protk 1.2.6.pre5 → 1.3.0.pre1

data/lib/protk/search_tool.rb

@@ -13,6 +13,15 @@ require 'optparse'
 require 'pathname'
 require 'protk/tool'
 
+class FastaDatabase
+  attr :name
+  attr :path
+  def initialize(name,path)
+    @name=name
+    @path=path
+  end
+end
+
 class SearchTool < Tool
 
   # Initializes commandline options common to all search tools.
@@ -21,158 +30,104 @@ class SearchTool < Tool
   def initialize(option_support=[])
     super(option_support)
 
-    if (option_support.include? :database)          
-      @options.database = "sphuman"
-      @option_parser.on( '-d', '--database dbname', 'Specify the database to use for this search. Default=sphuman' ) do |dbname|
-        options.database = dbname
-      end
-
+    if (option_support.include? :database)
+      add_value_option(:database,"sphuman",['-d', '--database dbname', 'Specify the database to use for this search. Can be a named protk database or the path to a fasta file'])        
     end
     
     if ( option_support.include? :enzyme )
-       @options.enzyme = "Trypsin"
-       @option_parser.on('--enzyme enz', 'Enzyme') do |enz|
-         @options.enzyme=enz
-       end
+      add_value_option(:enzyme,"Trypsin",['--enzyme enz', 'Enzyme'])
     end
 
     if ( option_support.include? :modifications )
-      @options.var_mods = ""
-      @option_parser.on('--var-mods vm', 'Variable modifications' ) do |vm|
-        @options.var_mods = vm
-      end
-
-      @options.fix_mods = ""
-      @option_parser.on('--fix-mods fm', 'Fixed modifications' ) do |fm|
-        @options.fix_mods = fm
-      end
+      add_value_option(:var_mods,"",['--var-mods vm','Variable modifications. These should be provided in a comma separated list'])
+      add_value_option(:fix_mods,"",['--fix-mods fm','Fixed modifications. These should be provided in a comma separated list'])
     end
 
     if ( option_support.include? :instrument )
-      @options.instrument = "ESI-QUAD-TOF"
-      @option_parser.on('--instrument instrument', 'Instrument') do |instrument|
-       @options.instrument=instrument
-     end
+      add_value_option(:instrument,"ESI-QUAD-TOF",['--instrument instrument', 'Instrument'])
     end
 
     if ( option_support.include? :mass_tolerance_units )
-      @options.fragment_tolu="Da"
-      @option_parser.on('--fragment-ion-tol-units tolu', 'Fragment ion mass tolerance units (Da or mmu). Default=Da' ) do |tolu|
-        @options.fragment_tolu = tolu
-      end
-        
-      @options.precursor_tolu="ppm"
-      @option_parser.on('--precursor-ion-tol-units tolu', 'Precursor ion mass tolerance units (ppm or Da). Default=ppm' ) do |tolu|
-        @options.precursor_tolu = tolu
-      end
+      add_value_option(:fragment_tolu,"Da",['--fragment-ion-tol-units tolu', 'Fragment ion mass tolerance units (Da or mmu). Default=Da'])      
+      add_value_option(:precursor_tolu,"ppm",['--precursor-ion-tol-units tolu', 'Precursor ion mass tolerance units (ppm or Da). Default=ppm'])
     end
 
     if ( option_support.include? :mass_tolerance )
-      @options.fragment_tol=0.65
-      @option_parser.on( '-f', '--fragment-ion-tol tol', 'Fragment ion mass tolerance (unit dependent). Default=0.65' ) do |tol|
-        @options.fragment_tol = tol
-      end
-      
-      @options.precursor_tol=200
-      @option_parser.on( '-p', '--precursor-ion-tol tol', 'Precursor ion mass tolerance in (ppm if precursor search type is monoisotopic or Da if it is average). Default=200' ) do |tol|
-        @options.precursor_tol = tol.to_f
-      end
+      add_value_option(:fragment_tol,0.65,['-f', '--fragment-ion-tol tol', 'Fragment ion mass tolerance (unit dependent). Default=0.65'])
+      add_value_option(:precursor_tol,200,['-p','--precursor-ion-tol tol', 'Precursor ion mass tolerance. Default=200'])
     end
     
     if ( option_support.include? :precursor_search_type )
-      @options.precursor_search_type="monoisotopic"
-      @option_parser.on( '-a', '--search-type type', 'Use monoisotopic or average precursor masses. (monoisotopic or average)' ) do |type| 
-        @options.precursor_search_type = type
-      end
+      add_value_option(:precursor_search_type,"monoisotopic",['-a', '--search-type type', 'Use monoisotopic or average precursor masses. (monoisotopic or average)'])
     end
 
     if ( option_support.include? :strict_monoisotopic_mass )
-      @options.strict_monoisotopic_mass=false
-      @option_parser.on( '-s', '--strict-monoisotopic-mass', 'Dont allow for misassignment of monoisotopic mass to another isotopic peak') do
-        @options.strict_monoisotopic_mass=true
-      end
+      add_boolean_option(:strict_monoisotopic_mass,false,['-s', '--strict-monoisotopic-mass', 'Dont allow for misassignment of monoisotopic mass to another isotopic peak'])
     end
 
     if ( option_support.include? :missed_cleavages )
-      @options.missed_cleavages=2
-      @option_parser.on( '-v', '--num-missed-cleavages num', 'Number of missed cleavages allowed' ) do |num| 
-        @options.missed_cleavages = num
-      end
+      add_value_option(:missed_cleavages,2,['-v', '--num-missed-cleavages num', 'Number of missed cleavages allowed'])
     end
 
     if ( option_support.include? :cleavage_semi )
-      @options.cleavage_semi=true
-      @option_parser.on( '--no-cleavage-semi', 'Dont allow up to 1 non tryptic terminus on peptides' ) do  
-        @options.cleavage_semi=false
-      end
+      add_boolean_option(:cleavage_semi,false,['--cleavage-semi', 'Search for peptides with up to 1 non-enzymatic cleavage site'])
     end
 
     if ( option_support.include? :respect_precursor_charges )
-      @options.respect_precursor_charges=false
-      @option_parser.on( '-q', '--respect-charges','Dont respect charges in the input file. Instead impute them by trying various options') do 
-        @options.respect_precursor_charges=true
-      end
+      add_boolean_option(:respect_precursor_charges,false,['-q', '--respect-charges','Dont respect charges in the input file. Instead impute them by trying various options'])
     end
 
     if ( option_support.include? :searched_ions )
-        @options.searched_ions = ""
-        @option_parser.on('--searched-ions si', 'Ion series to search (default=b,y)' ) do |si|
-          @options.searched_ions = si
-        end
+      add_value_option(:searched_ions,"",['--searched-ions si', 'Ion series to search (default=b,y)'])
+    end
+
+    if ( option_support.include? :multi_isotope_search )
+      add_boolean_option(:multi_isotope_search,false,["--multi-isotope-search","Expand parent mass window to include windows around neighbouring isotopic peaks"])
     end
 
     if ( option_support.include? :num_peaks_for_multi_isotope_search )
-        @options.num_peaks_for_multi_isotope_search="0"
-        @option_parser.on("--num-peaks-for-multi-isotope-search np","Number of peaks to include in multi-isotope search") do |np|
-          @options.num_peaks_for_multi_isotope_search=np
-        end
+      add_value_option(:num_peaks_for_multi_isotope_search,0,["--num-peaks-for-multi-isotope-search np","Number of peaks to include in multi-isotope search"])
     end
 
     if ( option_support.include? :glyco)
-      @options.glyco = false
-      @option_parser.on( '-g', '--glyco', 'Expect N-Glycosylation modifications as variable mod in a search or as a parameter when building statistical models' ) do 
-        @options.glyco = true
-      end
+      add_boolean_option(:glyco,false,['-g','--glyco', 'Expect N-Glycosylation modifications as variable mod in a search or as a parameter when building statistical models'])
+    end
+
+    if ( option_support.include? :acetyl_nterm)
+      add_boolean_option(:acetyl_nterm,false,['-y','--acetyl-nterm', 'Expect N-terminal acetylation as a variable mod in a search or as a parameter when building statistical models'])
     end
 
     if ( option_support.include? :methionine_oxidation)
-      @options.methionine_oxidation = false
-      @option_parser.on( '-m', '--methionineo', 'Expect Oxidised Methionine modifications as variable mod in a search' ) do 
-        @options.methionine_oxidation = true
-      end
+      add_boolean_option(:methionine_oxidation,false,['-m', '--methionineo', 'Expect Oxidised Methionine modifications as variable mod in a search'])
     end
 
     if ( option_support.include? :carbamidomethyl)
-      @options.carbamidomethyl = false
-      @option_parser.on( '-c', '--carbamidomethyl', 'Expect Carbamidomethyl C modifications as fixed mod in a search' ) do 
-        @options.carbamidomethyl = true
-      end
+      add_boolean_option(:carbamidomethyl,false,['-c', '--carbamidomethyl', 'Expect Carbamidomethyl C modifications as fixed mod in a search'])
     end
     
     if ( option_support.include? :maldi)
-      @options.maldi=false
-      @option_parser.on( '-l', '--maldi', 'Run a search on MALDI data') do
-        @options.maldi=true
-      end
+      add_boolean_option(:maldi,false,['-l', '--maldi', 'Run a search on MALDI data'])
     end
+
+    @option_parser.summary_width=40
+
       
   end
-  
-  
-  def jobid_from_filename(filename)
-    jobid="protk"
-    jobnum_match=filename.match(/(.{1,10}).*?\./)
-    if (jobnum_match!=nil)
-      jobid="#{self.jobid_prefix}#{jobnum_match[1]}"
-    end
-    return jobid
-  end
-    
-  # Based on the database setting and global database path, find the most current version of the required database
-  # This function returns the name of the database with an extension appropriate to the database type
-  #
-  def current_database(db_type,db=@options.database)
-    return Constants.new.current_database_for_name(db)
+      
+
+  def database_info
+    case
+      when Pathname.new(@options.database).exist? # It's an explicitly named db  
+        db_path=Pathname.new(@options.database).expand_path.to_s
+        db_name=Pathname.new(@options.database).basename.to_s
+      else
+        db_path=Constants.new.current_database_for_name @options.database
+        db_name=@options.database
+    end
+    FastaDatabase.new(db_name,db_path)
   end
-    
-end
+
+end
+
+
+

data/lib/protk/setup_rakefile.rake

@@ -143,9 +143,10 @@ tpp_download_file = download_task tpp_url, tpp_packagefile
 # Build
 file tpp_installed_file => [@build_dir,tpp_download_file] do
 	sh %{cp #{@download_dir}/#{tpp_packagefile} #{@build_dir}}
+#	use_perl_locallib_cmd="echo hi"
 	use_perl_locallib_cmd="eval $(perl -I#{perl_dir}/lib/perl5 -Mlocal::lib=#{perl_dir})"
-	sh %{#{use_perl_locallib_cmd};cpanm --local-lib=#{env.protk_dir}/perl5 XML::Parser}
-	sh %{#{use_perl_locallib_cmd};cpanm --local-lib=#{env.protk_dir}/perl5 CGI --force}
+#	sh %{#{use_perl_locallib_cmd};cpanm --local-lib=#{env.protk_dir}/perl5 XML::Parser}
+#	sh %{#{use_perl_locallib_cmd};cpanm --local-lib=#{env.protk_dir}/perl5 CGI --force}
 
 	sh %{cd #{@build_dir};tar -xvzf TPP-#{tpp_version}.tgz}
 
@@ -173,6 +174,8 @@ file tpp_installed_file => [@build_dir,tpp_download_file] do
 			f.write subs_text
 		end
 	end
+	sh %{cd #{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src;echo '' > ../perl/tpp_models.pl;echo '' > ../perl/exporTPP.pl;echo '' > ../CGI/show_nspbin.pl;echo '' > ../CGI/tpp_gui/tpp_gui.pl}
+
 	build_cmd = "#{use_perl_locallib_cmd};cd #{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src ; make -s"
 	install_cmd = "#{use_perl_locallib_cmd};cd #{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src ; make install"
 	env.log build_cmd, :info
@@ -354,9 +357,9 @@ file tandem_installed_file => [@build_dir,"#{@download_dir}/#{tandem_packagefile
     tandem_dirname = "#{tandem_packagefile.chomp('.zip')}"
 
     if ( tandem_platform=="linux") #Must compile
-    	tandem_src_dir = "#{@build_dir}/#{tandem_dirname}/#{tandem_dirname}/src/"
+    	tandem_src_dir = "#{@build_dir}/#{tandem_dirname}/src/"
     	sh %{cd #{tandem_src_dir}; make}
-    	sh %{cd #{@build_dir}; cp -r ./#{tandem_dirname}/#{tandem_dirname}/bin  #{env.tandem_root}/}
+    	sh %{cd #{@build_dir}; cp -r ./#{tandem_dirname}/bin  #{env.tandem_root}/}
     else
     	sh %{cd #{@build_dir}; cp -r ./#{tandem_packagefile.chomp('.zip')}/* #{env.tandem_root}/}
     	sh %{chmod u+x #{env.gpmtandem}}
@@ -402,8 +405,9 @@ task :galaxyenv => protk_galaxy_envfile
 
 # multitask :downloads => FileList["nr","env_nr","gi_taxid_prot.zip","taxdmp.zip"]
 
+task :base => [:perl_locallib]
 
-task :all => [:tpp,:omssa,:blast,:msgfplus,:pwiz,:openms,:galaxyenv]
+task :all => [:tpp,:omssa,:blast,:msgfplus,:pwiz,:openms]
 
 # Special task when installing via toolshed
 #

data/lib/protk/tandem_search_tool.rb

@@ -0,0 +1,256 @@
+require 'protk/search_tool'
+
+class String
+  def xtandem_modification_motif?
+		# 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.
+		ismotif=false
+		case self
+		when /[\(\)\{\}\!]/,/\[.*\]/
+			ismotif=true
+		end
+		ismotif
+  end
+end
+
+class TandemSearchTool < SearchTool
+	attr :defaults_path
+	attr :taxonomy_path
+	attr :default_data_path
+
+	attr :supported_xtandem_keys
+
+	def initialize
+
+		super([
+			:database,
+			:explicit_output,
+			:over_write,
+			:enzyme,
+			:modifications,
+			:mass_tolerance_units,
+			:mass_tolerance,
+			:multi_isotope_search,
+			:missed_cleavages,
+			:cleavage_semi,
+			:methionine_oxidation,
+			:glyco,
+			:acetyl_nterm,
+			:threads
+  			])
+
+		@xtandem_keys_with_single_multiplicity = {
+			:fragment_tol => "spectrum, fragment monoisotopic mass error",
+			:missed_cleavages => "scoring, maximum missed cleavage sites",
+			:cleavage_semi => "protein, cleavage semi",
+			:precursor_tolu => "spectrum, parent monoisotopic mass error units",
+			:multi_isotope_search => "spectrum, parent monoisotopic mass isotope error",
+			:fragment_tolu => "spectrum, fragment monoisotopic mass error units",
+			:acetyl_nterm => "protein, quick acetyl",
+			:output_spectra => "output, spectra",
+			:threads => "spectrum, threads"
+		}
+
+		@xtandem_keys_for_precursor_tol = {
+			:precursor_tol => ["spectrum, parent monoisotopic mass error minus", "spectrum, parent monoisotopic mass error plus"]
+		}
+
+		@defaults_path="#{File.dirname(__FILE__)}/data/tandem_params.xml"
+		@taxonomy_path="#{File.dirname(__FILE__)}/data/taxonomy_template.xml"
+		@default_data_path="#{File.dirname(__FILE__)}/data/"
+		
+		@option_parser.banner = "Run an X!Tandem msms search on a set of mzML input files.\n\nUsage: tandem_search.rb [options] file1.mzML file2.mzML ..."
+		@options.output_suffix="_tandem"
+
+		add_value_option(:tandem_params,"isb_native",['-T', '--tandem-params tandem', 'Either the full path to an xml file containing a complete set of default parameters, or one of the following (isb_native,isb_kscore,gpm). Default is isb_native'])
+		add_boolean_option(:keep_params_files,false,['-K', '--keep-params-files', 'Keep X!Tandem parameter files'])
+		add_boolean_option(:output_spectra,false,['--output-spectra', 'Include spectra in the output file'])
+
+	end
+
+	private
+	# Galaxy changes things like @ to __at__ we need to change it back
+	#
+	def decode_modification_string(mstring)
+  		mstring.gsub!("__at__","@")
+  		mstring.gsub!("__oc__","{")
+  		mstring.gsub!("__cc__","}")
+  		mstring.gsub!("__ob__","[")
+  		mstring.gsub!("__cb__","]")
+  		mstring
+	end
+
+	def set_option(std_params, tandem_key, value)
+  		notes = std_params.find("/bioml/note[@type=\"input\" and @label=\"#{tandem_key}\"]")
+  		throw "Exactly one parameter named (#{tandem_key}) is required in parameter file" unless notes.length==1
+  		notes[0].content=value.to_s
+	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 tandem_keys_in_params_file(default_params_path)
+		params_parser=XML::Parser.file(default_params_path)
+		default_params=params_parser.parse
+		input_nodes=default_params.find('/bioml/note[@type="input"]')
+		defined_keys=[]
+		input_nodes.each do |node|
+			defined_keys << node.attributes['label']
+		end
+		defined_keys
+	end
+
+	def taxon_from_taxonomy_file(taxo_path)
+		taxo_parser=XML::Parser.file(taxo_path)
+		taxo_doc=taxo_parser.parse
+		taxon_nodes=taxo_doc.find('/bioml/taxon')
+		throw "Exactly one taxon entry allowed in taxonomy file but found #{taxon_nodes.length}" unless taxon_nodes.length==1
+		taxon_nodes[0].attributes['label']
+	end
+
+	def generate_parameter_doc(std_params,output_path,input_path,db_info,taxo_path)
+
+		#
+		# The TandemSearchTool class has a special defaults system 
+		# Defaults are read from (a) The commandline (b) A defaults file (c) commandline defaults.
+		# The ideal priority order is a -> b -> c
+		#
+		# In order to support this we need to read the defaults file and check options defined there
+		# against those defined on the commandline
+		#
+		# In addition, we support some default parameter files built-in to protk. These are treated the same
+		# but are specified if the user provides a keyword rather than a path
+		#
+		default_params_notes=std_params.find('/bioml/note[@type="input" and @label="list path, default parameters"]')
+		throw "Exactly one list path, default parameters note is required in the parameter file" unless default_params_notes.length==1
+
+		is_file=File.exists?(self.tandem_params)
+		if is_file
+			default_params_notes[0].content="#{self.tandem_params}"			
+		else			
+			default_params_notes[0].content="#{@default_data_path}tandem_#{self.tandem_params}_defaults.xml"
+		end
+
+
+		keys_in_params_file=tandem_keys_in_params_file(default_params_notes[0].content)
+		keys_on_commandline=@options_defined_by_user.keys
+
+		# Set the input and output paths 
+		#
+		set_option(std_params,"spectrum, path",input_path)
+		set_option(std_params,"output, path",output_path)
+
+		# Taxonomy and Database
+		#  
+		set_option(std_params,"list path, taxonomy information",taxo_path)
+		set_option(std_params,"protein, taxon",db_info.name)
+
+
+
+		# set_option(std_params, "protein, cleavage semi", self.cleavage_semi ? "yes" : "no")
+
+		# Simple options (unique with a 1:1 mapping to parameters from this tool)
+		#
+		@xtandem_keys_with_single_multiplicity.each_pair do |commandline_option_key, xtandem_key|  
+			if (!keys_in_params_file.include?(xtandem_key) || keys_on_commandline.include?(commandline_option_key))
+				opt_val=self.send(commandline_option_key)
+				if opt_val.is_a?(TrueClass) || opt_val.is_a?(FalseClass)
+					opt_val = opt_val ? "yes" : "no"
+				end
+				append_option(std_params,xtandem_key,opt_val.to_s) 
+			end
+		end
+
+		# Precursor mass tolerance is a special case as it requires two xtandem options
+		#
+		@xtandem_keys_for_precursor_tol.each_pair do |commandline_option_key, xtandem_keys|  
+			xtandem_keys.each do |xtandem_key|
+				if (!keys_in_params_file.include?(xtandem_key) || keys_on_commandline.include?(commandline_option_key))
+					append_option(std_params,xtandem_key,(self.precursor_tol.to_f*0.5).to_s)
+				end
+			end
+		end
+
+		# Per residue Fixed and Variable Modifications
+		#
+		# These can be added using a variety of methods in xtandem
+		#
+		# residue, potential modification mass
+		# residue, modification mass
+		# residue, potential modification motif
+		#
+		# We support these primarily via the var_mods and fix_mods commandline params
+		# Modification masses and/or motifs can be entered via these arguments
+		#
+
+		var_mods = self.var_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject {|e| e.empty? }
+		var_mods=var_mods.collect {|mod| decode_modification_string(mod) }
+
+		# var_mods allows motif's as well as standard mods. These should be in a separate array
+		var_motifs = [].replace(var_mods)
+		var_mods.delete_if {|mod| mod.xtandem_modification_motif? }
+		var_motifs.keep_if {|mod| mod.xtandem_modification_motif? }
+
+		fix_mods = self.fix_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject { |e| e.empty? }
+		fix_mods=fix_mods.collect {|mod| decode_modification_string(mod)}
+
+		# We also support the --glyco and --methionineo shortcuts.
+		# Add these here. No check is made for duplication
+		#
+		var_motifs << "0.998@N!{P}[ST]" if self.glyco
+		var_mods << "15.994915@M" if self.methionine_oxidation
+
+		append_option(std_params,"residue, modification mass",fix_mods.join(",")) unless fix_mods.length==0
+		append_option(std_params,"residue, potential modification mass",var_mods.join(",")) unless var_mods.length==0
+		append_option(std_params,"residue, potential modification motif",var_motifs.join(",")) unless var_motifs.length==0
+
+		std_params
+  
+	end
+
+	public
+	def taxonomy_doc(db_info)
+		throw "Invalid input db_info must be a FastaDatabase object" unless db_info.class==FastaDatabase
+		database_path=db_info.path
+		taxon=db_info.name
+		# Parse taxonomy template file
+		#
+		taxo_parser=XML::Parser.file(@taxonomy_path)
+		taxo_doc=taxo_parser.parse
+
+		taxon_label=taxo_doc.find('/bioml/taxon')
+		throw "Exactly one taxon label is required in the taxonomy_template file" unless taxon_label.length==1
+		taxon_label[0].attributes['label']=taxon
+
+		db_file=taxo_doc.find('/bioml/taxon/file')
+		throw "Exactly one database file is required in the taxonomy_template file" unless db_file.length==1
+		db_file[0].attributes['URL']=database_path
+
+		taxo_doc
+	end
+
+
+	def params_doc(db_info,taxo_path,input_path,output_path)
+		params_parser=XML::Parser.file(@defaults_path)
+		std_params=params_parser.parse
+
+		
+		throw "Invalid input db_info must be a FastaDatabase object" unless db_info.class==FastaDatabase
+
+    	generate_parameter_doc(std_params,output_path,input_path,db_info,taxo_path)
+    end
+
+
+
+end