protk 1.2.0 → 1.2.1

data/bin/omssa_search.rb

@@ -40,6 +40,20 @@ search_tool.option_parser.on( '--galaxy-index-dir dir', 'Specify galaxy index di
   search_tool.options.galaxy_index_dir=dir
 end
 
+search_tool.options.omx_output=nil
+search_tool.option_parser.on( '--omx-output path', 'Specify path for additional OMX output (optional).' ) do |path|
+  search_tool.options.omx_output=path
+end
+
+if ( ENV['PROTK_OMSSA_NTHREADS'] )
+  search_tool.options.nthreads=ENV['PROTK_OMSSA_NTHREADS']
+else
+  search_tool.options.nthreads=0
+end
+search_tool.option_parser.on( '--nthreads num', 'Number of search threads to use. Default is to use the value in environment variable PROTK_OMSSA_NTHREADS or else to autodetect' ) do |num|
+  search_tool.options.nthreads=num
+end
+
 search_tool.option_parser.parse!
 
 # Environment with global constants
@@ -96,7 +110,7 @@ ARGV.each do |filename|
   
     # The basic command
     #
-    cmd = "#{make_blastdb_cmd} #{genv.omssacl} -d #{current_db} -fm #{input_path} -op #{output_path} -w"
+    cmd = "#{make_blastdb_cmd} #{genv.omssacl} -nt #{search_tool.nthreads} -d #{current_db} -fm #{input_path} -op #{output_path} -w"
 
     #Missed cleavages
     #
@@ -118,6 +132,10 @@ ARGV.each do |filename|
       end
     end
 
+    if ( search_tool.omx_output )
+      cmd << " -ox #{search_tool.omx_output} "
+    end
+
 
     # Precursor tolerance
     #
@@ -150,8 +168,9 @@ ARGV.each do |filename|
 
     # Variable Modifications
     #
-    if ( search_tool.var_mods !="" && !search_tool.var_mods =~/None/) # Checking for none is to cope with galaxy input
+    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
@@ -163,7 +182,7 @@ ARGV.each do |filename|
       end
     end
 
-    if ( search_tool.fix_mods !="" && !search_tool.fix_mods=~/None/)
+    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}"    
@@ -182,7 +201,7 @@ ARGV.each do |filename|
       end
     end
     
-    if ( search_tool.searched_ions !="" && !search_tool.searched_ions=~/None/)
+    if ( search_tool.searched_ions !="" && !(search_tool.searched_ions=~/None/))
       searched_ions=search_tool.searched_ions.split(",").collect{ |mod| mod.lstrip.rstrip }.reject { |e| e.empty? }.join(",")
       if ( searched_ions!="")
         cmd << " -i #{searched_ions}"
@@ -204,19 +223,23 @@ ARGV.each do |filename|
     # Intensity cut-off
     cmd << " -ci #{search_tool.intensity_cut_off}"
     
+    # Send output to logfile. OMSSA Logging does not play well with Ruby Open4
+    cmd << " -logfile omssa.log"
+
     # Up to here we've formulated the omssa command. The rest is cleanup
     p "Running:#{cmd}"
     
     # Add retention time corrections
     #
     if (search_tool.options.add_retention_times)
-      cmd << "; #{rt_correct_bin} #{output_path} #{input_path} "
+      # TODO: Really correct rts
+#      cmd << "; #{rt_correct_bin} #{output_path} #{input_path} "
     end
     
     # Correct the pepXML file 
     #
-    cmd << "; #{repair_script_bin} -N #{input_path} -R mgf #{output_path} --omssa-itol #{search_tool.fragment_tol}"
-    genv.log("Running repair script command #{cmd}",:info)
+#    cmd << "; #{repair_script_bin} -N #{input_path} -R mgf #{output_path} --omssa-itol #{search_tool.fragment_tol}"
+#    genv.log("Running repair script command #{cmd}",:info)
     
     # Run the search
     #
@@ -234,4 +257,4 @@ ARGV.each do |filename|
   #
   make_blastdb_cmd=""
 
-end
+end

data/bin/protk_setup.rb

@@ -41,8 +41,5 @@ env=Constants.new
 
 toolname=ARGV.shift
 
-p ARGV
-p toolname
-
 p "Installing #{toolname}"
 tool.install toolname

data/bin/sixframe.rb

@@ -40,12 +40,6 @@ file.each do |entry|
         position_start = position
         position_end = position_start + orf.length*3 -1
 
-        if ( frame > 3)
-            position_start = length - (position - 1)
-            position_end = position_start - orf.length * 3 + 1          
-        end
-
-
         # Create accession compliant with NCBI naming standard
         # See http://www.ncbi.nlm.nih.gov/books/NBK7183/?rendertype=table&id=ch_demo.T5
         ncbi_scaffold_id = entry.entry_id.gsub('|','_').gsub(' ','_')

data/bin/tandem_search.rb

@@ -48,7 +48,18 @@ search_tool.option_parser.on( '--tandem-output tandem_output', 'Keep X! Tandem O
 end
 
 search_tool.options.thresholds_type = 'isb_kscore'
-search_tool.option_parser.on( '--thresholds-type thresholds_type', 'Threshold Type (tandem_default, isb_native, isb_kscore, scaffold)' ) do |thresholds_type|
+search_tool.option_parser.on( '--thresholds-type thresholds_type', 'Threshold Type (tandem_default, isb_native, isb_kscore, scaffold, system_default)' ) do |thresholds_type|
+  # This options sets up various X! Tandem thresholds. 
+  #  - system_default: Don't change any defaults just use
+  #      the defaults for this TPP install as is.
+  #  - tandem_default: These thresholds are found on the 
+  #      tandem api page. http://www.thegpm.org/tandem/api/index.html
+  #  - isb_native: These are the defaults found in 
+  #      isb_default_input_native.xml distributed with TPP 4.6.
+  #  - isb_kscore: These are the defaults found in 
+  #      isb_default_input_kscore.xml distributed with TPP 4.6.
+  #  - scaffold: These are the defaults recommend by Proteome Software
+  #      for use with Scaffold.
   search_tool.options.thresholds_type = thresholds_type
 end
 
@@ -57,11 +68,12 @@ search_tool.option_parser.on( '--algorithm algorithm', "Scoring algorithm (kscor
   search_tool.options.algorithm = algorithm
 end
 
-search_tool.options.cleavage_semi = false
-search_tool.option_parser.on( '--cleavage-semi' ) do 
-  search_tool.options.cleavage_semi = true
+search_tool.options.cleavage_semi = true
+search_tool.option_parser.on( '--no-cleavage-semi' ) do
+  search_tool.options.cleavage_semi = false
 end
 
+
 search_tool.options.n_terminal_mod_mass=nil
 search_tool.option_parser.on('--n-terminal-mod-mass mass') do |mass|
     search_tool.options.n_terminal_mod_mass = mass
@@ -122,9 +134,16 @@ def decode_modification_string(mstring)
   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
+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)
+
   # Set the input and output paths 
   #
   input_notes=std_params.find('/bioml/note[@type="input" and @label="spectrum, path"]')
@@ -184,6 +203,78 @@ def generate_parameter_doc(std_params,output_path,input_path,taxo_path,current_d
     isotopic_error[0].content="no"
   end
   
+  if search_tool.tandem_output
+    # If one is interested in the tandem output (e.g. for consumption by Scaffold)
+    # want to store additional information.
+    set_option(std_params, "output, spectra", "yes")
+  end
+
+  thresholds_type = search_tool.thresholds_type
+
+  if thresholds_type != "system_default"
+
+    maximum_valid_expectation_value = "0.1"
+    if thresholds_type == "scaffold"
+      maximum_valid_expectation_value = "1000"
+    end 
+    
+    minimum_ion_count = "4"
+    case thresholds_type 
+    when "isb_kscore", "isb_native"
+      minimum_ion_count = "1"
+    when "scaffold"
+      minimum_ion_count = "0"
+    end
+    
+    minimum_peaks = "15"
+    case thresholds_type
+    when "isb_native"
+      minimum_peaks = "6"
+    when "isb_kscore"
+      minimum_peaks = "10"
+    when "scaffold"
+      minimum_peaks = "0"
+    end
+    
+    minimum_fragement_mz = "150"
+    case thresholds_type
+    when "isb_native"
+      minimum_fragement_mz = "50"
+    when "isb_kscore"
+      minimum_fragement_mz = "125"
+    when "scaffold"
+      minimum_fragement_mz = "0"
+    end
+    
+    minimum_parent_mh = "500" # tandem and isb_native defaults
+    case thresholds_type
+    when "isb_kscore"
+      minimum_parent_mh = "600"
+    when "scaffold"
+      minimum_parent_mh = "0"
+    end
+    
+    use_noise_suppression = "yes"
+    if thresholds_type == "isb_kscore" or thresholds_type == "scaffold"
+      use_noise_suppression = "no"
+    end
+    
+    dynamic_range = "100.0"
+    case thresholds_type
+    when "isb_kscore"
+      dynamic_range = "10000.0"
+    when "scaffold"
+      dynamic_range = "1000.0"
+    end
+
+    set_option(std_params, "spectrum, dynamic range", dynamic_range)
+    set_option(std_params, "spectrum, use noise suppression", use_noise_suppression)
+    set_option(std_params, "spectrum, minimum parent m+h", minimum_parent_mh)
+    set_option(std_params, "spectrum, minimum fragment mz", minimum_fragement_mz)
+    set_option(std_params, "spectrum, minimum peaks", minimum_peaks)
+    set_option(std_params, "scoring, minimum ion count", minimum_ion_count)
+    set_option(std_params, "output, maximum valid expectation value", maximum_valid_expectation_value)
+  end
   
   # Fixed and Variable Modifications
   #

data/bin/template_search.rb

@@ -0,0 +1,144 @@
+#!/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
+#
+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 msms search on a set of msms spectrum input files.\n\nUsage: template_search.rb [options] file1.mzML file2.mzML ..."
+search_tool.options.output_suffix="_template"
+
+search_tool.options.custom_option="default"
+search_tool.option_parser.on('--custom-opt value','Custom option relevant to this tool only (Default default)') do |val|
+  search_tool.options.custom_option=val
+end
+  
+search_tool.option_parser.parse!
+
+# Set search engine specific parameters on the SearchTool object
+#
+msgf_bin="#{genv.msgf_bin}/MSGFPlus.jar"
+
+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

data/bin/uniprot_annotation.rb

@@ -0,0 +1,130 @@
+#!/usr/bin/env ruby
+#
+# This file is part of Protk
+# Created by Ira Cooke 24/3/2013
+#
+# Retrieve annotation information for proteins from the Uniprot Swissprot database
+#
+# 
+require 'protk/constants'
+require 'protk/command_runner'
+require 'protk/tool'
+require 'protk/swissprot_database'
+require 'protk/bio_sptr_extensions'
+
+
+# Setup specific command-line options for this tool. Other options are inherited from Tool
+#
+tool=Tool.new({:explicit_output=>true})
+tool.option_parser.banner = "Retrieve information from the Uniprot database given a list of ID's.\n\n\
+Usage: uniprot_annotation.rb [options] input.tsv"
+
+tool.options.id_column=1
+tool.option_parser.on(  '--id-column num', 'Specify a column for ids (default is column 1)' ) do |col|
+  tool.options.id_column=col.to_i
+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
+end
+
+tool.option_parser.parse!
+
+# Obtain a global environment object
+genv=Constants.new
+
+input_file=ARGV[0]
+
+swissprotdb=SwissprotDatabase.new(genv)
+
+output_file=nil
+
+if ( tool.explicit_output==nil)
+  output_file=$stdout
+else
+  output_file=File.open(tool.explicit_output,'w+')
+end
+
+ac_column = tool.id_column-1
+
+db_fields = {
+  'recname'=>"Primary Name",
+  'cd'=>"CD Antigen Name",
+  'altnames'=>"Alternate Names", 
+  'location' => "Subcellular Location",
+  'function' => "Known Function",
+  'similarity' => "Similarity",
+  'tissues' => "Tissue Specificity",
+  'disease' => "Disease Association",
+  'domain' => "Domain",
+  'subunit' => "Sub Unit",
+  'nextbio' => "NextBio",
+  'ipi' => "IPI",
+  'intact' => "Interactions",
+  'pride' => 'Pride',
+  'ensembl'=> 'Ensembl',
+  'num_transmem'=>"Transmembrane Regions",
+  'signalp'=>'Signal Peptide',
+  'ref_dump'=>'References',
+  'tax_dump'=>'Taxonomy Cross Ref',
+  'species_dump'=>'Species',
+  'feature_dump'=>'Feature Table',
+  'seq_dump' => 'AA Sequence'
+  }
+
+hyperlink_fields = {
+  'uniprot_link'=>"Uniprot Link",
+  'nextbio_link'=>'NextBio Link',
+  'intact_link'=>"Interactions Link",
+  'pride_link'=>"Pride Link",
+  'ensembl_link'=>"Ensembl Link"
+}
+
+if tool.fields !=nil
+  fields = tool.fields.split(",").collect { |f| f.lstrip.rstrip }.reject {|e| e.empty? }
+  db_fields = db_fields.select { |k| fields.include? k }
+  hyperlink_fields = hyperlink_fields.select { |k| fields.include? k}
+end
+
+output_file.write db_fields.values.join("\t")
+if ( hyperlink_fields.count > 0 )
+  output_file.write("\t")
+  output_file.write hyperlink_fields.values.join("\t")
+end
+output_file.write("\n")
+
+line_num=0
+File.foreach(input_file) { |line|  
+  input_cols=line.split("\t")
+  throw "Not enough columns in line #{line_num}" unless input_cols.count > ac_column
+  accession=input_cols[ac_column].chomp
+
+  sptr_entry=swissprotdb.get_entry_for_name(accession)
+
+  if ( sptr_entry==nil)
+    genv.log("No entry for #{accession} in uniprot database",:warn)  
+  else
+
+    db_values = db_fields.collect { |key,value|     
+      sptr_entry.send(key)
+    }
+
+    hyperlink_values = hyperlink_fields.collect { |key,value|
+      sptr_entry.send(key)
+    }
+
+    output_file.write db_values.join("\t")
+    if ( hyperlink_fields.count > 0 )
+      output_file.write("\t")
+      output_file.write hyperlink_values.join("\t")
+    end
+    output_file.write "\n"
+  end
+
+  line_num+=1
+
+}
+
+
+