{DEFAULT_MZ_PATH}'") {|v| opt.mspath = v }
-  op.on('--copy_mzxml', "copies mzXML files to outdir path"){|v| opt.copy_mzxml = v }
   op.on('--model <LCQ|Orbi|string>', "MS model      (xml)  d: '#{DEFAULT_MS_MODEL}'") {|v| opt.model = v }
   op.on('--mass_analyzer <string>',  "Mass Analyzer (xml)  d: '#{DEFAULT_MASS_ANALYZER}'") {|v| opt.mass_analyzer = v }
 
@@ -131,5 +143,5 @@ opt.params ||= DEFAULT_PARAMS_FILE
 opt.mass_analyzer ||= DEFAULT_MASS_ANALYZER
 opt.model ||= DEFAULT_MS_MODEL
 
-xml_objs = Sequest::PepXML.set_from_bioworks(bioworks_file, {:params => opt.params, :ms_data => opt.mspath, :out_path => opt.outdir, :model => model, :backup_db_path => opt.dbpath, :copy_mzxml => opt.copy_mzxml, :ms_mass_analyzer => opt.mass_analyzer, :print => true})
+xml_objs = Sequest::PepXML.set_from_bioworks(bioworks_file, {:params => opt.params, :ms_data => opt.mspath, :out_path => opt.outdir, :model => model, :backup_db_path => opt.dbpath, :copy_mzxml => opt.copy_mzxml, :ms_mass_analyzer => opt.mass_analyzer, :print => true, :all_hits => opt.all_hits, :deltacn_orig => opt.deltacn_orig, :sample_enzyme => opt.sample_enzyme})
 

data/bin/ms_to_lmat.rb

@@ -47,7 +47,8 @@ ARGV.each do |file|
   }
   args.merge!(opt)
   lmat = LMat.new.from_times_and_spectra(times, spectra, args)
-  outfile = file.sub(/\.mzXML$/, opt[:newext])
+  ext = File.extname(file)
+  outfile = file.sub(/#{Regexp.escape(ext)}$/, opt[:newext])
   if args[:ascii]
     outfile << "a"
     lmat.print(outfile)

data/bin/sqt_group.rb

@@ -0,0 +1,26 @@
+#!/usr/bin/ruby
+
+require 'optparse'
+require 'spec_id/sqt'
+
+$OUTFILE = 'bioworks.sqg'
+
+opts = OptionParser.new do |op|
+  op.banner = "usage: #{File.basename(__FILE__)} <file1>.sqt <file2>.sqt ..."
+  op.separator "outputs: 'bioworks.sqg'"
+  op.separator ""
+  op.separator "    A '.sqg' file is an ascii text file with a list"
+  op.separator "    of the sqt files (full path names) in that group."
+  op.separator ""
+  op.on('-o', '--output <filename>', 'a different output name') {|v| $OUTFILE }
+end
+
+if ARGV.size == 0
+  puts opts
+  exit
+end
+
+obj = SQTGroup.new
+obj.filenames = ARGV.to_a
+obj.to_sqg($OUTFILE)
+

data/changelog.txt

@@ -126,3 +126,39 @@ interfaces and implementations (using ArrayClass)
 ## version 0.3.1
 
 1. Bug fix in srf filtering (num_hits adjusted)
+
+## version 0.3.2
+
+1. Uses sequest peptide_mass_tolerance filter on srf group files by default
+now.
+
+## version 0.3.3
+
+1. Worked out minor kinks in prob_precision.rb
+
+## version 0.3.4
+
+1. filters >= +3 charged ions now.
+
+## version 0.3.5
+
+1. fixed creation of background distribution in validators (hash_by base_name,
+first_scan, charge now)
+
+## version 0.3.6
+
+1. split off bad_aa_est from bad_aa
+
+## version 0.3.7
+
+1. can deal with No_Enzyme searches now (while still capable of setting
+sample_enzyme)
+
+## version 0.3.8
+
+1. can set a decoy to target ratio for decoy validation
+2. added mass calculator in Mass::Calculator
+
+## version 0.3.9
+
+1. doesn't clobber mzdata filename in ms_to_lmat.rb conversion

data/lib/ms/msrun.rb

@@ -30,7 +30,9 @@ class MS::MSRun
     myopts = opts.dup ; myopts[:msrun] = self
     if file
       filetype_and_version = MS::Parser.filetype_and_version(file)
-      MS::Parser.new(filetype_and_version, :msrun).parse(file, myopts)
+      parser = MS::Parser.new(filetype_and_version, :msrun)
+      parser.parse(file, myopts)
+      #MS::Parser.new(filetype_and_version, :msrun).parse(file, myopts)
       (@filetype, @version) = filetype_and_version
     end
   end

data/lib/ms/parser/mzdata/dom.rb

@@ -51,23 +51,20 @@ class MS::Parser::MzData::DOM
     # %w(num msLevel retentionTime startMz endMz precursors spectrum)
 
     root = get_root_node_from_file(file)
-    scan_count = 0
     description = root.find_first('child::description')
     bioworks33 = is_bioworks33?(description)  
     spectrum_list = description.next
-    scans =
-    if bioworks33
-      [] #bioworks33 gives incorrect scan numbers!
-    else
-      Array(spectrum_list['count'].to_i)
-    end
+
+    scans = []
+
+    # bioworks 33 gives incorrect scan count
+    stated_num_scans = spectrum_list['count'].to_i
 
     # if I move from node to node, it means I've checked that it's a sequence
     # and that the elements are req'd
     if spectrum_list.child?
       spectrum_n = spectrum_list.child
       loop do
-        scan_count += 1
         scan = MS::Scan.new(9)
         id = spectrum_n["id"].to_i
         id_to_scan_hash[id] = scan
@@ -81,11 +78,9 @@ class MS::Parser::MzData::DOM
         spec_inst_n = spec_settings_n.find_first('child::spectrumInstrument')
         scan[1] = spec_inst_n['msLevel'].to_i
 
-        if bioworks33
-          scans << scan # we can't trust the scan count!
-        else
-          scans[scan_count] = scan
-        end
+        # we could use a scan_count, but in bioworks 33, we can't trust the
+        # scan count!  So, we just collect them
+        scans << scan 
 
         scan[3] = spec_inst_n['mzRangeStart'].to_f
         scan[4] = spec_inst_n['mzRangeStop'].to_f
@@ -149,7 +144,12 @@ class MS::Parser::MzData::DOM
       MS::MSRun.add_parent_scan(scans, opts[:spectra])
     end
     msrun_obj.scans = scans
-    msrun_obj.scan_count = scan_count
+    msrun_obj.scan_count = scans.size
+    unless bioworks33  # we know the scan count is off here
+      if msrun_obj.scan_count != stated_num_scans
+        warn "num collected scans (#{scans.size}) does not agree with stated num scans (#{stated_num_scans})!"
+      end
+    end
     msrun_obj.start_time = msrun_obj.scans.first.time
     msrun_obj.end_time = msrun_obj.scans.last.time
   end

data/lib/ms/scan.rb

@@ -28,7 +28,7 @@ class MS::Scan
     atts = %w(num ms_level time start_mz end_mz) 
     display = atts.map do |att|
       if val = send(att.to_sym)
-        "@#{att}=#{val}"
+        "#{att}=#{val}"
       else
         nil
       end
@@ -38,9 +38,9 @@ class MS::Scan
       if spectrum
         spectrum.mz.size
       else
-        nil
+        'nil'
       end
-    "<MS::Scan:#{__id__} " + display.join(", ") + "@precursors=#{precursors.inspect}" + "@spectrum=size:#{spec_display}" + ">"
+    "<MS::Scan:#{__id__} " + display.join(", ") + " precursors=#{precursors.inspect}" + " spectrum(size)=#{spec_display}" + " >"
   end
 
   # returns the string (space delimited): "ms_level num time [prec_mz prec_inten]"

data/lib/mspire.rb

@@ -1,4 +1,4 @@
 
 module Mspire
-  Version = '0.3.1'
+  Version = '0.3.9'
 end

data/lib/sample_enzyme.rb

@@ -23,6 +23,7 @@ class SampleEnzyme
   # For other enzymes, you must set :cut, :no_cut, :name, and :sense
   # will yield the object if you want to set the values that way
   def initialize(name=nil)
+    @num_missed_cleavages_regex = nil
     @sense = nil
     @cut = nil
     @no_cut = nil
@@ -62,6 +63,44 @@ class SampleEnzyme
     self.new.from_pepxml_node(node)
   end
 
+  # takes an amino acid sequence (e.g., -.PEPTIDK.L)
+  # returns the number of missed cleavages
+  def num_missed_cleavages(aaseq)
+    raise NotImplementedError, 'need to implement for N terminal sense'  if sense == 'N'
+    @num_missed_cleavages_regex = 
+      if @num_missed_cleavages_regex ; @num_missed_cleavages_regex
+      else
+        regex_string = "[#{@cut}]"
+        if @no_cut and @no_cut != ''
+          regex_string << "[^#{@no_cut}]"
+        end
+        /#{regex_string}/
+      end
+    arr = aaseq.scan(@num_missed_cleavages_regex)
+    num = arr.size
+    if aaseq[-1,1] =~ @num_missed_cleavages_regex
+      num -= 1
+    end
+    num
+  end
+
+  # requires full sequence (with heads and tails)
+  def num_tol_term(sequence)
+    raise NotImplementedError, 'need to implement for N terminal sense'  if sense == 'N'
+    no_cut = @no_cut || ''
+    num_tol = 0
+    first, middle, last = SpecID::Pep.split_sequence(sequence)
+    last_of_middle = middle[-1,1]
+    first_of_middle = middle[0,1]
+    if ( @cut.include?(first) && !no_cut.include?(first_of_middle) ) || first == '-'
+      num_tol += 1
+    end
+    if @cut.include?(last_of_middle) && !no_cut.include?(last) || last == '-'
+      num_tol += 1
+    end
+    num_tol
+  end
+
   # returns all peptides of missed cleavages <= 'missed_cleavages'
   # so 2 missed cleavages will return all no missed cleavage peptides
   # all 1 missed cleavages and all 2 missed cleavages.

data/lib/spec_id.rb

@@ -7,6 +7,7 @@ require 'spec_id/bioworks'
 require 'spec_id/sequest'
 require 'spec_id/proph/prot_summary'
 require 'spec_id_xml'
+require 'spec_id/sqt'
 require 'spec_id/mass'
 require 'fasta'
 
@@ -71,6 +72,10 @@ module SpecID
       Proph::ProtSummary.new(file)
     when 'pepproph'
       Proph::PepSummary.new(file)
+    when 'sqg'
+      SQTGroup.new(file)
+    when 'sqt'
+      SQTGroup.new([file])
     else
       abort "UNRECOGNIZED file type for #{file}"
     end
@@ -447,6 +452,8 @@ module SpecID
   def self.file_type(file)
     if file =~ /\.srg$/
       return 'srg'
+    elsif file =~ /\.sqg$/
+      return 'sqg'
     end
     if IO.read(file, 7,438) == 'Enzyme:'
       return 'srf'
@@ -461,6 +468,17 @@ module SpecID
       elsif lines =~ /<msms_pipeline_analysis.*<peptideprophet_summary/m
         return 'pepproph'
       end
+      # assumes the header of a sqt file is less than 200 lines ...
+      200.times do 
+        line = fh.gets
+        if line
+          lines << line
+        else ; break
+        end
+      end
+      if lines =~ /^H\tDatabase/ and lines =~ /^H\tSQTGenerator/
+        return 'sqt'
+      end
     end
   end
 

data/lib/spec_id/aa_freqs.rb

@@ -3,30 +3,27 @@ require 'fasta'
 module SpecID ; end
 
 class SpecID::AAFreqs
-  # a fasta object
-  attr_accessor :fasta
   # hash by capital one-letter amino acid symbols giving the frequency of
   # seeing that amino acid.  Frequencies should add to 1.
   attr_accessor :aafreqs
 
   # fasta is fasta object!
   def initialize(fasta=nil)
-    @fasta = fasta
-    if @fasta
-      @aafreqs = calculate_frequencies(@fasta)
+    if fasta
+      @aafreqs = calculate_frequencies(fasta.prots)
     end
   end
 
-  # creates an aafreqs hash based on fasta object
-  def calculate_frequencies(fasta)
+  # takes an enumerable of objects responding to :aaseq and creates an aafreqs hash 
+  def calculate_frequencies(objs)
     hash = {}
     total_aas = 0
     ('A'..'Z').each do |x|
       hash[x] = 0
     end
     hash['*'] = 0
-    fasta.prots.each do |prot|
-      aaseq = prot.aaseq
+    objs.each do |obj|
+      aaseq = obj.aaseq
       total_aas += aaseq.size
       aaseq.split('').each do |x|
         hash[x] += 1

data/lib/spec_id/digestor.rb

@@ -100,38 +100,37 @@ class Digestor
   #   The prot_aaseq is used if the mass_hash contains the keys
   #   :add_C_term_protein or :add_N_term_protein
   #
+  #   mass_hash requires the key :h_plus or :h depending on h_plus option.
   #   prot_aaseqs is parallel to pep_aaseqs_ar where each is a group of
   #   peptides matching a protein aaseq
-  #   returns another parallel array of passing proteins
+  #   returns another parallel array of passing peptides per protein
   def limit_sizes(prot_aaseqs, pep_aaseqs_ar, min_mh, max_mh, mass_hash, h_plus=false)
     if mass_hash.key?(:add_C_term_protein) or mass_hash.key?(:add_N_term_protein)
       raise NotImplementedError, "need to add ability to change weights of peptides from the ends of proteins"
     else
       # figure out how much must be added to each peptide
       # include the h2o, the h, and N and C terminal static mods
-      h_key = h_plus ? :h_plus : :h
-      final_add = mass_hash[:h2o] + mass_hash[h_key]
+      h_plus_key = h_plus ? :h_plus : :h
+      extra_add = mass_hash[h_plus_key]
       [:add_N_term_peptide, :add_C_term_peptide].each do |sym|
         if mass_hash.key?(sym)
-          final_add += mass_hash[sym]
+          extra_add += mass_hash[sym]
         end
       end
-      hash_by_aa_string = {}
-      mass_hash.each {|k,v| hash_by_aa_string[k.to_s] = mass_hash[k] }
+      mc = Mass::Calculator.new(mass_hash, extra_add)
+      
+      masses_per_group = pep_aaseqs_ar.map do  |pep_aaseqs|
+        mc.masses(pep_aaseqs)
+      end
 
-      pep_aaseqs_ar.map do  |pep_aaseqs|
-        pep_aaseqs.select do |aaseq|
-          sum = 0.0
-          aaseq.split('').each do |let|
-            if !hash_by_aa_string.key? let
-              puts 'NOT FOUND'
-              p let
-            end
-            sum += hash_by_aa_string[let]
+      masses_per_group.zip(pep_aaseqs_ar).map do |masses, aaseqs|
+        passing = []
+        aaseqs.zip(masses) do |aaseq, mh_plus|
+          if ( (mh_plus >= min_mh) and (mh_plus <= max_mh) )
+            passing << aaseq
           end
-          mh_plus = sum + final_add
-          ( (mh_plus >= min_mh) and (mh_plus <= max_mh) )
         end
+        passing
       end
     end
   end

data/lib/spec_id/mass.rb

@@ -29,13 +29,13 @@ class Mass
     :U => 150.95364,   # (selenocysteine) http://www.matrix-science.com/help/aa_help.html
     :X => 118.805716,  # the average of the mono masses of the 20 amino acids
     :* => 118.805716, # same as X
+    :Z => (129.04259 + 128.05858) / 2,  # average glutamic acid and glutamine
 
     # elements etc.
     :h => 1.00783,
     :h_plus => 1.00728,
     :o => 15.9949146,
     :h2o => 18.01056,
-
   }
   AVG = {
     :A => 71.0788,
@@ -64,6 +64,7 @@ class Mass
     :U => 150.03,   # (selenocysteine) http://www.matrix-science.com/help/aa_help.html
     :X => 118.88603, # the average of the masses of the 20 amino acids
     :* => 118.88603, # same as X
+    :Z => (129.1155+ 128.1307) / 2,  # average glutamic acid and glutamine
 
     # elements etc.
     :h => 1.00794,
@@ -112,5 +113,66 @@ class Mass
     end
     copy_hash
   end
+
+  # returns an array of masses parallel to array passed in
+  # If you want the mass with H+, then pass in the mass as h_plus
+  # The mass hash must repond to 
+  #   :h2o (water)
+  #   and at least the twenty amino acids (by string or symbol)
+  # The mass hash may respond to :add_N_term_peptide or :add_C_term_peptide
+  # in which case these will be added to the final mass
+  def self.masses(aaseqs, mass_hash=Mass::MONO, h_plus=0.0)
+    final_add = mass_hash[:h2o] + h_plus
+    [:add_N_term_peptide, :add_C_term_peptide].each do |sym|
+      if mass_hash.key?(sym)
+        final_add += mass_hash[sym]
+      end
+    end
+    hash_by_aa_string = {}
+    mass_hash.each {|k,v| hash_by_aa_string[k.to_s] = mass_hash[k] }
+
+    aaseqs.map do  |pep_aaseqs|
+      sum = 0.0
+      aaseq.split('').each do |let|
+        sum += hash_by_aa_string[let]
+      end
+      mh_plus = sum + final_add
+    end
+  end
+
+
+end
+
+class Mass::Calculator
+
+  # mass_hash must respond to :h2o or 'h2o'.  This is added to represent the
+  # tails of the peptide.  add_extra is outside of that (e.g., an H+)
+  def initialize(mass_hash, add_extra=0.0)
+    @mass_hash = mass_hash_to_s(mass_hash)
+    @final_add = @mass_hash['h2o'] + add_extra
+  end
+
+  def mass_hash_to_s(mass_hash)
+    new_hash = {}
+    mass_hash.each do |k,v|
+      new_hash[k.to_s] = v
+    end
+    new_hash
+  end
+
+  def masses(aaseqs)
+    aaseqs.map do |aaseq|
+      sum = @final_add  # <- add in the initialization
+      aaseq.split('').each do |let|
+        if @mass_hash.key? let
+          sum += @mass_hash[let]
+        else
+          abort "LETTER not found in mass_hash: #{let}"
+        end
+      end
+      sum
+    end
+  end
+
 end