mspire-simulator 0.1.2 → 0.2.0

data/lib/ms/merger.rb

@@ -33,65 +33,78 @@ class Merger
     b = weights.flatten.inject(:+)
     return a/b
   end
-  
+
   def self.merge(spectra,half_range)
-    @start = Time.now
     new_data = {}
     total = spectra.size
     k = 0
+    prog = Progress.new("Merging Overlaps:")
     spectra.each do |rt,val|
-      Progress.progress("Merging Overlaps:",(((k/total)*100).to_i))
+      if k.even?
+        num = (((k/total)*100).to_i)
+        prog.update(num)
+      end
       peaks = val.transpose
-      peaks.sort_by!{|a| a[0]}
+      peaks.sort_by!{|a| a[0]} #mz
       peaks = peaks.transpose
       mzs = peaks[0]
       ints = peaks[1]
       mzs.each_with_index do |mz,i|
-	next if mz.class == Hash
-	o_mz = mz
-	mz = mz.keys[0][0] if mz.class == Hash
-	range = (mz..mz+half_range)
-	if range.include?(mzs[i+1])
-	  metaA_mz = [o_mz, mzs[i+1]]
-	  meta_int = [ints[i],ints[i+1]]
-	  sum = meta_int.flatten.inject(:+).to_f
-	  i1 = ints[i]
-	  i1 = ints[i].flatten.inject(:+) if ints[i].class == Array
-	  frac1 = (i1/sum) * 100
-	  frac2 = (ints[i+1]/sum) * 100
-	  metaB_mz = {[w_avg(metaA_mz,meta_int),frac1,frac2] => metaA_mz}
-	  
-	  mzs[i] = nil; mzs[i+1] = metaB_mz
-	  ints[i] = nil; ints[i+1] = meta_int
-	end
+        next if mz.class == Hash
+        o_mz = mz
+        mz = mz.keys[0][0] if mz.class == Hash
+        range = (mz..mz+half_range)
+        if range.include?(mzs[i+1])
+          metaA_mz = [o_mz, mzs[i+1]]
+          meta_int = [ints[i],ints[i+1]]
+          sum = meta_int.flatten.inject(:+).to_f
+          i1 = ints[i]
+          i1 = ints[i].flatten.inject(:+) if ints[i].class == Array
+          frac1 = (i1/sum) * 100
+          frac2 = (ints[i+1]/sum) * 100
+          metaB_mz = {[w_avg(metaA_mz,meta_int),frac1,frac2] => metaA_mz}
+
+          mzs[i] = nil; mzs[i+1] = metaB_mz
+          ints[i] = nil; ints[i+1] = meta_int
+        end
       end
-      new_data[rt] = [mzs.compact,ints.compact]
+      spec = [mzs.compact,ints.compact]
+      spec.ms_level = val.ms_level
+      spec.ms2 = val.ms2
+      new_data[rt] = spec
       k += 1
     end
-    Progress.progress("Merging Overlaps:",100,Time.now-@start)
-    puts ''
+    prog.finish!
     return new_data
   end
-  
+
   def self.compact(spectra)
     @start = Time.now
     total = spectra.size
     k = 0
+    num = 0
+    prog = Progress.new("Merge Finishing:")
+    step = total/100.0
     spectra.each do |rt,val|
-      Progress.progress("Merge Finishing:",(((k/total)*100).to_i))
+      if k > step * (num + 1)
+        num = (((k/total)*100).to_i)
+        prog.update(num)
+      end
       mzs = val[0]
       ints = val[1]
       mzs.each_with_index do |m,i|
-	if m.class == Hash
-	  mzs[i] = m.keys[0][0]
-	  ints[i] = ints[i].flatten.inject(:+)
-	end
+        if m.class == Hash
+          mzs[i] = m.keys[0][0]
+          ints[i] = ints[i].flatten.inject(:+)
+        end
       end
-      spectra[rt] = [mzs,ints]
+      spec = [mzs,ints]
+      spec.ms_level = val.ms_level
+      spec.ms2 = val.ms2
+      spectra[rt] = spec
       k += 1
     end
-    Progress.progress("Merge Finishing:",100,Time.now-@start)
-    puts ''
+    prog.finish!
     return spectra
   end
 end

data/lib/ms/mzml_wrapper.rb

@@ -6,42 +6,87 @@ require 'mspire/mzml'
 class Mzml_Wrapper
 
   def initialize(spectra)
-  #spectra is a Hash rt=>[[mzs],[ints]]
-    @start = Time.now
-  
-    
+    #spectra is a Hash rt=>[[mzs],[ints]]
+    ms2_count = 0
     count = 0.0
     scan_number = 1
     specs = []
-    spectra.each do |rt,data|
-      Progress.progress("Converting to mzml:",(((count/spectra.size)*100).to_i))
+    prog = Progress.new("Converting to mzml:")
+    num = 0
+    total = spectra.size
+    step = total/100
+    spec_id = nil
+    t_rt = 0
+    spectra.sort.map do |rt,data|
+      if count > step * (num + 1)
+        num = (((count/total)*100).to_i)
+        prog.update(num)
+      end
+      if t_rt > rt
+	puts "OUT of ORDER"
+      end
+      t_rt = rt
+
+      ms_level = data.ms_level # method added to array class
 
       spc = Mspire::Mzml::Spectrum.new("scan=#{scan_number}") do |spec|
-	spec.describe_many!(['MS:1000127', ['MS:1000511', 1]])
-	spec.data_arrays = [
-	  Mspire::Mzml::DataArray.new(data[0]).describe!('MS:1000514'),  
-	  Mspire::Mzml::DataArray.new(data[1]).describe!('MS:1000515')   
-	]
-	spec.scan_list = Mspire::Mzml::ScanList.new do |sl|
-	  scan = Mspire::Mzml::Scan.new do |scan|
-	    scan.describe! 'MS:1000016', rt, 'UO:0000010'
-	  end
-	  sl << scan
-	end
+        spec.describe_many!(['MS:1000127', ['MS:1000511', 1]]) 
+        spec.data_arrays = [
+          Mspire::Mzml::DataArray.new(data[0]).describe!('MS:1000514'),  
+          Mspire::Mzml::DataArray.new(data[1]).describe!('MS:1000515')   
+        ]
+        spec.scan_list = Mspire::Mzml::ScanList.new do |sl|
+          scan = Mspire::Mzml::Scan.new do |scan|
+            scan.describe! 'MS:1000016', rt, 'UO:0000010'
+          end
+          sl << scan
+        end
+      end
+      specs<<spc
+      if ms_level == 2
+        #[rt,[mzs],[ints]]
+        ms2 = data.ms2
+        ms2.each do |data|
+          ms2_count += 1
+          scan_number += 1
+          spc2 = Mspire::Mzml::Spectrum.new("scan=#{scan_number}") do |spec|
+            spec.describe_many!(['MS:1000127', ['MS:1000511', 2]]) 
+            spec.data_arrays = [
+              Mspire::Mzml::DataArray.new(data[1]).describe!('MS:1000514'),  
+              Mspire::Mzml::DataArray.new(data[2]).describe!('MS:1000515')   
+            ]
+            spec.scan_list = Mspire::Mzml::ScanList.new do |sl|
+              scan = Mspire::Mzml::Scan.new do |scan|
+                scan.describe! 'MS:1000016', data[0], 'UO:0000010'
+              end
+              sl << scan
+            end
+            precursor = Mspire::Mzml::Precursor.new( spc.id )
+            si = Mspire::Mzml::SelectedIon.new
+            # the selected ion m/z:
+            si.describe! "MS:1000744", data.pre_mz
+            # the selected ion charge state
+            si.describe! "MS:1000041", data.pre_charge
+            # the selected ion intensity
+            si.describe! "MS:1000042", data.pre_int
+            precursor.selected_ions = [si]
+            spec.precursors = [precursor]
+          end
+          specs<<spc2
+        end
       end
       count += 1
       scan_number += 1
-      specs<<spc
     end
-  
-    
-  
+
+
+
     @mzml = Mspire::Mzml.new do |mzml|
-      mzml.id = 'ms1'
+      mzml.id = 'ms1_and_ms2'
       mzml.cvs = Mspire::Mzml::CV::DEFAULT_CVS
       mzml.file_description = Mspire::Mzml::FileDescription.new  do |fd|
-	fd.file_content = Mspire::Mzml::FileContent.new
-	fd.source_files << Mspire::Mzml::SourceFile.new
+        fd.file_content = Mspire::Mzml::FileContent.new
+        fd.source_files << Mspire::Mzml::SourceFile.new
       end
       default_instrument_config = Mspire::Mzml::InstrumentConfiguration.new("IC").describe!('MS:1000031')
       mzml.instrument_configurations << default_instrument_config
@@ -50,15 +95,15 @@ class Mzml_Wrapper
       default_data_processing = Mspire::Mzml::DataProcessing.new("did_nothing")
       mzml.data_processing_list << default_data_processing
       mzml.run = Mspire::Mzml::Run.new("simulated_run", default_instrument_config) do |run|
-	spectrum_list = Mspire::Mzml::SpectrumList.new(default_data_processing, specs)
-	run.spectrum_list = spectrum_list
+        spectrum_list = Mspire::Mzml::SpectrumList.new(default_data_processing, specs)
+        run.spectrum_list = spectrum_list
       end
     end
-    Progress.progress("Converting to mzml:",100,Time.now-@start)
-    puts ''
+    prog.finish!
+    puts "ms2 written = #{ms2_count}"
     return @mzml
   end
-  
+
   def to_xml(file)
     return @mzml.to_xml(file)
   end

data/lib/ms/noise.rb

@@ -5,47 +5,47 @@ require 'ms/rt/rt_helper'
 module MS
   module Noise
     module_function
-    def noiseify(density,max_mz)
-    # spectra is {rt => [[mzs],[ints]]}
-      @start = Time.now
+    def noiseify(opts,max_mz)
+      # spectra is {rt => [[mzs],[ints]]}
+      desity = opts[:noise_density]
+      max_int = opts[:noiseMaxInt]
+      min_int = opts[:noiseMinInt]
       @noise = {}
       r_times = Sim_Spectra.r_times
-      
-      count = 0.0
+      count = 0
+      prog = Progress.new("Adding noise:")
+      num = 0
+      total = r_times.size
+      step = total/100.0
       r_times.each do |rt|
-      
-	Progress.progress("Adding noise:",(((count/r_times.size)*100).to_i))
-      
-	nmzs = []
-	nints = []
-	
-	density.times do
-	  rmz = RThelper.RandomFloat(0.0,max_mz)
-	  rint = RThelper.RandomFloat(50,1000)
-	  
-	  nmzs<<rmz
-	  nints<<rint
-	end
-	@noise[rt] = [nmzs,nints]
-	count += 1
+        if count > step * (num + 1)
+          num = (((count/total)*100.0).to_i)
+          prog.update(num)
+        end
+        nmzs = []
+        nints = []
+        density.times do
+          rmz = RThelper.RandomFloat(0.0,max_mz)
+          rint = RThelper.RandomFloat(min_int,max_int)
+          nmzs<<rmz
+          nints<<rint
+        end
+        @noise[rt] = [nmzs,nints]
+        count += 1
       end
-      
-      Progress.progress("Adding noise:",100,Time.now-@start)
-      puts ''
-      
+      prog.finish!
       return @noise
     end
-    
-    
+
     def spec_drops(drop_percentage)
       r_times = Sim_Spectra.r_times
       l = r_times.length
       num_drops = drop_percentage * l
       num_drops.to_i.times do 
-	r_times.delete_at(rand(l+1))
+        r_times.delete_at(rand(l+1))
       end
       return r_times
     end
-    
+
   end
 end

data/lib/ms/rt/rt_helper.rb

@@ -1,6 +1,6 @@
 
 module RThelper
-  
+
   module_function
   def normalized_gaussian(x,mu,sd)
     x = x.to_f
@@ -8,7 +8,7 @@ module RThelper
     sd = sd.to_f
     return ((1/(Math.sqrt(2*(Math::PI)*(sd**2))))*(Math.exp(-(((x-mu)**2)/((2*sd)**2)))))
   end
-  
+
   module_function
   def gaussian(x,mu,sd,h)
     x = x.to_f
@@ -17,7 +17,7 @@ module RThelper
     h = h.to_f
     return h*Math.exp(-(x-mu)**2/(sd**2))
   end
-  
+
   module_function
   def RandomFloat(a,b)
     a = a.to_f

data/lib/ms/rt/rtgenerator.rb

@@ -8,74 +8,86 @@ require 'ms/rt/rt_helper'
 
 module MS
   module Rtgenerator
-    
+
     module_function
     def generateRT(peptides, one_d)
-      
-      @start = Time.now
+
       @r_times = Sim_Spectra.r_times
-      
+
       # Gets retention times from the weka model
       peptides = MS::Weka.predict_rts(peptides)
       MS::Weka.predict_ints(peptides)
+
+
+      #-----------------------------------------------------------------
+      prog = Progress.new("Generating retention times:")
+      num = 0
+      total = peptides.size
+      step = total/100.0
       
+      max_rt = 4*(@r_times.max/5)
+      r_end = max_rt + (@r_times.max/5)/2
+      r_start = @r_times.max/5
       
-      #-----------------------------------------------------------------
       peptides.each_with_index do |pep,ind|
-        Progress.progress("Generating retention times:",(((ind+1)/peptides.size.to_f)*100).to_i)
-	
-	
-	#Fit retention times into scan times
-	max_rt = @r_times.max 
-	p_rt = pep.p_rt * 10**-2
-	if p_rt > 1
-	  pep.p_rt = @r_times.max
-	  pep.p_rt_i = @r_times.index(pep.p_rt)
-	else
-	  pep.p_rt = @r_times.find {|i| i >= (p_rt * max_rt)}
-	  pep.p_rt_i = @r_times.index(pep.p_rt)
+	if ind > step * (num + 1)
+	  num = (((ind+1)/total.to_f)*100).to_i
+	  prog.update(num)
 	end
+
+
+        #Fit retention times into scan times
+        p_rt = pep.p_rt * 10**-2
+	percent_time = p_rt 
+	sx = RThelper.gaussian(percent_time,0.5,0.45,1.0) * Math.sqrt(pep.abu) #need to figure out what these values should be
+	pep.sx = sx 
 	
+
+        if p_rt > 1
+          pep.p_rt = @r_times.find {|i| i >= r_end}
+          pep.p_rt_i = @r_times.index(pep.p_rt)
+        else
+          pep.p_rt = @r_times.find {|i| i >= (p_rt * max_rt)}
+          pep.p_rt_i = @r_times.index(pep.p_rt)
+        end
+
         if pep.p_rt == nil
           puts "\n\n\t#{pep} TIME-> #{p_rt*max_rt} :: Peptide not predicted in time range: try increasing run time\n\n."
-	else
-	
-	#Give peptide retention times
-	  head_length = nil
-	  tail_length = nil
-	  if one_d
-	    head_length = 300.0
-	    tail_length = 701
-	  else
-	    head_length = 100.0
-	    tail_length = 300
-	  end
-
-	  a = @r_times.find {|i| i >= (pep.p_rt-head_length)}
-	  b = @r_times.find {|i| i >= (pep.p_rt+tail_length)}
-	  a = @r_times.index(a)
-	  b = @r_times.index(b)
-	  
-	  if a == nil
-	    a = @r_times[0]
-	  end
-	  
-	  if b == nil
-	    b = @r_times[@r_times.length-1]
-	  end
-	  
-	  pep.set_rts(a,b)
+        else
 
-	end
+          #Give peptide retention times
+          head_length = nil
+          tail_length = nil
+          if one_d
+            head_length = 300.0
+            tail_length = 701
+          else
+            head_length = 100.0 * sx
+            tail_length = 300 * sx
+          end
+
+          a = @r_times.find {|i| i >= (pep.p_rt-head_length)}
+          b = @r_times.find {|i| i >= (pep.p_rt+tail_length)}
+          a = @r_times.index(a)
+          b = @r_times.index(b)
+
+          if a == nil
+            a = @r_times[0]
+          end
+
+          if b == nil
+            b = @r_times[@r_times.length-1]
+          end
+
+          pep.set_rts(a,b)
+
+        end
       end
       #-----------------------------------------------------------------
-      
-  
-      Progress.progress("Generating retention times:",100,Time.now-@start)
-      puts ""
-      
+      prog.finish!
+
       return peptides
-      
+
     end    
   end
 end

data/lib/ms/rt/weka.rb

@@ -3,7 +3,7 @@ require 'csv'
 
 module MS
   module Weka
-  #James Dalg
+    #James Dalg
     module_function
     def predict_rts(peptides)
       #mz,charge,intensity,rt,A,R,N,D,B,C,E,Q,Z,G,H,I,L,K,M,F,P,S,T,W,Y,V,J,mass,hydro,pi
@@ -13,12 +13,12 @@ module MS
         data<<pep.aa_counts
       end
       arff = make_rt_arff(Time.now.nsec.to_s,data)
-      
+
       path = Gem.bin_path('mspire-simulator', 'mspire-simulator').split(/\//)
       dir = path[0..path.size-3].join("/")
       system("java weka.classifiers.functions.MultilayerPerceptron -T #{arff} -l #{dir}/lib/weka/M5Rules.model -p 24 > #{arff}.out")
       system("rm #{arff}")
-      
+
       #extract what was predicted by weka model
       file = File.open("#{arff}.out","r")
       count = 0
@@ -31,38 +31,38 @@ module MS
       system("rm #{arff}.out")
       return peptides
     end
-    
-    
-    
+
+
+
     def predict_ints(peptides)
       data = []
       peptides.each do |pep|
-	array = []
-	array<<pep.mono_mz<<pep.charge<<pep.mass<<pep.p_rt
+        array = []
+        array<<pep.mono_mz<<pep.charge<<pep.mass<<pep.p_rt
         data << array.concat(pep.aa_counts)
       end
       arff = make_int_arff(Time.now.nsec.to_s,data)
-      
+
       path = Gem.bin_path('mspire-simulator', 'mspire-simulator').split(/\//)
       dir = path[0..path.size-3].join("/")
       system("java weka.classifiers.trees.M5P -T #{arff} -l #{dir}/lib/weka/M5P.model -p 27 > #{arff}.out")
       system("rm #{arff}")
-      
+
       #extract what was predicted by weka model
       file = File.open("#{arff}.out","r")
       count = 0
       while line = file.gets
         if line =~ /(\d*\.\d{0,3}){1}/
-	  peptides[count].p_int = line.match(/(\d*\.\d{0,3}){1}/)[0].to_f
+          peptides[count].p_int = line.match(/(\d*\.\d{0,3}){1}/)[0].to_f
           count += 1
         end
       end
       system("rm #{arff}.out")
       return peptides
     end
-    
-    
-    
+
+
+
     #James Dalg
     def make_rt_arff(sourcefile, training)
       sourcefile<<".arff"
@@ -105,9 +105,9 @@ module MS
       end
       return sourcefile
     end
-    
-    
-        #James Dalg
+
+
+    #James Dalg
     def make_int_arff(sourcefile, training)
       sourcefile<<".arff"
       File.open(sourcefile, "wb") do |f| # need to cite f.puts (not %Q)? if so http://www.devdaily.com/blog/post/ruby/how-write-text-to-file-ruby-example