mspire 0.8.6 → 0.8.6.1

data/VERSION

@@ -1 +1 @@
-0.8.6
+0.8.6.1

data/lib/mspire.rb

@@ -3,4 +3,5 @@ require 'mspire/mass/aa' # requires mspire/mass & therefore mspire/molecular_for
 
 module Mspire
   VERSION = IO.read(File.join(File.dirname(__FILE__), '..', 'VERSION')).chomp
+  CITE = "Prince JT, Marcotte EM. mspire: mass spectrometry proteomics in Ruby. Bioinformatics. 2008. 24(23):2796-7."
 end

data/script/accession_quantifier.rb

@@ -0,0 +1,176 @@
+#!/usr/bin/env ruby
+
+require 'optparse'
+require 'ostruct'
+require 'set'
+
+require 'mspire/mzml'
+require 'mspire/digester'
+require 'mspire/mascot/dat'
+
+
+class Array
+
+  def sum
+    inject( nil ) { |sum,x| sum ? sum + x.to_f : x.to_f }
+  end
+
+  def weighted_mean(weights_array)
+    w_sum = weights_array.sum
+    w_prod = 0
+    self.each_index {|i| w_prod += self[i] * weights_array[i].to_f}
+    w_prod.to_f / w_sum.to_f
+  end
+end
+
+
+opt = OpenStruct.new( {
+  max_rt_before: 60,
+  max_rt_after: 60,
+  mz_window: 0.01,
+  scan_id_regex: Regexp.new("(.*)"),
+  # the regex I use:
+  #scan_id_regex: Regexp.new("id_([^\\.]+)"),
+} )
+
+
+opts = OptionParser.new do |op|
+  op.banner = "usage: #{File.basename(__FILE__)} [OPTS] <mzML> <dat> <accession> ..."
+  op.separator "output: <TBD>"
+  op.separator ""
+  op.separator "options: "
+  op.on("--max_rt_before <#{opt.max_rt_before}>", Float, "(sec) max RT to look before") {v| opt.max_rt_before = v }
+  op.on("--max_rt_after <#{opt.max_rt_after}>", Float, "(sec) max RT to look after") {v| opt.max_rt_after = v }
+  op.on("--mz_window <#{opt.mz_window}>", Float, "(Th) window around m/z value") {|v| opt.mz_window = v }
+  op.on("--scan_id_regex <#{opt.scan_id_regex.source}>", "scan") {|v| opt.scan_id_regex = Regexp.new(v) } 
+  op.on("--add-filename", "adds the filename to output files") {|v| opt.add_filename = v }
+end
+opts.parse!
+
+if ARGV.size < 3
+  puts opts
+  exit
+end
+
+(mzml_file, dat_file, *accessions_array) = ARGV
+
+accessions = Set.new(accessions_array)
+
+# block yields the retention time in seconds and stops iteration if the block returns nil/false
+def create_chromatogram(mzml, index_enum, mz, mz_window, ms_level=1, &block)
+  chromatogram = []
+  while index=index_enum.next
+    break unless spectrum=mzml[index] 
+    next unless ms_level===spectrum.ms_level
+    break unless block.call( spectrum.retention_time )
+    mzs = spectrum.mzs
+    ints = spectrum.intensities
+    index = spectrum.find_nearest_index(mz)
+
+    lwin_mz = mz - (mz_window/2.0)
+    hwin_mz = mz + (mz_window/2.0)
+
+    
+    ints_in_range = []
+    index.upto(Float::INFINITY) do |i| 
+      if mzs[i] <= hwin_mz
+        ints_in_range << ints[i]
+      else
+        break
+      end
+    end
+    (index-1).downto(0) do |i|
+      if mzs[i] >= lwin_mz
+        ints_in_range << ints[i]
+      else
+        break
+      end
+    end
+    if ints_in_range.size > 0
+      chromatogram << [spectrum.retention_time, ints_in_range.reduce(:+)]
+    end
+  end
+  chromatogram
+end
+
+Pephit = Struct.new(:spectrum_id, :exp_mz, :charge, :seq, :accessions, :var_mods_string, :chromatogram)
+
+pephits = []
+Mspire::Mascot::Dat.open(dat_file) do |dat|
+  dat.each_peptide(1) do |pephit|     
+    intersecting_accessions = accessions & pephit.protein_hits_info.map(&:accession)
+    if intersecting_accessions.size > 0
+      query = dat.query(pephit.query_num)
+      z = query.charge
+      exp_mr = pephit.mr + pephit.delta
+      exp_mz = (exp_mr + (z * Mspire::Mass::H_PLUS)) / z
+      md=opt.scan_id_regex.match(query.title)
+      if md
+        spectrum_id = md[1]
+      end
+      pephits << Pephit.new(spectrum_id, exp_mz, z, pephit.seq, intersecting_accessions.to_a, pephit.var_mods_string)
+    end
+  end
+end
+
+puts "Found: #{pephits.size} pephits"
+exit unless pephits.size > 0
+
+Mspire::Mzml.open(mzml_file) do |mzml|
+  spec_index = mzml.index_list[:spectrum]
+
+  tic = mzml.map {|spec| spec.fetch_by_acc('MS:1000285').to_f }.reduce(:+)
+  divisor = tic.to_f/1e7
+  
+  id_to_index = {}
+  spec_index.ids.each_with_index {|id,index| id_to_index[id] = index }
+
+
+  pephits.each do |pephit|
+    print "." ; $stdout.flush
+
+    ms1_spec_id = mzml[pephit.spectrum_id].precursors.first.spectrum_id
+    index = id_to_index[ms1_spec_id]
+    spectrum = mzml[index]
+
+    orig_rt = spectrum.retention_time
+    lo_rt = orig_rt - opt.max_rt_before
+    hi_rt = orig_rt + opt.max_rt_after
+
+    first_chunk = create_chromatogram(mzml, index.downto(0), pephit.exp_mz, opt.mz_window) {|rt| rt >= lo_rt }
+    last_chunk = create_chromatogram(mzml, (index+1).upto(Float::INFINITY), pephit.exp_mz, opt.mz_window) {|rt| rt <= hi_rt }
+
+    chromatogram = (first_chunk + last_chunk).sort
+    chromatogram.each {|pair| pair[1] /= divisor }
+
+    pephit.chromatogram = chromatogram
+  end
+end
+puts "finished with mzml"
+
+pephits.group_by {|pephit| [pephit.seq, pephit.charge, pephit.var_mods_string] }.map do |group, sub_pephits|
+  puts "grouping: #{group.join(', ')}"
+  avg_exp_mz = sub_pephits.map(&:exp_mz).reduce(:+) / sub_pephits.size
+  new_chrom = sub_pephits.flat_map(&:chromatogram).uniq.sort
+  cpephit = Pephit.new("(#{sub_pephits.size})", avg_exp_mz, *[:charge, :seq, :accessions, :var_mods_string].map {|key| sub_pephits.first.send(key) }, new_chrom)
+
+  fileparts = [cpephit.seq, cpephit.charge, cpephit.var_mods_string]
+  if opt.add_filename
+    fileparts.unshift(dat_file.chomp(File.extname(dat_file)))
+  end
+  filename = fileparts.join(".") + ".tsv"
+
+  puts "writing: #{filename}"
+  File.open(filename, 'w') do |out|
+    cpephit.each_pair do |k,v|
+      out.puts "# #{k}: #{v}" unless k.to_sym == :chromatogram
+    end
+    out.puts
+    out.puts "rt(sec)\tnorm_intensity"
+    cpephit.chromatogram.each do |row|
+      out.puts row.join("\t")
+    end
+  end
+end
+
+

data/script/mzml_to_mgf.rb

@@ -3,6 +3,13 @@
 require 'mspire/mzml'
 require 'optparse'
 
+
+# returns '3+' for 3 or '2-' for -2
+def mascot_charge(val)
+  "#{val}#{val > 0 ? '+' : '-'}"
+end
+
+
 opt = {
   filter_zero_intensity: true,
   retention_times: true,
@@ -10,7 +17,7 @@ opt = {
 opts = OptionParser.new do |op|
   op.banner = "usage: #{File.basename($0)} <file>.mzML ..."
   op.separator "outputs: <file>.mgf"
-  #op.on("--no-filter-zeros", "won't remove values with zero intensity") {|v| opt[:filter_zero_intensity] = false }
+  op.on("--no-filter-zeros", "won't remove values with zero intensity") {|v| opt[:filter_zero_intensity] = false }
   # the default is set in ms/msrun/search.rb -> set_opts
   op.on("--no-retention-times", "won't include RT even if available") {|v| opt[:retention_times] = false }
 end
@@ -22,25 +29,35 @@ if ARGV.size == 0
   exit
 end
 
+filter_zeros = opt[:filter_zero_intensity]
+
 ARGV.each do |file|
-  if File.exist?(file)
+  basename = file.chomp(File.extname(file))
+  outfile = basename + ".mgf"
+
+  File.open(outfile, 'w') do |out|
     Mspire::Mzml.foreach(file).with_index do |spectrum,i|
       next unless spectrum.ms_level > 1
-      puts "BEGIN IONS"
+      out.puts "BEGIN IONS"
       # id, spectrumid, 
       rt = spectrum.retention_time
-      title = [i, "id_#{spectrum.id}", "rt_#{rt.round}"].join('.')
-      puts "TITLE=#{title}"
-      puts "RTINSECONDS=#{rt}" if opt[:retention_times]
-      puts "PEPMASS=#{spectrum.precursor_mz}"
-      puts "CHARGE=#{spectrum.precursor_charge}+"
+      title_ar = [i, "id_#{spectrum.id}"]
+      title_ar.push("rt_#{rt.round}") if opt[:retention_times]
+      title = title_ar.join('.')
+      out.puts "TITLE=#{title}"
+      out.puts "RTINSECONDS=#{rt}" if opt[:retention_times]
+      out.puts "PEPMASS=#{spectrum.precursor_mz}"
+      if z=spectrum.precursor_charge
+        out.puts "CHARGE=#{mascot_charge(z)}"
+      end
+
       spectrum.each do |mz,int|
-        puts [mz, int].join(" ")
+        unless filter_zeros && (int==0.0)
+          out.puts([mz, int].join(" ")) 
+        end
       end
-      puts "END IONS"
-      puts ""
+      out.puts "END IONS"
+      out.puts ""
     end
-  else
-    puts "missing file: #{file} [skipping]"
   end
 end

data/script/peptide_hit_qvalues_to_spectral_counts_table.rb

@@ -4,6 +4,7 @@ require 'andand'
 require 'set'
 require 'ruport'
 
+require 'mspire'
 require 'mspire/ident/peptide_hit/qvalue'
 require 'mspire/ident/peptide_hit'
 require 'mspire/ident/protein_group'
@@ -96,11 +97,14 @@ group names can be arbitrarily defined
   opt :qspec_decibans, "report bayesfactor in decibans"
   opt :qspec_normalize, "normalize spectral counts per run", :default => false
   opt :qspec_keep_files, "keep a copy of the files submitted and returned from Qspec", :default => false
+  opt :version_tag, "pass in a version tag (e.g. pass in git describe --tags) for version record", :type => String
   opt :write_subset, "(dev use only) write subset db", :default => false
 end
 
+commandline_incantation = __FILE__ + " " + ARGV.join(" ")
 opt = opts.parse(ARGV)
 opt[:count_type] = opt[:count_type].to_sym
+outfile = opt[:outfile] || outfile
 
 $VERBOSE = opt.delete(:verbose)
 
@@ -271,5 +275,15 @@ if opt[:peptides]
   hits_table.to_tsv(pephits_outfile, :footer => ["parallel to #{outfile}"])
 end
 
-intro = ["samples: #{samplename_to_filename}", "options: #{opt}"]
+intro = [
+  "",
+  "ruby: #{RUBY_VERSION}",
+  "software: mspire #{Mspire::VERSION}", 
+  "cite: #{Mspire::CITE}",
+  "samples: #{samplename_to_filename}", 
+"options: #{opt}", 
+"commandline: #{commandline_incantation}"
+]
+intro.insert(3, "version_tag: #{opt[:version_tag]}") if opt[:version_tag]
+
 counts_table.to_tsv(outfile, :footer => intro)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: mspire
 version: !ruby/object:Gem::Version
-  version: 0.8.6
+  version: 0.8.6.1
   prerelease: 
 platform: ruby
 authors:
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-04-03 00:00:00.000000000 Z
+date: 2013-04-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -314,6 +314,7 @@ files:
 - obo/ms.obo
 - obo/unit.obo
 - schema/peptide_hit_qvalues.pqh.tsv
+- script/accession_quantifier.rb
 - script/download_uniprotkb_db.rb
 - script/fasta_to_peptide_centric_db.rb
 - script/mascot_dat_to_peptide_hit_qvalues.rb