protk 1.2.4 → 1.2.5

data/bin/protxml_to_table.rb

@@ -19,6 +19,11 @@ include LibXML
 tool=Tool.new([:explicit_output])
 tool.option_parser.banner = "Convert a protXML file to a tab delimited table.\n\nUsage: protxml_to_table.rb [options] file1.protXML"
 
+# tool.options.proteinid_regex=".*?\|.*?\|(.*)"
+# tool.option_parser.on( '--regex rexpr', 'Regex' ) do |regex| 
+#   tool.options.proteinid_regex=regex
+# end
+
 exit unless tool.check_options 
 
 if ( ARGV[0].nil? )
@@ -48,7 +53,7 @@ end
 
 
 column_headers=[
-	"group_number","group_probability","protein_name",
+	"group_number","group_probability","protein_name","protein_id","indistinguishable_proteins",
 	"protein_probability","coverage","peptides",
 	"num_peptides","confidence"
 ]
@@ -62,13 +67,25 @@ protein_groups.each do |protein_group|
 
 	proteins=protein_group.find("./#{protxml_ns_prefix}protein", protxml_ns)
 
-	proteins.each do |protein|  
+	proteins.each do |protein| 
+
+		indis_proteins=protein.find("./#{protxml_ns_prefix}indistinguishable_protein", protxml_ns)
+		indis_proteins_summary=""
+		indis_proteins.each { |iprot| indis_proteins_summary<<"#{iprot.attributes['protein_name']};" }
+
+		protein_id=""
+		if protein.attributes['protein_name'] =~ /.*?\|.*?\|(.*)/
+			protein_id=protein.attributes['protein_name'].match(/.*?\|.*?\|(.*)/)[1]
+		end
+
 		column_values=[]
 
 		column_values << protein_group.attributes['group_number']
 		column_values << protein_group.attributes['probability']
 
 		column_values << protein.attributes['protein_name']
+		column_values << protein_id
+		column_values << indis_proteins_summary
 		column_values << protein.attributes['probability']
 		column_values << protein.attributes['percent_coverage']
 		column_values << protein.attributes['unique_stripped_peptides']

data/bin/sixframe.rb

@@ -41,9 +41,11 @@ end
 
 inname=ARGV.shift
 
-outfile=File.open("#{inname}.translated.fasta",'w')
+outfile=nil
 if ( tool.explicit_output != nil)
   outfile=File.open(tool.explicit_output,'w')
+else
+  outfile=File.open("#{inname}.translated.fasta",'w')
 end
 
 

data/bin/tandem_search.rb

@@ -149,6 +149,37 @@ def set_option(std_params, tandem_key, value)
   notes[0].content=value
 end
 
+def append_option(std_params, tandem_key, value)
+  notes = std_params.find("/bioml/note[@type=\"input\" and @label=\"#{tandem_key}\"]")
+  if notes.length == 0
+    node = XML::Node.new('note')
+    node["type"] = "input"
+    node["label"] = tandem_key
+    node.content = value
+    std_params.find('/bioml')[0] << node
+  else
+    throw "Exactly one parameter named (#{tandem_key}) is required in parameter file" unless notes.length==1    
+    notes[0].content = append_string(notes[0].content, value)
+  end
+end
+
+def collapse_keys(std_params, tandem_key)
+    mods=std_params.find('/bioml/note[@type="input" and @label="#{tandem_key}"]')
+    if not mods
+      first_mod = mods[0]
+      rest_mods = mods[1..-1]
+      rest_mods.each{ |node| first_mod.content = append_string(first_mod.content, node.content); node.remove!}
+    end
+end
+
+def append_string(first, second)
+  if first.empty?
+    second
+  else
+    "#{first},#{second}"
+  end
+end
+
 def generate_parameter_doc(std_params,output_path,input_path,taxo_path,current_db,search_tool,genv)
   set_option(std_params, "protein, cleavage semi", search_tool.cleavage_semi ? "yes" : "no")
   set_option(std_params, "scoring, maximum missed cleavage sites", search_tool.missed_cleavages)
@@ -301,7 +332,11 @@ def generate_parameter_doc(std_params,output_path,input_path,taxo_path,current_d
     mods=std_params.find('/bioml/note[@type="input" and @id="methionine-oxidation-variable"]')
     mods.each{ |node| node.remove!}        
   end  
-  
+
+  # Merge all remaining id based modification into single modification. 
+  collapse_keys(std_params, "residue, potential modification mass")
+  collapse_keys(std_params, "residue, modification mass")
+
   var_mods = search_tool.var_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject {|e| e.empty? }
   var_mods=var_mods.collect {|mod| decode_modification_string(mod) }
   fix_mods = search_tool.fix_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject { |e| e.empty? }
@@ -313,31 +348,17 @@ def generate_parameter_doc(std_params,output_path,input_path,taxo_path,current_d
   var_mods.each do |vm|
 
     mod_type="potential modification mass"
-    mod_type = "potential modification motif" if ( vm=~/[\[\]\(\)\{\}\!]/ )      
-    mod_id_label = "custom-variable-mod-#{mod_id.to_s}"
-    mod_id=mod_id+1
-    mnode=XML::Node.new('node')
-    mnode["id"]=mod_id_label
-    mnode["type"]="input"
-    mnode["label"]="residue, #{mod_type}"
-    mnode.content=vm
-    
-    root_bioml_node << mnode
+    mod_type = "potential modification motif" if motif?(vm)
+    label="residue, #{mod_type}"
+    append_option(std_params, label, vm)
   end
   
   mod_id=1
   fix_mods.each do |fm|
     mod_type="modification mass"
-    mod_type = "modification motif" if ( fm=~/[\[\]\(\)\{\}\!]/ )      
-    mod_id_label = "custom-fixed-mod-#{mod_id.to_s}"
-    mod_id=mod_id+1
-    mnode=XML::Node.new('node')
-    mnode["id"]=mod_id_label
-    mnode["type"]="input"
-    mnode["label"]="residue, #{mod_type}"
-    mnode.content=fm
-    
-    root_bioml_node << mnode
+    mod_type = "modification motif" if motif?(fm)
+    label="residue, #{mod_type}"
+    append_option(std_params, label, fm)
   end
 
   #p root_bioml_node
@@ -345,6 +366,13 @@ def generate_parameter_doc(std_params,output_path,input_path,taxo_path,current_d
   
 end
 
+def motif?(mod_string)
+  # 124@[ is not a modification motif, it is a residue (N-term) modification,
+  # so when checking if modification is a motif look for paired square brackets.
+  mod_string =~ /[\(\)\{\}\!]/ or mod_string =~ /\[.*\]/
+end
+
+
 def generate_taxonomy_doc(taxo_doc,current_db,search_tool)
 
   taxon_label=taxo_doc.find('/bioml/taxon')
@@ -425,8 +453,8 @@ ARGV.each do |filename|
     # Run the search
     #
     job_params= {:jobid => search_tool.jobid_from_filename(filename)}
-    job_params[:queue]="lowmem"
-    job_params[:vmem]="900mb"
+    job_params[:queue]="sixteen"
+    job_params[:vmem]="12gb"
     code = search_tool.run(cmd,genv,job_params,jobscript_path)
     throw "Command failed with exit code #{code}" unless code==0
   else

data/bin/toppas_pipeline.rb

@@ -15,7 +15,7 @@ require 'libxml'
 
 include LibXML
 
-tool=Tool.new([:explicit_output, :background,:over_write])
+tool=Tool.new([:background,:over_write])
 tool.option_parser.banner = "Execute a toppas pipeline with a single inputs node\n\nUsage: toppas_pipeline.rb [options] input1 input2 ..."
 
 tool.options.outdir = ""
@@ -28,6 +28,11 @@ tool.option_parser.on( '--toppas-file f',"the toppas file to run" ) do |file|
   tool.options.toppas_file = file
 end
 
+tool.options.threads = "1"
+tool.option_parser.on( '--threads t',"Number of threads to use" ) do |tr|
+  tool.options.threads=tr
+end
+
 exit unless tool.check_options 
 
 if ( ARGV[0].nil? )
@@ -67,13 +72,13 @@ throw "outdir is a required parameter" if tool.outdir==""
 throw "toppas-file is a required parameter" if tool.toppas_file==""
 throw "outdir must exist" unless Dir.exist?(tool.outdir)
 
-trf_path = "#{tool.toppas_file}.trf"
+trf_path = "#{Pathname.new(Tempfile.new(tool.toppas_file).path).basename.to_s}.trf"
 
 generate_trf(ARGV,trf_path)
 
 cmd=""
 cmd<<"export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:#{genv.openms_root}/lib;
-#{genv.executepipeline} -in #{Pathname.new(tool.toppas_file).realpath.to_s} -out_dir #{Pathname.new(tool.outdir).realpath.to_s} -resource_file #{Pathname.new(trf_path).realpath.to_s}"
+#{genv.executepipeline} -in #{Pathname.new(tool.toppas_file).realpath.to_s} -out_dir #{Pathname.new(tool.outdir).realpath.to_s} -resource_file #{Pathname.new(trf_path).realpath.to_s} -threads #{tool.threads}"
 
 run_pipeline(genv,tool,cmd,tool.outdir,tool.jobid_from_filename(tool.toppas_file))
 

data/bin/uniprot_annotation.rb

@@ -24,6 +24,11 @@ tool.option_parser.on(  '--id-column num', 'Specify a column for ids (default is
   tool.options.id_column=col.to_i
 end
 
+tool.options.flatfiledb="swissprot"
+tool.option_parser.on(  '--flatfiledb dbname', 'Specify path to a Uniprot flatfile' ) do |dbname|
+  tool.options.flatfiledb=dbname
+end
+
 tool.options.fields=nil
 tool.option_parser.on(  '--fields flds', 'A comma separated list of fields to extract' ) do |flds|
   tool.options.fields=flds
@@ -42,7 +47,7 @@ genv=Constants.new
 
 input_file=ARGV[0]
 
-swissprotdb=SwissprotDatabase.new(genv)
+swissprotdb=SwissprotDatabase.new(genv,tool.flatfiledb)
 
 output_file=nil
 

data/ext/protk/{protk.c → decoymaker/decoymaker.c}

@@ -1,6 +1,4 @@
 #include <ruby.h>
-
-
 /*                                                                                                 */
 /* make_random.c - make random protein sequence database using Markov chain with transitional      */
 /* probabilities from amino acid frequencies in a real database in FASTA format                    */
@@ -25,7 +23,8 @@
 #define MAX_SEQUENCE_LENGTH 20000
 #define MAX_LINE_LENGTH 20000 /* large enough to read in long header lines */
 
-static VALUE protk_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in,VALUE output_file_in,char *prefix_string_in) {
+
+static VALUE decoymaker_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in,VALUE output_file_in,char *prefix_string_in) {
   char *input_file = RSTRING_PTR(input_file_in);
   long sequences_to_generate = NUM2INT(db_length_in);
   char * output_file = RSTRING_PTR(output_file_in);
@@ -148,7 +147,7 @@ static VALUE protk_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in
                 measured_aa_freq[a]++;
       }
   }
-	  else {a=floor(20*(float)rand()/RAND_MAX);MP[a][i]++; measured_aa_freq[a]++;} // replace B, X, Z etc. with random amino acid to preserve size distribution
+    else {a=floor(20*(float)rand()/RAND_MAX);MP[a][i]++; measured_aa_freq[a]++;} // replace B, X, Z etc. with random amino acid to preserve size distribution
   }
   MP[20][pl]++;
       measured_aa_freq[20]++; // MP[20][n] is the number of sequences of length n in the database 
@@ -178,12 +177,12 @@ static VALUE protk_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in
        x=(double)row_sum[j]*((double)rand()/RAND_MAX);
        partial_sum=MP[0][j]; i=1;
        while (partial_sum<x) {partial_sum+=MP[i][j]; i++;}
-	  if (j>=MAX_SEQUENCE_LENGTH) i=21; /* terminate when sequence has reached MAX_SEQUENCE_LENGTH */
+    if (j>=MAX_SEQUENCE_LENGTH) i=21; /* terminate when sequence has reached MAX_SEQUENCE_LENGTH */
        if (i<21)
        {
          random_sequence[j]=AMINO_ACIDS[i-1];j++;generated_aa_freq[i-1]++;
        }
-	  else /* i==21, i.e. protein sequence terminated */
+    else /* i==21, i.e. protein sequence terminated */
        {
          k=0; generated_aa_freq[20]++; generated_pl_sum+=j;
          for(l=0;l<j;l++) 
@@ -196,7 +195,7 @@ static VALUE protk_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in
         }
 
         random_sequence_output[k]='\0';
-	      if (!(k%61)) random_sequence_output[k-1]='\0'; /* remove extra newline for sequence length multiple of 60 */
+        if (!(k%61)) random_sequence_output[k-1]='\0'; /* remove extra newline for sequence length multiple of 60 */
         fprintf(outp,">%srp%li\n%s\n",prefix_string,protein,random_sequence_output);
         break;
       }
@@ -222,14 +221,13 @@ static VALUE protk_make_decoys(VALUE self,VALUE input_file_in,VALUE db_length_in
 
 }
 
-  /* ruby calls this to load the extension */
-void Init_protk(void) {
-  /* assume we haven't yet defined Hola */
-  VALUE klass = rb_define_class("Protk",
-    rb_cObject);
 
-  /* the hola_bonjour function can be called
-   * from ruby as "Hola.bonjour" */
+void Init_decoymaker(void) 
+{
+  VALUE klass = rb_define_class("Decoymaker",rb_cObject);
+
   rb_define_singleton_method(klass,
-    "make_decoys", protk_make_decoys, 4);
+    "make_decoys", decoymaker_make_decoys, 4);
+
+
 }

data/ext/protk/decoymaker/extconf.rb

@@ -0,0 +1,3 @@
+require 'mkmf'
+
+create_makefile('protk/decoymaker')

data/ext/protk/simplealign/extconf.rb

@@ -0,0 +1,3 @@
+require 'mkmf'
+
+create_makefile('protk/simplealign')

data/lib/protk/data/FeatureFinderIsotopeWavelet.ini

@@ -1,10 +1,10 @@
 <?xml version="1.0" encoding="ISO-8859-1"?>
-<PARAMETERS version="1.3" xsi:noNamespaceSchemaLocation="http://open-ms.sourceforge.net/schemas/Param_1_3.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
+<PARAMETERS version="1.4" xsi:noNamespaceSchemaLocation="http://open-ms.sourceforge.net/schemas/Param_1_4.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
   <NODE name="FeatureFinderIsotopeWavelet" description="Detects two-dimensional features in LC-MS data.">
-    <ITEM name="version" value="1.9.0" type="string" description="Version of the tool that generated this parameters file." tags="advanced" />
+    <ITEM name="version" value="1.10.0" type="string" description="Version of the tool that generated this parameters file." tags="advanced" />
     <NODE name="1" description="Instance &apos;1&apos; section for &apos;FeatureFinderIsotopeWavelet&apos;">
-      <ITEM name="in" value="" type="string" description="input file" tags="input file,required" restrictions="*.mzML" />
-      <ITEM name="out" value="" type="string" description="output file" tags="output file,required" restrictions="*.featureXML" />
+      <ITEM name="in" value="" type="string" description="input file" tags="input file,required" supported_formats="*.mzML" />
+      <ITEM name="out" value="" type="string" description="output file" tags="output file,required" supported_formats="*.featureXML" />
       <ITEM name="log" value="" type="string" description="Name of log file (created only when specified)" tags="advanced" />
       <ITEM name="debug" value="0" type="int" description="Sets the debug level" tags="advanced" />
       <ITEM name="threads" value="1" type="int" description="Sets the number of threads allowed to be used by the TOPP tool" />
@@ -12,9 +12,9 @@
       <ITEM name="test" value="false" type="string" description="Enables the test mode (needed for internal use only)" tags="advanced" restrictions="true,false" />
       <NODE name="algorithm" description="Algorithm section">
         <ITEM name="max_charge" value="3" type="int" description="The maximal charge state to be considered." restrictions="1:" />
-        <ITEM name="intensity_threshold" value="3" type="float" description="The final threshold t&apos; is build upon the formula: t&apos; = av+t*sd, where t is the intensity_threshold, av the average intensity within the wavelet transformed signal and sd the standard deviation of the transform. If you set intensity_threshold=-1, t&apos; will be zero.#br#As the &apos;optimal&apos; value for this parameter is highly data dependent, we would recommend to start with -1, which will also extract features with very low signal-to-noise ratio. Subsequently, one might increase the threshold to find an optimized trade-off between false positives and true positives. Depending on the dynamic range of your spectra, suitable value ranges include: -1, [0:10], and if your data features even very high intensity values, t can also adopt values up to around 30. Please note that this parameter is not of an integer type, s.t. you can also use t:=0.1, e.g." />
+        <ITEM name="intensity_threshold" value="-1" type="float" description="The final threshold t&apos; is build upon the formula: t&apos; = av+t*sd, where t is the intensity_threshold, av the average intensity within the wavelet transformed signal and sd the standard deviation of the transform. If you set intensity_threshold=-1, t&apos; will be zero.#br#As the &apos;optimal&apos; value for this parameter is highly data dependent, we would recommend to start with -1, which will also extract features with very low signal-to-noise ratio. Subsequently, one might increase the threshold to find an optimized trade-off between false positives and true positives. Depending on the dynamic range of your spectra, suitable value ranges include: -1, [0:10], and if your data features even very high intensity values, t can also adopt values up to around 30. Please note that this parameter is not of an integer type, s.t. you can also use t:=0.1, e.g." />
         <ITEM name="intensity_type" value="ref" type="string" description="Determines the intensity type returned for the identified features. &apos;ref&apos; (default) returns the sum of the intensities of each isotopic peak within an isotope pattern. &apos;trans&apos; refers to the intensity of the monoisotopic peak within the wavelet transform. &apos;corrected&apos; refers also to the transformed intensity with an attempt to remove the effects of the convolution. While the latter ones might be preferable for qualitative analyses, &apos;ref&apos; might be the best option to obtain quantitative results. Please note that intensity values might be spoiled (in particular for the option &apos;ref&apos;), as soon as patterns overlap (see also the explanations given in the class documentation of FeatureFinderAlgorihtmIsotopeWavelet)." tags="advanced" restrictions="ref,trans,corrected" />
-        <ITEM name="check_ppm" value="true" type="string" description="Enables/disables a ppm test vs. the averagine model, i.e. potential peptide masses are checked for plausibility. In addition, a heuristic correcting potential mass shifts induced by the wavelet is applied." tags="advanced" restrictions="true,false" />
+        <ITEM name="check_ppm" value="false" type="string" description="Enables/disables a ppm test vs. the averagine model, i.e. potential peptide masses are checked for plausibility. In addition, a heuristic correcting potential mass shifts induced by the wavelet is applied." tags="advanced" restrictions="true,false" />
         <ITEM name="hr_data" value="false" type="string" description="Must be true in case of high-resolution data, i.e. for spectra featuring large m/z-gaps (present in FTICR and Orbitrap data, e.g.). Please check a single MS scan out of your recording, if you are unsure." restrictions="true,false" />
         <NODE name="sweep_line" description="">
           <ITEM name="rt_votes_cutoff" value="5" type="int" description="Defines the minimum number of subsequent scans where a pattern must occur to be considered as a feature." tags="advanced" restrictions="0:" />

data/lib/protk/gapped_aligner.rb

@@ -0,0 +1,264 @@
+require 'bio'
+require 'matrix'
+
+class PeptideFragment
+	attr_accessor :start
+	attr_accessor :end
+	attr_accessor :seq
+end
+
+class PeptideToGeneAlignment
+	attr_accessor :gene_seq
+	attr_accessor :pep_seq
+	attr_accessor :trace
+
+	def initialize(gene,peptide,trace)
+		@gene_seq = gene
+		@pep_seq = peptide
+		@trace = trace
+	end
+
+	def inspect
+		descr = "#{@gene_seq}\n"
+
+		pep_triples=""
+		@pep_seq.each_char { |c| pep_triples<<c;pep_triples<<c;pep_triples<<c }
+		
+		# gene_seq_triples=""
+		# Bio::Sequence::NA.new(@gene_seq).translate.each_char do |c| 
+		# 	gene_seq_triples<<c;gene_seq_triples<<c;gene_seq_triples<<c 
+		# end
+
+		# descr << "#{gene_seq_triples}\n"
+		
+		pepi=0
+		@trace.each_with_index do |move, i|  
+			if move==1
+				descr<<"-"
+			elsif move==0
+				descr<<"#{pep_triples[pepi]}"
+				pepi+=1
+			end
+		end
+		descr<<"\n"
+		puts descr
+	end
+
+	def fragments
+		frags=[]
+		in_fragment=false
+		@trace.each_with_index do |move,i|
+			if move==0
+				frags << [i,0] unless in_fragment #Start a fragment
+				in_fragment=true	
+			else
+				frags.last[1]=i-1 if in_fragment #End a fragment
+				in_fragment=false
+			end					
+		end
+		if frags.last[1]==0
+			frags.last[1]=@trace.length-1
+		end
+		frags
+	end
+
+	def gaps
+		gps=[]
+		in_start_end=true
+		in_gap=false
+		@trace.each_with_index do |move, i| 
+			if move==0
+				in_start_end=false
+				if in_gap #Ending a gap
+					gps.last[1]=i
+				end
+				in_gap=false
+			else
+				if !in_start_end && !in_gap #Starting a gap
+					in_gap=true
+					gps<<[i,0]
+				end
+			end				
+		end
+		#Remove gaps that have zero length (Trailing)
+		gps=gps.collect do |gp| 
+			rv=gp
+			if gp[1]==0
+				rv=nil
+			end		
+			rv
+		end
+		gps.compact!
+		gps
+	end
+
+end
+
+# Uses a dynamic programming algorithm (Smith-Waterman like) to align a peptide sequence to a nucleotide.
+# This aligner assumes you are doing protogenomics and just want to assume that
+#    (a) The entire peptide sequence matches (with gaps) to the DNA sequence
+#
+class GappedAligner
+
+	def initialize
+		@big_penalty = -1000000000
+		@gap_open_penalty = -10000
+		@gap_extend_penalty = -1
+		@end_gap_penalty = 0
+		@match_bonus = 400
+
+		@match_move=0
+		@aadel_move=-1
+		@nadel_move=1
+		@triplet_offsets = [[0,-2,-1],[-1,0,-2],[-2,-1,0]]
+	end
+
+	def aa_deletion()
+		return @big_penalty
+	end
+
+	def score_na_deletion(move_type)
+		if move_type==@nadel_move
+			return @gap_extend_penalty
+		end
+		return @gap_open_penalty
+	end
+
+	def score_match(aa,na)
+		if aa==na
+			return @match_bonus		
+		end
+		return @big_penalty
+	end
+
+	def traceback(from_row,from_col,dpmoves)
+		last_move = dpmoves[from_row][from_col]
+		last_row = from_row-1
+		last_col = from_col-1
+		if last_move==@aadel_move
+			last_col+=1
+		elsif last_move==@nadel_move			
+			last_row+=1
+		end
+
+		if last_col==0 && last_row==0
+			return [last_move]
+		else
+			throw "Beyond end of array" if last_col<0 || last_row <0
+
+			return traceback(last_row,last_col,dpmoves).push(last_move)
+		end
+	end
+
+	def next_frame(previous_frame)
+		(previous_frame+1) % 3
+	end
+
+	def translate_na_at(j,frame,gene_seq)
+		rm = j % 3 
+		start_pos=j+@triplet_offsets[rm][frame]
+		if start_pos < 0
+			return '-'
+		else
+			return gene_seq[start_pos,3].translate
+		end
+	end
+
+	def save_matrix(dpmatrix,pep_triples,gene_seq,name)
+		matfile=File.open("#{name}.csv", "w+")
+		matfile.write(",,")
+		gene_seq.each_char { |na| matfile.write("#{na},")  }
+		matfile.write("\n")
+		dpmatrix.each_with_index { |row,ri|  
+			if ri>0
+				matfile.write("#{pep_triples[ri-1]},")
+			else
+				matfile.write(",")
+			end
+			row.each { |col|  
+				matfile.write("#{col},")
+			}
+			matfile.write("\n")
+		}
+		matfile.close()
+	end
+
+	def calculate_dp(pep_seq,gene_seq)
+		gene_seq = Bio::Sequence::NA.new(gene_seq)
+		nrow = pep_seq.length*3+1
+		ncol = gene_seq.length+1
+
+		throw "Peptide sequence is longer than gene" if nrow > ncol
+
+		pep_triples=""
+		pep_seq.each_char { |c| pep_triples<<c;pep_triples<<c;pep_triples<<c }
+
+		dpmoves=Matrix.build(nrow,ncol) {|r,c| 0 }.to_a
+		dpmatrix=Matrix.build(nrow,ncol) { |r,c| 0 }.to_a
+		dpframes=Matrix.build(nrow,ncol) { |r,c| 0 }.to_a
+		# before_gap_positions = Matrix.build(nrow,ncol) { |r,c| 0 }.to_a
+
+		# Boundary conditions
+		(0..(nrow-1)).each { |i| 
+			dpmatrix[i][0] = aa_deletion*i 
+			dpmoves[i][0] = @aadel_move
+		}
+		(0..(ncol-1)).each { |j| 
+			dpmatrix[0][j] = @end_gap_penalty*j 
+			dpmoves[0][j] = @nadel_move
+			dpframes[0][j] = j % 3
+		}
+		dpmoves[0][0]=0
+		dpframes[0][0]=0
+
+		(1..(nrow-1)).each do |i|
+			(1..(ncol-1)).each do |j|
+				aa = pep_triples[i-1]
+
+				translated_na = translate_na_at(j-1,dpframes[i-1][j-1],gene_seq)
+
+				match = score_match(aa,translated_na) + dpmatrix[i-1][j-1]
+
+				nadel = score_na_deletion(dpmoves[i][j-1]) + dpmatrix[i][j-1]
+
+				# if (translated_na=="R") && (pep_seq=="FR") && (aa == "R")
+					# require 'debugger';debugger					
+				# end
+
+				if match >= nadel
+					dpmatrix[i][j] = match					
+					dpmoves[i][j] = @match_move
+					dpframes[i][j] = dpframes[i-1][j-1]
+				else
+					dpmatrix[i][j] = nadel
+					dpmoves[i][j] = @nadel_move
+					dpframes[i][j] = next_frame(dpframes[i][j-1])
+				end
+
+			end
+		end
+
+		# Find best end-point
+		end_score = dpmatrix[nrow-1].max
+		end_j = dpmatrix[nrow-1].index(end_score)
+
+		save_matrix(dpmatrix,pep_triples,gene_seq,"dpmatrix")
+		save_matrix(dpmoves,pep_triples,gene_seq,"moves")
+		save_matrix(dpframes,pep_triples,gene_seq,"frames")
+#		require 'debugger';debugger
+
+		traceback(nrow-1,end_j,dpmoves)
+	end
+
+
+	def align pep_seq, gene_seq
+
+		trace = calculate_dp(pep_seq,gene_seq)
+		alignment = PeptideToGeneAlignment.new(gene_seq,pep_seq,trace)
+		# puts alignment
+		# require 'debugger';debugger
+
+		return alignment
+	end
+
+end