protk 1.1.0.pre

data/lib/protk/protxml.rb

@@ -0,0 +1,137 @@
+require 'rubygems'
+require 'rexml/document'
+require 'rexml/xpath'
+
+
+class ProtXML 
+    
+  attr_accessor :groups
+    
+    
+  def indistinguishable_proteins_from_protein(protein_element)
+    iprots=[]
+    REXML::XPath.each(protein_element,"./indistinguishable_protein") do |ipel|
+      ipel_attributes={}
+      ipel.attributes.each_attribute { |att| ipel_attributes[att.expanded_name.to_sym]=att.value }
+      iprots.push(ipel_attributes[:protein_name])
+    end    
+    iprots
+  end
+
+  def peptides_from_protein(protein_element)
+    peptides=[]
+    REXML::XPath.each(protein_element,"./peptide") do |pel|
+      peptide={}
+
+      pel.attributes.each_attribute { |att| peptide[att.expanded_name.to_sym]=att.value }
+      modifications=pel.get_elements("./modification_info")
+      mods=modifications.collect {|mp| mp.attribute("modified_peptide").value }
+      peptide[:modifications] = mods
+      peptides.push(peptide)
+    end
+    peptides
+  end
+    
+  def proteins_from_group(group_element)
+    proteins=[]
+    REXML::XPath.each(group_element,"./protein") do |pel|
+      protein={}
+      pel.attributes.each_attribute { |att| protein[att.expanded_name.to_sym]=att.value }
+      protein[:peptides]=peptides_from_protein(pel)      
+      protein[:indistinguishable_prots]=indistinguishable_proteins_from_protein(pel)
+      proteins.push(protein)
+    end
+    proteins
+  end
+    
+  def init_groups
+    @groups=[]
+    REXML::XPath.each(@doc.root,"//protein_group") do |gel|
+      group={}
+      group[:group_probability]=gel.attributes["probability"].to_f
+      group[:proteins]=proteins_from_group(gel)
+      groups.push group
+    end
+    @groups
+  end
+
+
+  def initialize(file_name)
+    @doc=REXML::Document.new(File.new(file_name))
+    @groups=self.init_groups
+  end
+  
+  def peptide_sequences_from_protein(prot)
+    peptides=prot[:peptides]
+    sequences=[]
+    peptides.each do |pep| 
+      if ( pep[:modifications].length > 0 )
+        pep[:modifications].each {|pmod| 
+          sequences.push(pmod) }
+      else
+        sequences.push(pep[:peptide_sequence])
+      end
+    end
+    sequences
+  end
+  
+  def protein_to_row(prot)
+    protein_row=[]
+    protein_row.push(prot[:protein_name])
+    protein_row.push(prot[:probability])
+    
+    indistinct=prot[:indistinguishable_prots]
+    indist_string="#{prot[:protein_name]};"
+    indistinct.each { |pr| indist_string<<"#{pr};"}
+    indist_string.chop!
+    protein_row.push(indist_string)
+    
+    protein_row.push(prot[:peptides].length)
+    
+    peptide_string=""
+    peptide_sequences_from_protein(prot).each {|pep| peptide_string<<"#{pep};" }
+    peptide_string.chop!
+    
+    protein_row.push(peptide_string)
+    protein_row
+  end
+  
+  # Convert the entire prot.xml document to row format
+  # Returns an array of arrays. Each of the sub-arrays is a row.
+  # Each row should contain a simple summary of the protein.
+  # A separate row should be provided for every protein (including indistinguishable ones)
+  # The first row will be the header
+  # 
+  # Proteins with probabilities below a threshold are excluded
+  #
+  def as_rows(threshold_probability)
+    
+    rows=[]
+    rows.push(["Accession","Probability","Indistinguishable Proteins","Num Peptides","Peptides"])
+    
+    proteins=[]
+    @groups.each do |grp|
+      grp[:proteins].each {|prot| 
+        if ( prot[:probability].to_f >= threshold_probability)
+          proteins.push(prot)
+        end
+      }
+    end
+    
+    proteins.each do |prot|
+      protein_row=protein_to_row(prot)
+      rows.push(protein_row)
+      
+      indistinguishables=prot[:indistinguishable_prots]
+      indistinguishables.each do |indist|
+        indist_row=protein_row.clone
+        indist_row[0]=indist
+        rows.push(indist_row)
+      end
+      
+    end
+    
+    rows
+  end
+  
+end

data/lib/protk/randomize.rb

@@ -0,0 +1,7 @@
+require 'protk/protk'
+
+class Randomize
+	def self.make_decoys input_path, db_len, output_path, prefix
+		Protk.make_decoys input_path.to_s, db_len.to_i,  output_path.to_s, prefix.to_s
+	end
+end

data/lib/protk/search_tool.rb

@@ -0,0 +1,182 @@
+#
+# This file is part of protk
+# Created by Ira Cooke 15/12/2010
+#
+# Provides common functionality used by all msms search tools.
+#
+# It allows;
+# 1. Specification of the search database using a simple name ... this class provides the necessary search for the actual file
+# 2. Output files to be specified via a prefix or suffix to be added to the name of the corresponding input file
+# 
+
+require 'optparse'
+require 'pathname'
+require 'protk/tool'
+
+class SearchTool < Tool
+
+  # Initializes commandline options common to all search tools.
+  # Individual search tools can add their own options, but should use Capital letters to avoid conflicts
+  #
+  def initialize(option_support={})
+    super(option_support)
+
+    if (option_support[:database]==true)          
+          
+      @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
+
+    end
+    
+    if ( option_support[:msms_search_detailed_options]==true)      
+       @options.allowed_charges="1+,2+,3+"
+       @option_parser.on(  '--allowed-charges ac', 'Allowed precursor ion charges. Default=1+,2+,3+' ) do |ac|
+         @options.allowed_charges = ac
+       end       
+       
+       @options.enzyme = "Trypsin"
+       @option_parser.on('--enzyme enz', 'Enzyme') do |enz|
+         @options.enzyme=enz
+       end
+       
+       @options.instrument = "ESI-QUAD-TOF"
+       @option_parser.on('--instrument instrument', 'Instrument') do |instrument|
+         @options.instrument=instrument
+       end
+       
+       
+       @options.var_mods = ""
+        @option_parser.on('--var-mods vm', 'Variable modifications (Overrides -g)' ) do |vm|
+          @options.var_mods = vm
+        end
+
+        @options.fix_mods = ""
+        @option_parser.on('--fix-mods fm', 'Fixed modifications (Overrides -c and -m options)' ) do |fm|
+          @options.fix_mods = fm
+        end
+
+        @options.searched_ions = ""
+        @option_parser.on('--searched-ions si', 'Ion series to search (default=b,y)' ) do |si|
+          @options.searched_ions = si
+        end
+
+        
+        @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
+        
+        @options.email=""
+        @option_parser.on('--email em', 'User email.') do |em|
+          @options.email = em
+        end
+
+        @options.username=""
+        @option_parser.on('--username un', 'Username.') do |un|
+          @options.username = un
+        end
+
+        @options.mascot_server="www.matrixscience.com"
+        @option_parser.on( '-S', '--server url', 'The url to the cgi directory of the mascot server' ) do |url| 
+          @options.mascot_server=url
+        end
+
+        @options.mascot_server=""
+        @option_parser.on('--username un', 'Username.') do |un|
+          @options.username = un
+        end
+
+        @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
+
+       
+     end
+    
+    if ( option_support[:msms_search]==true)      
+      @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
+      
+      @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
+      
+      @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
+      
+      @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
+      
+      @options.missed_cleavages=2
+      @option_parser.on( '-v', '--num-missed-cleavages num', 'Number of missed cleavages allowed' ) do |num| 
+        @options.missed_cleavages = num
+      end
+    
+      @options.carbamidomethyl=true
+      @option_parser.on( '-c', '--no-carbamidomethyl', 'Run a search without a carbamidomethyl fixed modification' ) do 
+        @options.carbamidomethyl = false
+      end
+
+      @options.methionine_oxidation=false
+      @option_parser.on( '-m', '--methionine-oxidation', 'Run a search with oxidised methionines as a variable modification' ) do 
+        @options.methionine_oxidation = true
+      end
+      
+    end
+      
+    if ( option_support[:glyco]==true)
+
+      @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
+
+    end
+    
+    if ( option_support[:maldi]==true)
+      @options.maldi=false
+      @option_parser.on( '-l', '--maldi', 'Run a search on MALDI data') do
+        @options.maldi=true
+      end
+    end
+      
+  end
+  
+  
+  def jobid_from_filename(filename)
+    jobid="protk"
+    jobnum_match=filename.match(/(.{1,10})\.d/)
+    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)
+  end
+    
+end

data/lib/protk/setup_rakefile.rake

@@ -0,0 +1,245 @@
+
+require 'protk/constants.rb'
+require 'rbconfig'
+
+env=Constants.new
+
+@build_dir = "#{env.protk_dir}/tmp/build"
+@download_dir = "#{env.protk_dir}/tmp/download"
+
+directory @build_dir
+directory @download_dir
+
+def package_manager_name 
+	if RbConfig::CONFIG['host_os'] =~ /darwin/ 
+		return 'brew'
+	end
+	'apt-get'
+end
+
+def clean_build_dir
+	sh %{cd #{@build_dir}; rm -rf ./*}
+end
+
+def download_buildfile url, file
+	sh %{cd #{@download_dir}; wget  #{url}}
+end
+
+def download_task url, packagefile
+	file "#{@download_dir}/#{packagefile}" => @download_dir do 
+		download_buildfile "#{url}", "#{packagefile}"
+	end
+end
+
+#
+# Package manager
+#
+task :package_manager do
+	pmname = package_manager_name
+	needs_homebrew=false
+	sh "which #{pmname}" do |ok,res|
+		unless ok
+			throw "Missing package manager #{pmname}" unless  pmname=='brew'
+			needs_homebrew=true
+		end
+	end
+
+	if needs_homebrew
+		sh { "ruby -e \"$(curl -fsSkL raw.github.com/mxcl/homebrew/go)" }
+		sh { "brew update"}
+		sh { "brew tap homebrew/versions"}
+	end
+
+end
+
+#
+# System packages
+#
+task :system_packages => :package_manager do
+	# Gather package requirements 
+	pkgs=YAML::load(File.open("#{File.dirname(__FILE__)}/data/#{package_manager_name}_packages.yaml"))
+
+	unique_packages=[]
+	apps=[]
+	installed_packages=[]
+	for pk in pkgs 
+    	unique_packages = pk[1] | unique_packages
+    	apps = apps.push pk[0]
+  	end
+
+  	# Install all packages
+  	#
+  	unique_packages.each { |pk|  
+  		sh "#{package_manager_name} install #{pk}" do |ok,res|
+  			p res
+  			installed_packages.push pk if ok
+  		end
+  	}
+
+end
+
+#
+# Perl local::lib
+#
+perl_locallib_version="1.008004"
+perl_locallib_packagefile="local-lib-#{perl_locallib_version}.tar.gz"
+perl_locallib_installed_file = "#{env.protk_dir}/perl5/lib/perl5/local/lib.pm"
+perl_locallib_url = "http://search.cpan.org/CPAN/authors/id/A/AP/APEIRON/local-lib-#{perl_locallib_version}.tar.gz"
+
+download_task perl_locallib_url, perl_locallib_packagefile
+
+file perl_locallib_installed_file =>  [@build_dir,"#{@download_dir}/#{perl_locallib_packagefile}"] do
+	sh %{cp #{@download_dir}/#{perl_locallib_packagefile} #{@build_dir}}
+	perl_dir = "#{env.protk_dir}/perl5"
+
+	sh %{cd #{@build_dir}; gunzip local-lib-#{perl_locallib_version}.tar.gz } 
+	sh %{cd #{@build_dir}; tar -xf local-lib-#{perl_locallib_version}.tar } 
+	sh "cd #{@build_dir}/local-lib-#{perl_locallib_version}; perl Makefile.PL --bootstrap=#{perl_dir}; make install" do |ok,res|
+#		clean_build_dir if ok
+	end
+
+	if !Pathname.new("~/.bashrc").exist? || File.read("~/.bashrc") =~ /Mlocal::lib/
+		sh "echo 'eval $(perl -I#{perl_dir}/lib/perl5 -Mlocal::lib=#{perl_dir})' >>~/.bashrc"
+	end
+
+	if !Pathname.new("~/.bash_profile").exist? || File.read("~/.bash_profile") =~ /Mlocal::lib/
+		sh "echo 'eval $(perl -I#{perl_dir}/lib/perl5 -Mlocal::lib=#{perl_dir})' >>~/.bash_profile"
+	end
+end
+
+task :perl_locallib => [perl_locallib_installed_file]
+
+
+#
+# Top Level Packages. 
+#
+
+
+#
+# TPP
+#
+tpp_version="4.6.1"
+tpp_packagefile="TPP-#{tpp_version}.tgz"
+tpp_installed_file = "#{env.xinteract}"
+tpp_url = "https://dl.dropbox.com/u/226794/TPP-4.6.1.tgz"
+
+download_task tpp_url, tpp_packagefile
+
+# Build
+file tpp_installed_file => [:perl_locallib,@build_dir,"#{@download_dir}/#{tpp_packagefile}"] do
+	sh %{cp #{@download_dir}/#{tpp_packagefile} #{@build_dir}}
+	sh %{cpanm --local-lib=#{env.protk_dir}/perl5 XML::Parser}
+	sh %{cpanm --local-lib=#{env.protk_dir}/perl5 XML::CGI --force}
+
+	sh %{cd #{@build_dir};tar -xvzf TPP-#{tpp_version}.tgz}
+
+	File.open("#{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src/Makefile.config.incl","wb") do |f|
+		f.write "TPP_ROOT=#{env.tpp_root}/\nTPP_WEB=/tpp/\nXSLT_PROC=/usr/bin/xsltproc\nCGI_USERS_DIR=${TPP_ROOT}cgi-bin/"
+	end
+
+	makefile_incl_path="#{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src/Makefile.incl"
+	makefile_incl_text=File.read("#{makefile_incl_path}")
+
+	# Homebrew specific modifications to makefiles
+	#
+	if ( package_manager_name=='brew')
+		File.open("#{makefile_incl_path}","w+") do |f|
+			subs_text = makefile_incl_text.gsub(/GD_LIB= \/opt\/local\/lib\/libgd.a \/opt\/local\/lib\/libpng.a/,"GD_LIB= /usr/local/lib/libgd.a /usr/local/opt/libpng12/lib/libpng.a") #We're using homebrew not fink or macports
+			subs_text = subs_text.gsub(/GD_INCL= -I \/opt\/local\/include\//,"GD_INCL= -I /usr/local/include/ -I /usr/local/opt/libpng12/include")
+			f.write subs_text
+		end	
+
+		makefile_path="#{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/CGI/Makefile"
+		makefile_text = File.read("#{makefile_path}")
+
+		File.open("#{makefile_path}","w+") do |f|
+			subs_text = makefile_text.gsub("cp -rfu","cp -rf")
+			f.write subs_text
+		end
+	end
+	sh %{cd #{@build_dir}/TPP-#{tpp_version}/trans_proteomic_pipeline/src ; make; make install}
+
+end
+
+task :tpp => tpp_installed_file
+
+
+
+#
+# omssa
+#
+def omssa_platform 
+	if RbConfig::CONFIG['host_os'] =~ /darwin/ 
+		return 'macos'
+	end
+	'linux'
+end
+
+omssa_packagefile="omssa-#{omssa_platform}.tar.gz"
+omssa_installed_file = "#{env.omssacl}"
+omssa_url = "ftp://ftp.ncbi.nih.gov/pub/lewisg/omssa/CURRENT/omssa-#{omssa_platform}.tar.gz"
+
+download_task omssa_url, omssa_packagefile
+
+# Install
+file omssa_installed_file => [@build_dir,"#{@download_dir}/omssa-#{omssa_platform}.tar.gz"] do
+	sh %{cp #{@download_dir}/#{omssa_packagefile} #{@build_dir}}
+    sh %{cd #{@build_dir}; gunzip omssa-#{omssa_platform}.tar.gz}
+    sh %{cd #{@build_dir}; tar -xvf omssa-#{omssa_platform}.tar}
+    sh %{mkdir -p #{env.omssa_root}}
+    sh %{cd #{@build_dir}; cp -r omssa-*.#{omssa_platform}/* #{env.omssa_root}/}
+end
+
+task :omssa => omssa_installed_file
+
+
+
+#
+# blast
+#
+def blast_platform 
+	if RbConfig::CONFIG['host_os'] =~ /darwin/ 
+		return 'universal-macosx'
+	end
+	'x64-linux'
+end
+
+blast_version="2.2.27+"
+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.makeblastdb}"
+
+download_task blast_url, blast_packagefile
+
+# Install
+file blast_installed_file => [@build_dir,"#{@download_dir}/#{blast_packagefile}"] do
+	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}/}
+end
+
+task :blast => blast_installed_file
+
+
+#
+# MSGFPlus
+#
+msgfplus_version="20120823"
+msgfplus_packagefile="MSGFPlus.20120823.zip"
+msgfplus_url="http://proteomics.ucsd.edu/Downloads/MSGFPlus.20120823.zip"
+msgfplus_installed_file="#{env.msgfplusjar}"
+
+download_task msgfplus_url, msgfplus_packagefile
+
+file msgfplus_installed_file => [@build_dir,"#{@download_dir}/#{msgfplus_packagefile}"] do
+	sh %{cp #{@download_dir}/#{msgfplus_packagefile} #{@build_dir}}
+    sh %{cd #{@build_dir}; unzip #{msgfplus_packagefile}}
+    sh %{mkdir -p #{env.msgfplus_root}}
+    sh %{cd #{@build_dir}; cp MSGFPlus.jar #{env.msgfplus_root}/}
+end
+
+task :msgfplus => msgfplus_installed_file
+
+task :all => [:tpp,:omssa,:blast]

data/lib/protk/setup_tool.rb

@@ -0,0 +1,19 @@
+require 'optparse'
+require 'pathname'
+require 'protk/tool'
+
+require 'rake'
+require 'pp'
+
+include Rake::DSL
+
+Rake.application.init 'protk_setup'
+Rake.application.rake_require 'protk/setup_rakefile'
+
+class SetupTool < Tool
+
+	def install toolname
+		Rake.application.invoke_task toolname
+	end
+
+end

data/lib/protk/spreadsheet_extensions.rb

@@ -0,0 +1,78 @@
+# Add a method to the Spreadsheet::Worksheet class to insert a column
+class Spreadsheet::Worksheet < Object
+    def insert_column(col,index)
+      # First check to see if the length of the column equals the number of rows
+      if ( col.length!=self.rows.length && self.rows.length!=0)
+        raise "The length of column #{col.length} does not equal the number of rows #{self.rows.length}"
+      end
+      if ( col.class!=Array || index.class!=Fixnum)
+        raise "Wrong arguments. Requires a column array and an integer index"
+      end
+      
+      # Check for special case where there are no rows yet and if so then insert as new rows
+      if ( self.rows.length==0)
+        col.each_index { |i|
+          self.insert_row(i,[col[i]])        
+        }
+      else
+        # Insert the column row by row. Probably inefficient but it works
+        rowi=0
+        self.each {|row|
+          row.insert(index,col[rowi])        
+          rowi+=1
+        }
+      end
+    end
+  end
+  
+  class Spreadsheet::Workbook < Object
+    
+    
+    # creates an output excel file (returning the workbook object), transcribing all original content up to the given number of rows
+    # Throws an error if the input contains more than 1 worksheet
+    #
+    def copyBook(numrows=0)
+     
+     if ( !numrows )
+       numrows=0
+     end
+     
+      # Create a new workbook from scratch for writing
+      outputBook = Spreadsheet::Workbook.new
+      outputSheet = outputBook.create_worksheet
+
+      # There should only be one worksheet in the input workbook
+      worksheets=self.worksheets
+      if ( self.worksheets.length != 1 )
+        puts "More than one worksheet in this excel file. This script only operates on single worksheets"
+      end
+
+      # Get the worksheet  
+      inputSheet=self.worksheet 0
+
+      # Figure out how many rows to convert if not specified
+      if ( numrows==0 || numrows > (inputSheet.row_count+1))
+        numrows=inputSheet.row_count
+      end
+
+
+      # Transcribe everything from the old worksheet to the new one
+      puts "Creating new spreadsheet with #{numrows} rows"
+      (0...[numrows,inputSheet.row_count].min).each { |r| 
+
+        outputSheet.insert_row(r,inputSheet.row(r))
+
+        newRow=outputSheet.row(r)
+
+        # After inserting the row make sure it doesn't contain any nil values
+        newRow.each_index { |ci| 
+          if ( newRow[ci]==nil)
+            newRow[ci]=""
+          end
+        }      
+      }
+      outputBook
+    end
+    
+    
+end