protk 1.1.0.pre

data/bin/manage_db.rb

@@ -0,0 +1,123 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 13/3/2012
+#
+# Manage named protein databases
+#
+#
+
+require 'protk/constants'
+require 'protk/manage_db_tool'
+require 'yaml'
+require 'pp'
+
+
+
+
+command=ARGV[0]
+ARGV[0] = "--help" if ( command==nil || command=="-h" || command=="help")
+tool=ManageDBTool.new(command)
+if ( tool.option_parser.banner=="")
+  tool.option_parser.banner = "Manage named protein databases.\nUsage: manage_db.rb <command> [options] dbname\nCommands are: add remove update list help\nType manage_db <command> -h to get help on a specific command"
+end
+
+tool.option_parser.parse!
+if ( ARGV[0]=="--help")
+  exit
+end
+
+command=ARGV.shift
+dbname=ARGV.shift
+
+if ( dbname!=nil)
+  dbname=dbname.downcase
+  throw "Database name should contain no spaces" if ( dbname=~/\s/)
+end
+
+genv=Constants.new()
+
+
+case command
+when "add"
+  throw "Must specify a database name" if dbname==nil 
+  throw "all is a reserved word and cannot be used as a database name" if ( dbname=="all")
+  throw "Database #{dbname} exists. Use --update-spec to overwrite." if genv.dbexist?(dbname) && !tool.update_spec
+  throw "Database #{dbname} cannot be updated because it doesn't exist" if !genv.dbexist?(dbname) && tool.update_spec
+
+  dbspec = tool.get_predefined_definition dbname
+  throw "#{dbname} is not a predefined database"  if tool.predefined && dbspec=={}
+
+  genv.log("Adding new database #{dbname}",:info) if !tool.update_spec
+  genv.log("Updating spec for #{dbname}",:info) if tool.update_spec
+
+  if dbspec=={} 
+     # Create the database specifiation dictionary (or make ammendments if a predefinition was used)
+     dbspec[:is_annotation_db]=tool.is_annotation_db
+     dbspec[:sources]=tool.sources
+     dbspec[:make_blast_index]= tool.make_blast_index
+     dbspec[:make_msgf_index]= tool.make_msgf_index
+     dbspec[:include_filters]= tool.is_annotation_db ? [] : tool.include_filters
+     dbspec[:id_regexes]= tool.is_annotation_db ? [] : tool.id_regexes
+     dbspec[:decoys]= tool.is_annotation_db ? false : tool.decoys
+     dbspec[:archive_old]= tool.is_annotation_db ? false : tool.archive_old
+     dbspec[:decoy_prefix]= tool.decoy_prefix
+     dbspec[:format] = tool.db_format
+  end  
+  tool.add dbspec, dbname
+
+when "update"
+  throw "Must specify a database name" if dbname==nil
+  if ( dbname=="all" )
+    dbnames=tool.all_database_names(genv)
+  else
+    dbnames=[dbname]
+  end
+  p dbnames
+  dbnames.each { |db|  
+    throw "Database #{db} does not exist" if !genv.dbexist?(db) 
+    dbdir="#{genv.protein_database_root}/#{db}"
+    throw "Could not find required spec file #{dbdir}/.protkdb.yaml" unless Pathname.new("#{dbdir}/.protkdb.yaml").exist?
+   runner=CommandRunner.new(genv)
+    runner.run_local("rake -f #{tool.rakefile_path} #{db} ")
+  } 
+when "list"
+  
+  gw_file_handle=nil
+  if tool.galaxy_write
+    pepxml_loc = "#{genv.galaxy_root}/tool-data/pepxml_databases.loc"
+    if ( Pathname.new(pepxml_loc).exist?  )
+      gw_file_handle=File.open(pepxml_loc,'w')
+    end
+    p "Warning: Could not find database loc file #{pepxml_loc}" unless Pathname.new(pepxml_loc).exist?
+  end
+  
+  
+  Dir.foreach(genv.protein_database_root) do |db_subdir|
+    db_specfile="#{genv.protein_database_root}/#{db_subdir}/.protkdb.yaml"
+    if ( Pathname.new(db_specfile).exist?)
+      spec=YAML.load_file(db_specfile)
+      case
+      when tool.galaxy || tool.galaxy_write
+        unless ( spec[:is_annotation_db] )
+          db_prettyname=db_subdir.gsub(/_/,' ').capitalize
+          loc_line= "#{db_prettyname}\t#{db_subdir}_\t#{db_subdir}\t#{db_subdir}_\n"
+          puts loc_line
+          if ( gw_file_handle )
+            gw_file_handle.write loc_line
+          end
+        end
+      when tool.verbose
+        puts "-- #{db_subdir} --\n"
+        PP.pp(spec)
+        puts "\n"
+      else
+        puts "#{db_subdir}\n"
+      end
+    end
+  end
+
+  gw_file_handle.close if ( gw_file_handle)
+
+end
+

data/bin/mascot_search.rb

@@ -0,0 +1,187 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 14/12/2010
+#
+# Runs an MS/MS search using the Mascot search engine
+#
+
+$VERBOSE=nil
+
+require 'protk/constants'
+require 'protk/command_runner'
+require 'protk/search_tool'
+require 'rest_client'
+
+# Environment with global constants
+#
+genv=Constants.new
+
+# Setup specific command-line options for this tool. Other options are inherited from SearchTool
+#
+search_tool=SearchTool.new({:msms_search=>true,:background=>false,:database=>true,:explicit_output=>true,:over_write=>true,:msms_search_detailed_options=>true})
+search_tool.jobid_prefix="o"
+
+search_tool.option_parser.banner = "Run a Mascot msms search on a set of mgf input files.\n\nUsage: mascot_search.rb [options] file1.mgf file2.mgf ..."
+search_tool.options.output_suffix="_mascot"
+
+search_tool.options.mascot_server="#{genv.default_mascot_server}/mascot/cgi/"
+#search_tool.option_parser.on( '-P', '--http-proxy url', 'The url to a proxy server' ) do |url| 
+#  search_tool.options.mascot_server=url
+#end
+
+#search_tool.options.http_proxy="http://squid.latrobe.edu.au:8080"
+#search_tool.option_parser.on( '-P', '--http-proxy url', 'The url to a proxy server' ) do |url| 
+#  search_tool.options.http_proxy=url
+#end
+
+search_tool.option_parser.parse!
+
+
+
+
+
+# Set search engine specific parameters on the SearchTool object
+#
+fragment_tol = search_tool.fragment_tol
+precursor_tol = search_tool.precursor_tol
+
+
+
+mascot_cgi=search_tool.mascot_server
+
+unless ( mascot_cgi =~ /^http:\/\//)
+  mascot_cgi  = "http://#{mascot_cgi}"
+end
+
+#
+RestClient.proxy=search_tool.http_proxy
+
+genv.log("Var mods #{search_tool.var_mods} and fixed #{search_tool.fix_mods}",:info)
+
+var_mods = search_tool.var_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject {|e| e.empty? }.join(",")
+fix_mods = search_tool.fix_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject { |e| e.empty? }.join(",")
+
+# None is given by a an empty galaxy multi-select list and we need to turn it into an empty string
+#
+var_mods=""  if var_mods=="None"
+fix_mods="" if fix_mods=="None"
+
+postdict={}
+
+# CHARGE
+#
+postdict[:CHARGE]=search_tool.allowed_charges
+
+# CLE
+#
+postdict[:CLE]=search_tool.enzyme
+
+# PFA
+#
+postdict[:PFA]=search_tool.missed_cleavages
+
+# COM (Search title)
+# 
+postdict[:COM]="Protk"
+
+# DB (Database)
+#
+postdict[:DB]=search_tool.database
+
+# INSTRUMENT
+#
+postdict[:INSTRUMENT]=search_tool.instrument
+
+# IT_MODS (Variable Modifications)
+#
+postdict[:IT_MODS]=var_mods
+
+# ITOL (Fragment ion tolerance)
+#
+postdict[:ITOL]=search_tool.fragment_tol
+
+# ITOLU (Fragment ion tolerance units)
+#
+postdict[:ITOLU]=search_tool.fragment_tolu
+
+# MASS (Monoisotopic and Average)
+#
+postdict[:MASS]=search_tool.precursor_search_type
+
+# MODS (Fixed modifications)
+#
+postdict[:MODS]=fix_mods
+
+# REPORT (What to include in the search report. For command-line searches this is pretty much irrelevant because we retrieve the entire results file anyway)
+#
+postdict[:REPORT]="AUTO"
+
+# TAXONOMY (Blank because we don't allow taxonomy)
+#
+postdict[:TAXONOMY]="All entries"
+
+# TOL (Precursor ion tolerance (Unit dependent))
+#
+postdict[:TOL]=search_tool.precursor_tol
+
+# TOLU (Tolerance Units)
+#
+postdict[:TOLU]=search_tool.precursor_tolu
+
+# Email
+#
+postdict[:USEREMAIL]=search_tool.email
+
+# Username
+#
+postdict[:USERNAME]=search_tool.username
+
+
+# FILE
+#
+postdict[:FILE]=File.new(ARGV[0])
+
+postdict[:FORMVER]='1.01'
+postdict[:INTERMEDIATE]=''
+
+genv.log("Sending #{postdict}",:info)
+
+postdict.each do |kv| 
+  p "#{kv}|\n"
+end
+
+search_response=RestClient.post "#{mascot_cgi}/nph-mascot.exe?1",  postdict
+
+genv.log("Mascot search response was #{search_response}",:info)
+
+# Look for an error if there is one
+error_result= /Sorry, your search could not be performed(.*)/.match(search_response)
+if ( error_result != nil )
+  p error_result[0]
+  genv.log("Mascot search failed with response #{search_response}",:warn)
+  throw "Mascot search failed with response #{search_response}"
+else
+
+  # Search for the location of the mascot data file in the response
+  results=/master_results_?2?\.pl\?file=\.*\/data\/(.*)\/(.+\.dat)/.match(search_response)
+  results_date=results[1]
+  results_file=results[2]
+
+
+  get_url= "#{mascot_cgi}/../x-cgi/ms-status.exe?Autorefresh=false&Show=RESULTFILE&DateDir=#{results_date}&ResJob=#{results_file}" 
+
+  if ( search_tool.explicit_output!=nil)
+    output_path=search_tool.explicit_output
+  else
+    output_path="#{results_file}"
+  end
+
+  # Download the results
+  #
+  require 'open-uri'
+  open("#{output_path}", 'wb') do |file|
+    file << open("#{get_url}").read
+  end
+end
+

data/bin/mascot_to_pepxml.rb

@@ -0,0 +1,44 @@
+#!/usr/bin/env ruby
+#
+# This file is part of MSLIMS
+# Created by Ira Cooke 12/4/2010
+#
+# Convert mascot dat files to pepxml. A wrapper for Mascot2XML
+#
+
+
+require 'protk/constants'
+require 'protk/search_tool'
+require 'protk/mascot_util'
+
+# Environment with global constants
+#
+genv=Constants.new
+
+tool=SearchTool.new({:database=>true,:explicit_output=>true,:over_write=>true})
+tool.option_parser.banner = "Convert mascot dat files to pep.xml files.\n\nUsage: mascot_to_pepxml.rb [options] file1.dat file2.dat ... "
+tool.option_parser.parse!
+
+
+ARGV.each do |file_name| 
+  name=file_name.chomp
+
+  this_dir=Pathname.new(name).dirname.realpath
+
+  if ( tool.explicit_output==nil )
+    new_basename="#{this_dir}/#{MascotUtil.input_basename(name)}_mascot2xml"      
+    cmd="cp #{name} #{new_basename}.dat"
+    cmd << "; #{genv.mascot2xml} #{new_basename}.dat -D#{tool.current_database :fasta}"
+    
+  else  #Mascot2XML doesn't support explicitly named output files so we move the file to an appropriate output filename after finishing
+    new_basename="#{this_dir}/#{MascotUtil.input_basename(name)}_mascot2xml"
+    cmd="cp #{name} #{new_basename}.dat"
+    cmd << "; #{genv.mascot2xml} #{new_basename}.dat -D#{tool.current_database :fasta}"
+    cmd << "; mv #{new_basename}.pep.xml #{tool.explicit_output}; rm #{new_basename}.dat"
+    repair_script="#{File.dirname(__FILE__)}/repair_run_summary.rb"     
+    cmd << "; #{repair_script} #{tool.explicit_output}"
+  end
+    
+  code = tool.run(cmd,genv,nil,nil)
+  throw "Command #{cmd} failed with exit code #{code}" unless code==0
+end

data/bin/msgfplus_search.rb

@@ -0,0 +1,191 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 14/12/2010
+#
+# Runs an MS/MS search using the MSGFPlus search engine
+#
+$VERBOSE=nil
+require 'protk/constants'
+require 'protk/command_runner'
+require 'protk/search_tool'
+
+
+# Setup specific command-line options for this tool. Other options are inherited from SearchTool
+#
+search_tool=SearchTool.new({:msms_search=>true,:background=>false,:glyco=>true,:database=>true,:explicit_output=>true,:over_write=>true,:msms_search_detailed_options=>true})
+search_tool.option_parser.banner = "Run an MSGFPlus msms search on a set of msms spectrum input files.\n\nUsage: msgfplus_search.rb [options] file1.mzML file2.mzML ..."
+search_tool.options.output_suffix="_msgfplus"
+
+search_tool.options.fragment_method=0
+search_tool.option_parser.on(  '--fragment-method method', 'Fragment method 0: As written in the spectrum or CID if no info (Default), 1: CID, 2: ETD, 3: HCD, 4: Merge spectra from the same precursor' ) do |method|
+  search_tool.options.fragment_method=method
+end
+
+search_tool.options.protocol=0
+search_tool.option_parser.on(  '--protocol p', '0: NoProtocol (Default), 1: Phosphorylation' ) do |p|
+  search_tool.options.protocol=p
+end
+
+search_tool.options.min_pep_length=6
+search_tool.option_parser.on(  '--min-pep-length p', 'Minimum peptide length to consider, Default: 6' ) do |p|
+  search_tool.options.min_pep_length=p
+end
+
+search_tool.options.max_pep_length=40
+search_tool.option_parser.on(  '--max-pep-length p', 'Maximum peptide length to consider, Default: 40' ) do |p|
+  search_tool.options.max_pep_length=p
+end
+
+search_tool.options.min_pep_charge=2
+search_tool.option_parser.on(  '--min-pep-charge c', 'Minimum precursor charge to consider if charges are not specified in the spectrum file, Default: 2' ) do |c|
+  search_tool.options.min_pep_charge=c
+end
+
+search_tool.options.max_pep_charge=3
+search_tool.option_parser.on(  '--max-pep-charge c', 'Maximum precursor charge to consider if charges are not specified in the spectrum file, Default: 3' ) do |c|
+  search_tool.options.max_pep_charge=c
+end
+
+search_tool.options.num_reported_matches=1
+search_tool.option_parser.on(  '--num-reported-matches n', 'Number of matches per spectrum to be reported, Default: 1' ) do |n|
+  search_tool.options.num_reported_matches=n
+end
+
+search_tool.options.add_features=false
+search_tool.option_parser.on(  '--add-features', 'output additional features' ) do 
+  search_tool.options.add_features=true
+end
+
+search_tool.options.java_mem="3500M"
+search_tool.option_parser.on('--java-mem mem','Java memory limit when running the search (Default 3.5Gb)') do |mem|
+  search_tool.options.java_mem=mem
+end
+  
+search_tool.option_parser.parse!
+
+# Environment with global constants
+#
+genv=Constants.new
+
+# Set search engine specific parameters on the SearchTool object
+#
+msgf_bin="#{genv.msgfplusjar}"
+
+case 
+when Pathname.new(search_tool.database).exist? # It's an explicitly named db
+  current_db=Pathname.new(search_tool.database).realpath.to_s
+else
+  current_db=search_tool.current_database :fasta
+end
+
+fragment_tol = search_tool.fragment_tol
+precursor_tol = search_tool.precursor_tol
+
+
+throw "When --output is set only one file at a time can be run" if  ( ARGV.length> 1 ) && ( search_tool.explicit_output!=nil ) 
+
+# Run the search engine on each input file
+#
+ARGV.each do |filename|
+
+  if ( search_tool.explicit_output!=nil)
+    output_path=search_tool.explicit_output
+  else
+    output_path="#{search_tool.output_base_path(filename.chomp)}.mzid"
+  end
+  
+  # (*.mzML, *.mzXML, *.mgf, *.ms2, *.pkl or *_dta.txt)
+  # Get the input file extension
+  ext = Pathname.new(filename).extname
+
+
+
+  input_path="#{search_tool.input_base_path(filename.chomp)}#{ext}"
+
+  # Only proceed if the output file is not present or we have opted to over-write it
+  #
+  if ( search_tool.over_write || !Pathname.new(output_path).exist? )
+  
+    # The basic command
+    #
+    cmd= "java -Xmx#{search_tool.java_mem} -jar #{msgf_bin} -d #{current_db} -s #{input_path} -o #{output_path} "
+
+    #Missed cleavages
+    #
+    throw "Maximum value for missed cleavages is 2" if ( search_tool.missed_cleavages > 2)
+    cmd << " -ntt #{search_tool.missed_cleavages}"
+
+    # Precursor tolerance
+    #
+    cmd << " -t #{search_tool.precursor_tol}#{search_tool.precursor_tolu}"
+    
+    # Instrument type
+    #
+    cmd << " -inst 2"
+    
+#    cmd << " -m 4"
+
+    cmd << " -addFeatures 1"
+
+    # Enzyme
+    #
+  #    if ( search_tool.enzyme!="Trypsin")
+  #      cmd << " -e #{search_tool.enzyme}"
+  #    end
+
+  mods_path="#{search_tool.input_base_path(filename.chomp)}.msgfplus_mods.txt"
+  mods_file=File.open(mods_path,'w+')
+
+    # Variable Modifications
+    #
+    if ( search_tool.var_mods !="" && !search_tool.var_mods =~/None/) # Checking for none is to cope with galaxy input
+      var_mods = search_tool.var_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject {|e| e.empty? }.join(",")
+      if ( var_mods !="" )
+        cmd << " -mv #{var_mods}"
+      end
+    else 
+      # Add options related to peptide modifications
+      #
+      if ( search_tool.glyco )
+        cmd << " -mv 119 "
+      end
+    end
+
+  # Fixed modifications
+  #
+    if ( search_tool.fix_mods !="" && !search_tool.fix_mods=~/None/)
+      fix_mods = search_tool.fix_mods.split(",").collect { |mod| mod.lstrip.rstrip }.reject { |e| e.empty? }.join(",")
+      if ( fix_mods !="")
+        cmd << " -mf #{fix_mods}"    
+      end
+    else
+      if ( search_tool.has_modifications )
+        cmd << " -mf "
+        if ( search_tool.carbamidomethyl )
+          cmd<<"3 "
+        end
+
+        if ( search_tool.methionine_oxidation )
+          cmd<<"1 "
+        end
+
+      end
+    end
+    
+    # Up to here we've formulated the omssa command. The rest is cleanup
+    p "Running:#{cmd}"
+    
+    # Run the search
+    #
+    job_params= {:jobid => search_tool.jobid_from_filename(filename) }
+    job_params[:queue]="lowmem"
+    job_params[:vmem]="900mb"    
+    search_tool.run(cmd,genv,job_params)
+
+
+  else
+    genv.log("Skipping search on existing file #{output_path}",:warn)       
+  end
+
+end