protk 1.3.1.pre3 → 1.4.0

data/lib/protk/manage_db_rakefile.rake

@@ -13,7 +13,7 @@ dbname=ARGV[0]
 
 # Load database spec file
 #
-$genv=Constants.new()
+$genv=Constants.instance()
 dbdir="#{$genv.protein_database_root}/#{dbname}"
 
 dbspec_file="#{dbdir}/.protkdb.yaml"

data/lib/protk/manage_db_tool.rb

@@ -12,7 +12,7 @@ require 'protk/tool'
 class ManageDBTool < Tool
 
   def add dbspec, dbname
-    genv=Constants.new()
+    genv=Constants.instance()
     dbdir="#{genv.protein_database_root}/#{dbname}"
     %x[mkdir -p #{dbdir}]
 

data/lib/protk/pepxml.rb

@@ -1,22 +1,174 @@
 require 'rubygems'
-require 'rexml/document'
-require 'rexml/xpath'
+require 'libxml'
+
+include LibXML
+
+# require 'rexml/document'
+# require 'rexml/xpath'
 
 class PepXML
+
+  attr_accessor :file_name
+
   def initialize(file_name)
-    @doc=REXML::Document.new(File.new(file_name))
+    @file_name=file_name
+
+    XML::Error.set_handler(&XML::Error::QUIET_HANDLER)
+    pepxml_parser=XML::Parser.file("#{file_name}")
+
+    @pepxml_ns_prefix="xmlns:"
+    @pepxml_ns="xmlns:http://regis-web.systemsbiology.net/pepXML"
+    @pepxml_doc=pepxml_parser.parse
+    if not @pepxml_doc.root.namespaces.default
+      @pepxml_ns_prefix=""
+      @pepxml_ns=nil
+    end
+  end
+
+
+  
+  # Obtain the database name from the given input file
+  #
+  def extract_db()
+    reader = XML::Reader.file(self.file_name)
+    throw "Failed to open xml file #{file_name}" unless reader!=nil
+
+    while(reader.read)
+      # For pep.xml files
+      #
+      if ( reader.name == "search_database" )
+        dbnode=reader.expand
+        dbvalue=dbnode['local_path']
+        reader.close
+        return dbvalue
+      end
+
+      # For prot.xml files
+      #
+      if ( reader.name == "protein_summary_header" )
+        dbnode=reader.expand
+        dbvalue=dbnode['reference_database']
+        reader.close
+        return dbvalue
+      end
+      
+      
+      
+    end
+
+  end
+
+
+
+  # Obtain the search engine name from the input file
+  # The name of the engine is returned in lowercase and should contain no spaces
+  # Names of common engines are searched for and extracted in simplified form if possible
+  #
+  def extract_engine()
+    reader = XML::Reader.file(self.file_name)
+    throw "Failed to open xml file #{file_name}" unless reader!=nil
+
+    while(reader.read)
+      if ( reader.name == "search_summary" )
+        dbnode=reader.expand
+        dbvalue=dbnode['search_engine']
+        reader.close
+        engine_name=dbvalue.gsub(/ /,"_")
+        engine_name=engine_name.gsub(/\(/,"")
+        engine_name=engine_name.gsub(/\)/,"")
+        engine_name=engine_name.gsub(/\!/,"")        
+        return engine_name.downcase
+      end
+    end
+  end
+
+
+  def extract_enzyme()
+    reader = XML::Reader.file(self.file_name)
+    throw "Failed to open xml file #{file_name}" unless reader!=nil
+
+    while(reader.read)
+      if ( reader.name == "sample_enzyme" )
+        dbnode=reader.expand
+        dbvalue=dbnode['name']
+        reader.close        
+        return dbvalue.downcase
+      end
+    end
+  end
+
+
+
+  def type_from_base_name(basename)
+    # A common error is for tools to include the extension in the base_name attribute.
+    # We exploit this to guess the type
+    ext_guess=""
+    case basename
+    when /.mgf$/
+      ext_guess="mgf"
+    when /.mzML$/
+      ext_guess="mzML"
+    when /.mzXML$/
+      ext_guess="mzXML"
+    else
+      ext_guess=""
+    end
+    ext_guess
+  end
+
+  def type_from_summary_attributes(atts)
+    if is_valid_type(atts["raw_data_type"])
+      return  atts["raw_data_type"]
+    end
+
+    if is_valid_type(atts["raw_data"])
+      return atts["raw_data"]
+    end
+    return ""
+  end
+
+  def is_valid_type(type)
+    case type
+    when /^mgf$/i
+      return true
+    when /^mzML$/i
+      return true
+    when /^mzXML$/i
+      return true
+    else
+      return false
+    end
   end
 
-  def find_runs() 
+
+  # TODO: Make this faster and more memory efficient by using XML::Reader as in the functions above
+  #
+  def find_runs()
+
+
+    run_summaries = @pepxml_doc.find("//#{@pepxml_ns_prefix}msms_run_summary", @pepxml_ns)
+
     runs = {}
-    REXML::XPath.each(@doc,"//msms_run_summary") do |summary|
+    run_summaries.each do |summary|
       base_name = summary.attributes["base_name"]
       if not runs.has_key?(base_name)
-        runs[base_name] = {:base_name => summary.attributes["base_name"],
-                           :type => summary.attributes["raw_data"]}
+        bn = summary.attributes["base_name"]
+
+        runs[base_name] = {:base_name => summary.attributes["base_name"]}
+
+        if is_valid_type(type_from_summary_attributes(summary.attributes))
+          runs[base_name][:type] = type_from_summary_attributes(summary.attributes)
+        elsif is_valid_type(type_from_base_name(bn))
+          runs[base_name][:type] = type_from_base_name(bn)
+        else
+          runs[base_name][:type] = "mzML" # Same guess as peptide prophet makes
+        end
+
       end
     end
     runs
   end
+
+
   
 end

data/lib/protk/plasmodb.rb

@@ -11,7 +11,7 @@ class PlasmoDB
     if ( env!=nil)
       @genv=env
     else
-      @genv=Constants.new
+      @genv=Constants.instance
     end
 
     database_file="#{@genv.protein_database_root}/#{@genv.plasmodb_annotation_database}/raw.txt"

data/lib/protk/prophet_tool.rb

@@ -21,63 +21,31 @@ class ProphetTool < SearchTool
     
     super(option_support)
 
-  end
-    
-
-
-  # Obtain the database name from the given input file
-  #
-  def extract_db(file_name)
-    reader = XML::Reader.file(file_name)
-    throw "Failed to open xml file #{file_name}" unless reader!=nil
-
-    while(reader.read)
-      # For pep.xml files
-      #
-      if ( reader.name == "search_database" )
-        dbnode=reader.expand
-        dbvalue=dbnode['local_path']
-        reader.close
-        return dbvalue
-      end
-
-      # For prot.xml files
-      #
-      if ( reader.name == "protein_summary_header" )
-        dbnode=reader.expand
-        dbvalue=dbnode['reference_database']
-        reader.close
-        return dbvalue
-      end
-      
-      
-      
+    if ( option_support.include? :probability_threshold )
+      add_value_option(:probability_threshold,0.05,['--p-thresh val', 'Probability threshold below which PSMs are discarded'])
     end
 
   end
-
-
-
-  # Obtain the search engine name from the input file
-  # The name of the engine is returned in lowercase and should contain no spaces
-  # Names of common engines are searched for and extracted in simplified form if possible
+   
+  # TODO: Deal with multiple enzyme combos
   #
-  def extract_engine(file_name)
-    reader = XML::Reader.file(file_name)
-    throw "Failed to open xml file #{file_name}" unless reader!=nil
+  def self.xinteract_code_for_enzyme(enzyme_name)
+
+  	codes = {
+  		'trypsin' => 'T',
+  		'stricttrypsin' => 'S',
+  		'chymotrypsin' => 'C',
+  		'ralphtrypsin' => 'R',
+  		'aspn' => 'A',
+  		'gluc' => 'G',
+  		'glucbicarb' => 'B',
+  		'cnbr' => 'M',
+  		'elastase' => 'E',
+  		'lysn' => 'L',
+  		'nonspecific' => 'N'
+  	}
+  	codes[enzyme_name]
 
-    while(reader.read)
-      if ( reader.name == "search_summary" )
-        dbnode=reader.expand
-        dbvalue=dbnode['search_engine']
-        reader.close
-        engine_name=dbvalue.gsub(/ /,"_")
-        engine_name=engine_name.gsub(/\(/,"")
-        engine_name=engine_name.gsub(/\)/,"")
-        engine_name=engine_name.gsub(/\!/,"")        
-        return engine_name.downcase
-      end
-    end
   end
 
 end

data/lib/protk/setup_rakefile.rake

@@ -1,8 +1,9 @@
 
 require 'protk/constants.rb'
+require 'rake/clean'
 require 'rbconfig'
 
-env=Constants.new
+env=Constants.instance
 
 @build_dir = "#{env.protk_dir}/tmp/build"
 @download_dir = "#{env.protk_dir}/tmp/download"
@@ -10,6 +11,8 @@ env=Constants.new
 directory @build_dir
 directory @download_dir
 
+CLEAN.include @build_dir, @download_dir
+
 def package_manager_name 
 	package_managers = ["brew","yum","apt-get"]
 
@@ -30,7 +33,7 @@ def clean_build_dir
 end
 
 def download_buildfile url, file
-	sh %{cd #{@download_dir}; wget  #{url}}
+	sh %{cd #{@download_dir}; wget -O #{file} #{url}}
 end
 
 def download_task url, packagefile
@@ -133,10 +136,10 @@ task :perl_locallib => [perl_locallib_installed_file]
 #
 # TPP
 #
-tpp_version="4.6.3"
+tpp_version="4.8.0"
 tpp_packagefile="TPP-#{tpp_version}.tgz"
 tpp_installed_file = "#{env.tpp_root}/bin/xinteract"
-tpp_url = "https://dl.dropbox.com/u/226794/TPP-4.6.3.tgz"
+tpp_url = "http://sourceforge.net/projects/sashimi/files/Trans-Proteomic%20Pipeline%20%28TPP%29/TPP%20v4.8%20%28philae%29%20rev%200/TPP_4.8.0-src.tgz/download"
 
 tpp_download_file = download_task tpp_url, tpp_packagefile
 
@@ -229,10 +232,11 @@ def blast_platform
 	'x64-linux'
 end
 
-blast_version="2.2.27+"
+blast_version="2.2.30+"
 blast_packagefile="ncbi-blast-#{blast_version}-#{blast_platform}.tar.gz"
 blast_url="ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/#{blast_version.chomp('+')}/#{blast_packagefile}"
 blast_installed_file="#{env.blast_root}/bin/makeblastdb"
+blast_required_bin=["makeblastdb", "blastdbcmd"]
 
 download_task blast_url, blast_packagefile
 
@@ -241,8 +245,10 @@ file blast_installed_file => [@build_dir,"#{@download_dir}/#{blast_packagefile}"
 	sh %{cp #{@download_dir}/#{blast_packagefile} #{@build_dir}}
     sh %{cd #{@build_dir}; gunzip #{blast_packagefile}}
     sh %{cd #{@build_dir}; tar -xvf #{blast_packagefile.chomp('.gz')}}
-    sh %{mkdir -p #{env.blast_root}}
-    sh %{cd #{@build_dir}; cp -r ncbi-blast-#{blast_version}/* #{env.blast_root}/}
+    sh %{mkdir -p #{env.blast_root}/bin}
+    blast_required_bin.each do |binary|
+        sh %{cd #{@build_dir}; cp -r ncbi-blast-#{blast_version}/bin/#{binary} #{env.blast_root}/bin/}
+    end
 end
 
 task :blast => blast_installed_file
@@ -251,7 +257,7 @@ task :blast => blast_installed_file
 #
 # MSGFPlus
 #
-msgfplus_version="20140210"
+msgfplus_version="20140630"
 msgfplus_packagefile="MSGFPlus.#{msgfplus_version}.zip"
 msgfplus_url="http://proteomics.ucsd.edu/Software/MSGFPlus/MSGFPlus.#{msgfplus_version}.zip"
 msgfplus_installed_file="#{env.msgfplus_root}/MSGFPlus.jar"
@@ -274,7 +280,7 @@ def pwiz_platform
 	if RbConfig::CONFIG['host_os'] =~ /darwin/ 
 		return 'darwin-x86-xgcc40'
 	end
-	'linux-x86_64-gcc42'
+	'linux-x86_64-gcc48'
 end
 
 def platform_bunzip
@@ -284,7 +290,7 @@ def platform_bunzip
 	'bunzip2'
 end
 
-pwiz_version="3_0_4388"
+pwiz_version="3_0_6790"
 pwiz_folder_name="pwiz-bin-#{pwiz_platform}-release-#{pwiz_version}"
 pwiz_packagefile="#{pwiz_folder_name}.tar.bz2"
 pwiz_url="https://dl.dropbox.com/u/226794/#{pwiz_packagefile}"
@@ -297,7 +303,8 @@ file pwiz_installed_file => [@build_dir,"#{@download_dir}/#{pwiz_packagefile}"]
     sh %{cd #{@build_dir}; #{platform_bunzip} -f #{pwiz_packagefile}}
     sh %{cd #{@build_dir}; tar -xvf #{pwiz_packagefile.chomp('.bz2')}}
     sh %{mkdir -p #{env.pwiz_root}}
-    sh %{cd #{@build_dir}; cp ./#{pwiz_folder_name}/* #{env.pwiz_root}/}
+    sh %{cd #{@build_dir}; cp ./msconvert #{env.pwiz_root}/}
+    sh %{cd #{@build_dir}; cp ./idconvert #{env.pwiz_root}/}    
 end
 
 task :pwiz => pwiz_installed_file

data/lib/protk/tandem_search_tool.rb

@@ -48,7 +48,8 @@ class TandemSearchTool < SearchTool
 			:fragment_tolu => "spectrum, fragment monoisotopic mass error units",
 			:acetyl_nterm => "protein, quick acetyl",
 			:output_spectra => "output, spectra",
-			:threads => "spectrum, threads"
+			:threads => "spectrum, threads",
+			:enzyme => "protein, cleavage site"
 		}
 
 		@xtandem_keys_for_precursor_tol = {
@@ -61,7 +62,7 @@ class TandemSearchTool < SearchTool
 		
 		@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"
-
+		@options.enzyme="[RK]|{P}"
 		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'])
@@ -71,12 +72,24 @@ class TandemSearchTool < SearchTool
 	private
 	# Galaxy changes things like @ to __at__ we need to change it back
 	#
-	def decode_modification_string(mstring)
+	def decode_galaxy_string(mstring)
   		mstring.gsub!("__at__","@")
   		mstring.gsub!("__oc__","{")
   		mstring.gsub!("__cc__","}")
   		mstring.gsub!("__ob__","[")
   		mstring.gsub!("__cb__","]")
+  		mstring.gsub!("__gt__",">")
+  		mstring.gsub!("__lt__","<")
+  		mstring.gsub!("__sq__","'")
+  		mstring.gsub!("__dq__","\"")
+  		mstring.gsub!("__cn__","\n")
+  		mstring.gsub!("__cr__","\r")
+  		mstring.gsub!("__tc__","\t")
+  		mstring.gsub!("__pd__","#")
+
+  		# For characters not allowed at all by galaxy
+  		mstring.gsub!("__pc__","|")
+
   		mstring
 	end
 
@@ -157,7 +170,6 @@ class TandemSearchTool < SearchTool
 		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)
@@ -168,7 +180,7 @@ class TandemSearchTool < SearchTool
 				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) 
+				append_option(std_params,xtandem_key,decode_galaxy_string(opt_val.to_s)) 
 			end
 		end
 
@@ -182,6 +194,7 @@ class TandemSearchTool < SearchTool
 			end
 		end
 
+
 		# Per residue Fixed and Variable Modifications
 		#
 		# These can be added using a variety of methods in xtandem
@@ -195,7 +208,7 @@ class TandemSearchTool < SearchTool
 		#
 
 		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=var_mods.collect {|mod| decode_galaxy_string(mod) }
 
 		# var_mods allows motif's as well as standard mods. These should be in a separate array
 		var_motifs = [].replace(var_mods)
@@ -203,7 +216,7 @@ class TandemSearchTool < SearchTool
 		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)}
+		fix_mods=fix_mods.collect {|mod| decode_galaxy_string(mod)}
 
 		# We also support the --glyco and --methionineo shortcuts.
 		# Add these here. No check is made for duplication