mspire-simulator 0.2.1 → 0.3.0

data/lib/ms/sim_modifications.rb

@@ -0,0 +1,40 @@
+
+require 'obo/ontology'
+
+class Modifications
+  def initialize(mods)
+    @modifications = mods
+    @modifications = @modifications.split(/_/)
+    if @modifications[0] != "false"
+      get_mods
+    end
+    return @modifications
+  end
+
+  def get_mods()
+    mods = {}
+    obo = Obo::Ontology.new(Obo::Ontology::DIR + '/mod.obo')
+    @modifications.each do |mod|
+      diff = nil
+      residue = mod[9..mod.size-1]
+      mod = (obo.id_to_element[mod[0..8]]).tagvalues
+      xref = mod['xref']
+      xref.each do |x|
+        if x =~ /DiffFormula/
+          diff = (x.split(/"/))[1]
+        end
+      end
+      if mods[residue] == nil
+        mods[residue] = [[mod['id'][0],diff]]
+      else
+        mds = mods[residue]
+        mds<<[mod['id'][0],diff]
+        mods[residue] = mds
+      end
+    end
+    @modifications = mods
+  end
+  
+  attr_reader :modifications
+  attr_writer :modifications
+end

data/lib/ms/sim_peptide.rb

@@ -1,117 +1,45 @@
-
 require 'mspire/isotope/distribution'
 
 module MS
   class Peptide
-    def initialize(sequence, charge, abu = 1.0)
+    def initialize(sequence, charge, abu = 1.0,db,id,prot_id,modifications)
       @abu = abu
       @p_rt = 0
       @p_int = 0
       @rts = []
-      @charge = charge #this is saved in the file name as well
+      @charge = charge 
+      
+      @mods = modifications
+
+      spec = calcSpectrum(sequence)
 
-      spec = calcSpectrum(sequence, @charge)
+      # TODO Ryan: alter this to handle variable and static mass modifications...Add it from the Katamari code
 
-      # TODO Ryan: alter this to handle variable and static mass modifications... Add it from the Katamari code
+      #core mzs, ints
+      db.execute "INSERT INTO core_spec VALUES(#{id},'#{spec.mzs}','#{spec.intensities}')"
 
-      @core_ints = spec.intensities.clone
-      @core_mzs = spec.mzs.clone
-      @mzs_file = ".m/#{sequence[0]}/#{sequence[0...15]}_#{charge}"
-      @ints_file = ".i/#{sequence[0]}/#{sequence[0...15]}_#{charge}"
-      file = File.open(@mzs_file, "w")
-      file.puts(sequence)
-      file.close
       @mono_mz = spec.mzs[spec.intensities.index(spec.intensities.max)]
       @mass = @mono_mz * @charge
       #U,O,X ???
+      @aa_counts = []
+      stm = "INSERT INTO aac VALUES(#{id},"
       amino_acids = ['A','R','N','D','B','C','E','Q','Z','G','H','I',
         'L','K','M','F','P','S','T','W','Y','V','J']
-      @aa_counts = amino_acids.map do |aa|
-        sequence.count(aa)
+      amino_acids.map do |aa|
+        count = sequence.count(aa)
+        stm<<"#{count},"
+        count
       end
-      @aa_counts<<0.0
-    end
-
-    attr_reader :mass, :charge, :mono_mz, :core_mzs, :p_rt, :p_int, :core_ints, :hydro, :pi, :aa_counts, :p_rt_i, :abu, :sx
-    attr_writer :mass, :charge, :mono_mz, :core_mzs, :p_rt, :p_int, :core_ints, :hydro, :pi, :aa_counts, :p_rt_i, :abu, :sx
-
-    def to_s
-      file = File.open(@mzs_file,"r")
-      seq = file.gets.chomp
-      file.close
-      "Peptide: #{seq}"
-    end
-
-    def sequence
-      file = File.open(@mzs_file,"r")
-      seq = file.gets.chomp
-      file.close
-      seq
+      stm<<"0.0)" #place holder for predicted values
+      stm = db.prepare(stm)
+      stm.execute
+      stm.close if stm
+      db.execute "INSERT INTO peptides VALUES(#{id},'#{sequence}', #{@mass}, #{charge}, #{@mono_mz}, #{@p_rt},NULL, #{@p_int}, #{@abu}, NULL,NULL,NULL,#{prot_id})"
     end
 
-    #---------------------------------------------------------------------------
-    def ints
-      file = File.open(@ints_file, "r")
-      line = file.gets.chomp.split(/;/)
-      file.close
-      ints = []
-      line.each do |iso|
-        ints<<iso.chomp.split(/,/).map!{|fl| fl.to_f}
-      end
-      return ints
-    end
-
-    def insert_ints(arr)
-      file = File.open(@ints_file, "a")
-      arr.each do |val|
-        file.print("#{val},")
-      end
-      file.print(";")
-      file.close
-    end
-
-    def mzs
-      file = File.open(@mzs_file, "r")
-      line = file.gets
-      line = file.gets.chomp.split(/;/)
-      file.close
-      mzs = []
-      line.each do |iso|
-        mzs<<iso.chomp.split(/,/).map!{|fl| fl.to_f}
-      end
-      return mzs
-    end
-
-    def insert_mzs(arr)
-      file = File.open(@mzs_file, "a")
-      arr.each do |val|
-        file.print("#{val},")
-      end
-      file.print(";")
-      file.close
-    end
-
-    def rts
-      return Sim_Spectra::r_times[@rts[0]..@rts[1]]
-    end
-
-    def set_rts(a,b)
-      @rts = [a,b]
-    end
-
-    def delete
-      if File.exists?(@mzs_file)
-        File.delete(@mzs_file)
-      end
-      if File.exists?(@ints_file)
-        File.delete(@ints_file)
-      end
-    end
-    #---------------------------------------------------------------------------
-
     # Calculates theoretical specturm
     #
-    def calcSpectrum(seq, charge)
+    def calcSpectrum(seq)
       #isotope.rb from Dr. Prince
       atoms = countAtoms(seq)
 
@@ -131,7 +59,7 @@ module MS
       var<<"Se"
       var<<atoms[6].to_s
 
-      mf = Mspire::MolecularFormula.from_string(var, charge)
+      mf = Mspire::MolecularFormula.from_string(var, @charge)
       spec = Mspire::Isotope::Distribution.spectrum(mf, :max, 0.001)
 
       spec.intensities.map!{|i| i = i*100.0}
@@ -143,6 +71,7 @@ module MS
     # Counts the number of each atom in the peptide sequence.
     #
     def countAtoms(seq)
+      atom_indexes = {'O' => 0,'N' => 1,'C' => 2,'H' => 3,'S' => 4,'P' => 5,'Se' => 6}
       o = 0
       n = 0
       c = 0
@@ -150,9 +79,12 @@ module MS
       s = 0
       p = 0
       se = 0
+      @charge.times {h += 1}
+      atom_counts = [(o + 1),n,c,(h + 2),s,p,se]
+      
       seq.each_char do |aa|
-
         #poly amino acids
+        #maybe in the future ignore fringe case amino acids
         #"X" is for any (I exclude uncommon "U" and "O")
         if aa == "X"
           aas = Mspire::Isotope::AA::ATOM_COUNTS.keys[0..19]
@@ -166,20 +98,46 @@ module MS
           aas = ["Q","E"]
           aa = aas[rand(2)]
         end
+        
+        #perform modification for residue
+        if @mods != nil
+          if @mods[aa] != nil
+            mods = @mods[aa]
+            mods.each do |mod|
+              mod[1].split(/\s/).each_slice(2) do |sl|
+                atom_counts[atom_indexes[sl[0]]] = atom_counts[atom_indexes[sl[0]]] + sl[1].to_i 
+              end
+            end
+          elsif seq[0] == aa and @mods["CT"] != nil#N-terminus
+            mods = @mods["CT"]
+            mods.each do |mod|
+              mod[1].split(/\s/).each_slice(2) do |sl|
+                atom_counts[atom_indexes[sl[0]]] = atom_counts[atom_indexes[sl[0]]] + sl[1].to_i 
+              end
+            end
+          elsif seq[-1] == aa and @mods["NT"] != nil#C-terminus
+            mods = @mods["NT"]
+            mods.each do |mod|
+              mod[1].split(/\s/).each_slice(2) do |sl|
+                atom_counts[atom_indexes[sl[0]]] = atom_counts[atom_indexes[sl[0]]] + sl[1].to_i 
+              end
+            end
+          end
+        end
 
         if aa !~ /A|R|N|D|C|E|Q|G|H|I|L|K|M|F|P|S|T|W|Y|V|U|O/
           puts "No amino acid match for #{aa}"
         else
-          o = o + Mspire::Isotope::AA::ATOM_COUNTS[aa][:o]
-          n = n + Mspire::Isotope::AA::ATOM_COUNTS[aa][:n]
-          c = c + Mspire::Isotope::AA::ATOM_COUNTS[aa][:c]
-          h = h + Mspire::Isotope::AA::ATOM_COUNTS[aa][:h]
-          s = s + Mspire::Isotope::AA::ATOM_COUNTS[aa][:s]
-          p = p + Mspire::Isotope::AA::ATOM_COUNTS[aa][:p]
-          se = se + Mspire::Isotope::AA::ATOM_COUNTS[aa][:se]
+          atom_counts[0] = atom_counts[0] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:o]
+          atom_counts[1] = atom_counts[1] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:n]
+          atom_counts[2] = atom_counts[2] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:c]
+          atom_counts[3] = atom_counts[3] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:h]
+          atom_counts[4] = atom_counts[4] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:s]
+          atom_counts[5] = atom_counts[5] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:p]
+          atom_counts[6] = atom_counts[6] + Mspire::Isotope::AA::ATOM_COUNTS[aa][:se]
         end
       end
-      return (o + 1),n,c,(h + 2) ,s,p,se
+      return atom_counts
     end
   end
 end

data/lib/ms/sim_spectra.rb

@@ -6,8 +6,8 @@ require 'ms/sim_feature'
 
 module MS
   class Sim_Spectra
-    def initialize(peptides,opts,one_d = false)
-      @data
+    def initialize(opts,one_d = false,db)
+      @opts = opts
       @max_mz
       sampling_rate = opts[:sampling_rate]
       run_time = opts[:run_time]
@@ -17,53 +17,39 @@ module MS
       @@r_times = []
       num_of_spec = sampling_rate*run_time
       spec_time = 1/sampling_rate
-      num_of_spec.to_i.times do
+      num_of_spec.to_i.times do |k|
         @@r_times<<spec_time+RThelper.RandomFloat(-var,var)
         spec_time = spec_time + (1/sampling_rate)
       end
       @@r_times = MS::Noise.spec_drops(drop_percentage)
 
-      pre_features = MS::Rtgenerator.generateRT(peptides,one_d)
+      MS::Rtgenerator.generateRT(one_d,db)
 
       #Features
-      features_o = MS::Sim_Feature.new(pre_features,opts,one_d)
-      @features = features_o.features
-      @data = features_o.data
-      @max_mz = features_o.max_mz
-      @spectra = @data.clone
-
-      @noise = nil
+      @features_o = MS::Sim_Feature.new(opts,one_d,db)
+      @max_mz = @features_o.max_mz
 
     end
 
-    def noiseify
-      @noise = MS::Noise.noiseify(opts,@max_mz)
-
-      @@r_times.each do |k|
-        s_v = @data[k]
-        n_v = @noise[k]
-        if s_v != nil
-          @spectra[k] = [s_v[0]+n_v[0],s_v[1]+n_v[1]]
-        else
-          @spectra[k] = [n_v[0],n_v[1]]
+    def noiseify(db)
+      @noise = MS::Noise.noiseify(@opts,@max_mz)
+      cent_id = @features_o.cent_id + 1
+      @noise.each do |key,val|
+        mzs = val[0]
+        ints = val[1]
+        mzs.each_with_index do |mz,index|
+          db.execute "INSERT INTO spectra VALUES(#{cent_id},NULL,#{key},#{mz},#{ints[index]},NULL)"
+          cent_id += 1
         end
       end
-
-      return @noise
     end
 
     def self.r_times
       @@r_times
     end
 
-    attr_reader :data, :max_mz, :spectra, :noise, :features
-    attr_writer :data, :max_mz, :spectra, :noise, :features
+    attr_reader :max_mz
+    attr_writer :max_mz
 
   end
 end
-
-#charge ratio: take both charge states, determine pH effective
-#more small peaks from lesser charge states
-
-#one_d
-#fit to other labs data - different machine

data/lib/ms/sim_trollop.rb

@@ -27,19 +27,20 @@ module MS
                 trypsin,\n \t\tv8_e_trypsin,
                 v8_de_trypsin",
                 :default => "trypsin" 
+                opt :missed_cleavages, "Number of missed cleavages during digestion", :default => 2
                 opt :sampling_rate, "How many scans per second", :default => 0.5 
                 opt :run_time, "Run time in seconds", :default => 1000.0 
                 opt :noise, "Noise on or off", :default => "true"
                 opt :noise_density, "Determines the density of white noise", :default => 10
-		opt :noiseMaxInt, "The max noise intensity level", :default => 1000
-		opt :noiseMinInt, "The minimum noise intensity level", :default => 50
+                opt :noiseMaxInt, "The max noise intensity level", :default => 1000
+                opt :noiseMinInt, "The minimum noise intensity level", :default => 50
                 opt :pH, "The pH that the sample is in - for determining charge", :default => 2.6
                 opt :out_file, "Name of the output file", :default => "test.mzml"
                 opt :contaminants, "Fasta file containing contaminant sequences", :default => "testFiles/contam/hum_keratin.fasta"
                 opt :dropout_percentage, "Defines the percentage of random dropouts in the run. 0.0 <= percentage < 1.0", :default => 0.01
                 opt :shuffle, "Option shuffles the scans to simulate 1d data", :default => "false"
                 opt :one_d, "Turns on one dimension simulation; run_time is automatically set to 300.0", :default => "false"
-                opt :truth, "Determines truth file type; false gives no truth file; one of: xml or csv", :default => "false"
+                opt :truth, "Determines truth file type; false gives no truth file; one of: 'xml' or 'csv' or 'xml_csv' (for both)", :default => "csv"
                 opt :front, "Fronting chromatography parameter", :default => 6.65
                 opt :tail, "Tailing chromatography parameter", :default => 0.30
                 opt :mu, "Expected value of the chromatography curve", :default => 25.0
@@ -53,7 +54,11 @@ module MS
                 opt :mzml, "Mzml file to extract simulation parameters from", :default => "nil"
                 opt :generations, "If an mzml file is provided this specifies the number of generations for the curve fitting algorithm", :default => 30000
                 opt :mass_label, "Specify a mass tag pattern", :default => 0
-                opt :modifications, "Use a specific modifications file, or read them from a header of the fasta file, perhaps... TBD..."
+                opt :ms2s, "Number of peptide ms2s to perform on each scan", :default => 1
+                opt :ms2, "Turn on/off ms2 (true == on)", :default => "true"
+                opt :databaseName, "Name of database file", :default => "peptides_[Time.now.sec]"
+                opt :memory, "Determines whether to store the database in memory or write to file (false == write to file) Note: if true no database file will be accessible after simulation", :default => "false"
+                opt :modifications, "To define residue or termini modifications. Enter a string Id1R1_Id2R2_ ... where Idi is a modification Id from http://psidev.cvs.sourceforge.net/viewvc/psidev/psi/mod/data/PSI-MOD.obo and Ri is the residue/terminus to apply it to (c-term = CT, n-term = NT)", :default => "false"
 
       end
 

data/lib/ms/tr_file_writer.rb

@@ -1,188 +1,71 @@
 require 'progress'
 
-#if m/z value is in "[m/z, percentage contributed to peak]" it's a
-#merged peak. 
 module MS
   class Txml_file_writer
-    def self.write(features,spectra,file_name)
-      @spectra = spectra
+    def self.write(db,file_name,opts)
+      prog = Progress.new("Writing xml:")
       file = File.open("#{file_name}_truth.xml","w")
-
-      r_times = spectra.keys.sort
+      peps = db.execute "SELECT * FROM peptides"
 
       file.puts "<?xml version=\"1.0\" encoding=\"UTF-8\"?>"
       file.puts "<simulated_peptides>"
-      total = features.size.to_f
-      prog = Progress.new("Writing xml:")
+      file.puts "<simulator_options>\n"
+      opts.each do |k,v|
+        file.puts "\t#{k}=#{v},"
+      end
+      file.puts "</simulator_options>\n"
+      total = peps.size.to_f
+      
       num = 0
       step = total/100.0
-      features.each_with_index do |fe,k|
-        sequence = fe.sequence
-        charge = fe.charge
-        mzs = fe.mzs
-        ints = fe.ints
-        rts = fe.rts
+      
+      peps.each do |pep|
+        k = pep[0]
         if k > step * (num + 1)
           num = (((k/total)*100).to_i)
           prog.update(num)
         end
+        sequence = pep[1]
+        charge = pep[3]
+        cents = db.execute "SELECT * FROM spectra WHERE pep_id=#{k}"
+        
         file.puts "\t<simulated_peptide sequence=\"#{sequence}\" charge=\"#{charge.round}\">"
-        mzs.each_with_index do |mzs,i|
-          tags = ""
-          centroids = ""
-          tags<<"\t\t<lc_centroids isotopic_index=\"#{i}\">"
-          mzs.each_with_index do |mz,ind|
-            if ints[i][ind] > 0.9
-              index = get_ind(mz,rts[ind])
-              centroids<<"#{r_times.index(rts[ind])},#{index.inspect};"
-            end
-          end
-          if centroids != ""
-            tags<<centroids
-            tags<<"</lc_centroids>\n"
-            file<<tags
-          end
+        tags = ""
+        tags<<"\t\t<centroids>\n"
+        centroids = ""
+        cents.each do |cent|
+          centroids<<"\t\t\tcent_id=#{cent[0]},pep_id=#{cent[1]},rt=#{cent[2]},mz=#{cent[3]},int=#{cent[4]},merge_id=#{cent[5]}\n"
         end
+        tags<<centroids
+        tags<<"\t\t</centroids>\n"
+        file<<tags
         file.puts "\t</simulated_peptide>"
       end
       file.puts "</simulated_peptides>"
       file.close
-
       prog.finish!
     end
-
-    def self.get_ind(mz,rt)
-      index = nil
-      if @spectra[rt] != nil
-        mzs = @spectra[rt][0]
-        ints = @spectra[rt][1]
-        mzs.each_with_index do |m, i|
-          if m == mz
-            index = i
-          elsif m.class == Hash
-            if ind = m.values[0].index(mz)
-              index = [i,m.keys[0][ind+1]]
-            end
-          end
-        end
-      end
-      return index
-    end
   end
 
   class Tcsv_file_writer
-    def self.write(full_spectra,spectra,noise,features,file_name)
-      @spectra = full_spectra
-
-      #create indices for real peaks
-      ind_hash = create_indicies(features)
-
-      #create data structure with indices
-      data = data_with_indicies(full_spectra,spectra,noise,ind_hash)
-
-      #group by retention time
-      data = data.group_by{|d| d[0]}
+    def self.write(db,file_name,opts)
+      prog = Progress.new("Writing csv:")
+      spectra = db.execute "SELECT * FROM spectra AS S JOIN peptides AS P ON S.pep_id=P.Id"
+      total = spectra.size
 
       #write
       file = File.open("#{file_name}_truth.csv","w")
-      file.puts "rt,mz,int,index"
-      total = data.size.to_f
+      file.puts "simulator_options=#{opts.inspect}"
+      file.puts "rt,mz,int,centroid_id,merge_id,peptide_id,protien_id,seq,charge,abu"
       count = 0
-      prog = Progress.new("Writing csv(process 2 of 2):")
+      
       num = 0
       step = total/100
-      data.each_value do |val|
-        if count > step * (num + 1)
-          num = (((count/total)*100).to_i)
-          prog.update(num)
-        end
-        val.each do |a|
-          if a[3] >= 1
-            file.puts "#{a[0]},#{a[1]},#{a[2]},#{a[3]}"
-          else
-            file.puts "#{a[0]},#{a[1]},#{a[2]},#{0}"
-          end
-        end
-        count += 1
+      spectra.each do |cent|
+        file.puts "#{cent[2]},#{cent[3]},#{cent[4]},#{cent[0]},#{cent[5]},#{cent[1]},#{cent[18]},#{cent[7]},#{cent[9]},#{cent[14]}"
       end
       file.close
       prog.finish!
     end
-
-    def self.get_merged_mz(mz,rt)
-      m_mz = nil
-      int = nil
-      mzs = @spectra[rt][0]
-      ints = @spectra[rt][1]
-      mzs.each_with_index do |m, i|
-        if m == mz
-          m_mz = mz
-          int = ints[i]
-        elsif m.class == Hash
-          if ind = m.values[0].index(mz)
-            m_mz = [m.keys[0][0],m.keys[0][ind+1]]
-            int = ints[i].flatten.inject(:+)
-          end
-        end
-      end
-      return m_mz,int
-    end
-
-    def self.create_indicies(features)
-      ind_hash = {}
-      features.each_with_index do |pep,i|
-        pep.mzs.each_with_index do |m_ar,j|
-          m_ar.each do |mz|
-            ind_hash[mz] = "#{i + 1}.#{j + 1}".to_f
-          end
-        end
-      end
-      return ind_hash
-    end
-
-    def self.data_with_indicies(full_spectra,spectra,noise,ind_hash)
-      count = 1
-      time_i = 0.0
-      data = []
-      total = spectra.length
-      prog = Progress.new("Writing csv(process 1 of 2):")
-      num = 0
-      step = total/100
-      spectra.each do |k,v|
-        if time_i > step * (num + 1)
-          num = (((time_i/total)*100).to_i)
-          prog.update(num)
-        end
-
-        merged_d = full_spectra[k]
-        merged_mzs = merged_d[0]
-        merged_ints = merged_d[1]
-
-        if noise != "false"
-          n_data = noise[k]
-        end
-
-        if v != nil
-          v.each_slice(2) do |m,i|
-            m.each_with_index do |mz,index|
-              peak_index = ind_hash[mz]
-              mz,int = get_merged_mz(mz,k)
-              data<<[k,mz.inspect,int,peak_index]
-            end
-          end
-        end
-
-        if noise != "false"
-          n_data.each_slice(2) do |m,i|
-            m.each_with_index do |mz,index|
-              mz,int = get_merged_mz(mz,k)
-              data<<[k,mz.inspect,int,0]
-            end
-          end
-        end
-        time_i += 1
-      end
-      return data
-    end
   end
 end