protk 1.2.4 → 1.2.5

data/bin/add_retention_times.rb

@@ -0,0 +1,89 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 14/12/2010
+#
+# Attempts to add retention times to pepxml by looking up retention times in a raw file
+#
+
+require 'protk/constants'
+require 'protk/command_runner'
+require 'protk/tool'
+require 'libxml'
+require 'protk/mascot_util'
+include LibXML
+
+# Environment with global constants
+#
+genv=Constants.new
+
+tool=Tool.new([:over_write])
+tool.option_parser.banner = "Look up retention times in a raw file and \
+add them to a pepxml file.\n\nUsage: add_retention_times.rb [options] file1.pep.xml file2.mgf"
+
+exit unless tool.check_options 
+
+if ( ARGV[0].nil? || ARGV[1].nil? )
+    puts "You must supply an input file"
+    puts tool.option_parser 
+    exit
+end
+
+pepxml_file=ARGV[0]
+mgf_file=ARGV[1]
+
+pepxml_parser=XML::Parser.file(pepxml_file)
+
+begin
+	"Creating mascot spectrum id table"
+	rt_table=MascotUtil.index_mgf_times(mgf_file)
+rescue
+	puts "Unable to index retention times in mgf file"
+	exit
+end
+
+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
+
+
+
+queries=pepxml_doc.find("//#{pepxml_ns_prefix}spectrum_query", pepxml_ns)
+
+queries.each do |query|
+
+	atts=query.attributes
+	spect=atts["spectrum"]
+
+
+	throw "No spectrum found for spectrum_query #{query}" unless ( spect!=nil)
+
+	retention_time = rt_table[spect]
+	if retention_time==nil
+		retention_time=rt_table[spect.chop]
+		if retention_time==nil
+			retention_time=rt_table[spect.chop.chop]
+		end
+	end
+	if ( retention_time!=nil)
+      
+		if ( query.attributes["retention_time_sec"]!=nil )
+			puts "A retention time value is already present" 
+			exit
+		end
+
+		if ( query.attributes["retention_time_sec"]==nil || over_write)
+			query.attributes["retention_time_sec"]=retention_time       
+     		# p queries[i].attributes["retention_time_sec"] 
+     	end
+    else 
+    		puts "No retention time found for spectrum #{spect}"
+	end
+end
+
+pepxml_doc.save(pepxml_file)

data/bin/augustus_to_proteindb.rb

@@ -0,0 +1,193 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+# Created by Ira Cooke 4/9/2013
+#
+# 
+
+require 'protk/constants'
+require 'protk/tool'
+require 'bio'
+
+tool=Tool.new([:explicit_output])
+tool.option_parser.banner = "Create a protein database from Augustus gene prediction output that is suitable for later processing by proteogenomics tools.\n\nUsage: augustus_to_proteindb.rb [options] augustus.gff3"
+
+tool.options.add_transcript_info=false
+tool.option_parser.on( '--info', 'Include CDS coordinates' ) do  
+  tool.options.add_transcript_info=true
+end
+
+exit unless tool.check_options 
+
+if ( ARGV[0].nil? )
+    puts "You must supply an input file"
+    puts tool.option_parser 
+    exit
+end
+
+inname=ARGV.shift
+
+$add_transcript_info=tool.add_transcript_info
+
+$print_progress=true
+
+outfile=nil
+if ( tool.explicit_output != nil)
+  outfile=File.open(tool.explicit_output,'w')
+else
+  outfile=$stdout
+  $print_progress=false
+end
+
+
+def get_transcript_lines(gene_lines)
+  transcripts=[]
+  gene_lines.each do |line|  
+    if line =~ /transcript\t(\d*?)\t/
+      transcripts << line
+    end
+  end
+  transcripts
+end
+
+def get_cds_lines(gene_lines)
+  coding_sequences=[]
+  gene_lines.each do |line|  
+    if line =~ /CDS\t(\d*?)\t/
+      coding_sequences << line
+    end
+  end
+  coding_sequences
+end
+
+$capturing_protein=false
+
+def capture_protein_start(line)
+    if line=~/protein sequence = \[/
+      $capturing_protein=true     
+    end
+end
+
+def at_protein_end(line)
+  if $capturing_protein && line =~ /# .*?\]/
+    return true    
+  end
+  return false
+end
+
+def get_protein_sequence_lines(gene_lines)
+  $capturing_protein=false
+  proteins=[]
+  current_protein_lines=[]
+  gene_lines.each do |line|  
+    capture_protein_start(line)
+    if at_protein_end(line)
+      current_protein_lines << line
+      proteins << current_protein_lines 
+      current_protein_lines=[]
+      $capturing_protein=false   
+    else
+      current_protein_lines << line if $capturing_protein
+    end
+  end
+  proteins
+end
+
+def cds_to_header_text(coding_sequence,transcript_id)
+#  require 'debugger';debugger
+  imatch=coding_sequence.match(/CDS\t(\d+)\t(\d+).*?([-\+]{1}.*?Parent=#{transcript_id})$/)
+  if imatch==nil
+    return ""
+  end
+  istart=imatch[1]
+  iend=imatch[2]
+  "#{istart}|#{iend}"
+end
+
+def sequence_fasta_header(transcript_line,coding_sequences,scaffold)
+
+  tmatch=transcript_line.match(/transcript\t(\d+)\t(\d+).*?([-\+]{1}).*?ID=(.*?);/)
+#  require 'debugger'; debugger
+  tstart=tmatch[1]
+  tend=tmatch[2]
+  tstrand="fwd"
+  tstrand = "rev" if tmatch[3]=="-"
+
+  tid=tmatch[4]
+  header=">lcl|#{scaffold}_#{tstrand}_#{tid} #{tstart}|#{tend}"
+  if $add_transcript_info
+    coding_sequences.each { |coding_sequence| header << " #{cds_to_header_text(coding_sequence,tid)}" }
+  end
+  header
+end
+
+def protein_sequence(protein_lines)
+  seq=""
+  protein_lines.each_with_index do |line, i|  
+      seq << line.match(/(\w+)\]?$/)[1]
+ end
+
+  seq
+end
+
+def parse_gene(gene_lines)
+
+  geneid=gene_lines[0].match(/start gene (.*)/)[1]
+  transcripts=get_transcript_lines(gene_lines)
+  coding_sequences=get_cds_lines(gene_lines)
+  proteins=get_protein_sequence_lines(gene_lines)
+  fasta_string=""
+  throw "transcripts/protein mismatch" unless transcripts.length == proteins.length
+  transcripts.each_with_index do |ts, i|  
+    fh=sequence_fasta_header(ts,coding_sequences,$current_scaffold)
+    fasta_string << "#{fh}\n"
+    ps=protein_sequence(proteins[i])  
+    fasta_string << "#{ps}\n"
+  end
+
+  gene_lines=[]
+  $capturing_gene=false
+  fasta_string
+end
+
+def capture_scaffold(line)
+  if line =~ /-- prediction on sequence number.*?name = (.*)\)/
+    $current_scaffold=line.match(/-- prediction on sequence number.*?name = (.*)\)/)[1]
+    if ( $print_progress)
+      puts $current_scaffold
+    end
+  end
+end
+
+def capture_gene_start(line)
+  if line =~ /# start gene/
+    $capturing_gene=true
+  end
+end
+
+def at_gene_end(line)
+  if line =~ /# end gene/
+    return true
+  end
+  return false
+end
+
+$current_scaffold=""
+gene_lines=[]
+$capturing_gene=false
+
+
+File.open(inname).each_with_index do |line, line_i|  
+  line.chomp!
+  capture_scaffold(line)
+  capture_gene_start(line)
+
+  if at_gene_end(line)
+    gene_string=parse_gene(gene_lines)
+    outfile.write gene_string
+    gene_lines=[]
+  else
+    gene_lines << line if $capturing_gene
+  end
+
+end

data/bin/blastxml_to_table.rb

@@ -0,0 +1,72 @@
+#!/usr/bin/env ruby
+#
+# This file is part of protk
+#
+# 
+
+require 'protk/constants'
+require 'protk/tool'
+require 'bio'
+require 'protk/fastadb'
+require 'bio-blastxmlparser'
+
+
+tool=Tool.new([:explicit_output])
+tool.option_parser.banner = "Dump BLAST xml to tabular format.\n\nUsage: blastxml_to_table.rb blast.xml"
+
+tool.options.database=nil
+tool.option_parser.on( '-d filename','--database filename', 'Database used for BLAST search. If provided, hit sequences will be looked up in this database' ) do |file| 
+  tool.options.database=file
+end
+
+exit unless tool.check_options 
+
+#require 'debugger';debugger
+
+exit unless ARGV.length == 1
+input_file=ARGV[0]
+
+out_file=$stdout
+if ( tool.explicit_output != nil)
+  out_file=File.open(tool.explicit_output, "w")
+end
+
+$fastadb = nil
+if tool.database
+	$fastadb=FastaDB.new(tool.database)
+end
+
+def generate_line(hsp,hit,query,hit_seq=nil)
+	line="#{query.query_id}\t#{hit.hit_id}\t#{hit.hit_num}\t#{hit.hit_def}\t#{hsp.hsp_num}\t#{hsp.bit_score}\t#{hsp.evalue}\t#{hsp.qseq}\t#{hsp.hseq}"
+	if hit_seq
+		line << "\t#{hit_seq}"
+	end
+	line<<"\n"
+	line
+end
+
+def fetch_hit_seq(hit)
+	hit_seq=nil
+	if $fastadb
+		hit_seq=$fastadb.fetch(hit.hit_id).first.aaseq
+	end
+	hit_seq
+end
+
+blast = Bio::BlastXMLParser::XmlSplitterIterator.new(input_file).to_enum
+
+blast.each do |query|  
+	query.hits.each do |hit|
+#	hit=query.hits.first
+#	if hit
+		hit_seq=fetch_hit_seq(hit)
+		hit.hsps.each do |hsp|
+			out_file.write generate_line(hsp,hit,query,hit_seq)
+		end
+#	end
+	end
+end
+
+#require 'debugger';debugger
+
+#puts "Hi"

data/bin/feature_finder.rb

@@ -10,6 +10,7 @@ require 'protk/command_runner'
 require 'protk/tool'
 require 'protk/openms_defaults'
 require 'libxml'
+require 'tempfile'
 
 include LibXML
 
@@ -62,7 +63,9 @@ end
 
 throw "Cannot use explicit output in combination with multiple input files" if ( tool.explicit_output && ARGV.length>1)
 
-ini_file="#{Pathname.new(ARGV[0]).dirname.realpath.to_s}/feature_finder.ini"
+input_basename=Pathname.new(ARGV[0]).basename.to_s
+ini_file_name="#{Pathname.new(Tempfile.new(input_basename).path).basename.to_s}_feature_finder.ini"
+ini_file="#{Pathname.new(ini_file_name).dirname.realpath.to_s}/#{ini_file_name}"
 
 generate_ini(tool,ini_file)
 
@@ -74,6 +77,9 @@ ARGV.each do |filen|
   output_dir=Pathname.new(input_basename).dirname.realpath.to_s
   output_base=Pathname.new(input_basename).basename.to_s
   output_file = "#{output_dir}/#{tool.output_prefix}#{output_base}#{tool.output_suffix}.featureXML"
+  if ( tool.explicit_output )
+    output_file = "#{output_dir}/#{tool.explicit_output}"
+  end
 
   if ( tool.over_write || !Pathname.new(output_file).exist? )
     output_base_filename=Pathname.new(output_file).basename.to_s

data/bin/make_decoy.rb

@@ -11,6 +11,8 @@ require 'libxml'
 require 'protk/constants'
 require 'protk/command_runner'
 require 'protk/tool'
+require 'protk/randomize'
+require 'tempfile'
 require 'bio'
 
 include LibXML
@@ -60,8 +62,14 @@ output_file = tool.explicit_output if tool.explicit_output!=nil
 
 genv=Constants.new()
 
-Randomize.make_decoys #{input_file} #{db_length} #{output_file} #{tool.prefix_string}"
-cmd << "cat #{input_file} >> #{output_file}" if ( tool.append ) 
+decoys_tmp_file = Pathname.new(Tempfile.new("random").path).basename.to_s;
+
+Randomize.make_decoys input_file, db_length, decoys_tmp_file, tool.prefix_string
+cmd = "cat #{input_file} #{decoys_tmp_file} >> #{output_file}; rm #{decoys_tmp_file}" if ( tool.append ) 
+
+# Randomize.make_decoys raw_db_filename, db_length, decoys_filename, decoy_prefix
+#   cmd = "cat #{raw_db_filename} #{decoys_filename} >> #{decoy_db_filename}; rm #{decoys_filename}"
+
 p cmd
 # Run the conversion
 #

data/bin/mascot_search.rb

@@ -124,7 +124,7 @@ search_tool.option_parser.on('--username un', 'Username.') do |un|
     search_tool.options.username = un
 end
 
-search_tool.options.httpproxy=""
+search_tool.options.httpproxy=nil
 search_tool.option_parser.on( '--proxy url', 'The url to a proxy server' ) do |urll| 
     search_tool.options.httpproxy=urll
 end
@@ -144,6 +144,11 @@ search_tool.option_parser.on( '--export format', 'Save results in a specified fo
     search_tool.options.export_format=format
 end
 
+search_tool.options.timeout=200
+search_tool.option_parser.on( '--timeout seconds', 'Timeout for sending data file to mascot in seconds' ) do |seconds| 
+    search_tool.options.timeout=seconds.to_i
+end
+
 exit unless search_tool.check_options 
 
 if ( ARGV[0].nil? )
@@ -161,8 +166,7 @@ unless ( mascot_cgi =~ /^http[s]?:\/\//)
     mascot_cgi  = "http://#{mascot_cgi}"
 end
 
-RestClient.proxy=search_tool.httpproxy
-
+RestClient.proxy=search_tool.httpproxy if search_tool.httpproxy
 $genv.log("Var mods #{search_tool.var_mods} and fixed #{search_tool.fix_mods}",:info)
 
 cookie=""
@@ -177,7 +181,13 @@ end
 postdict = search_params_dictionary search_tool, ARGV[0]
 $genv.log("Sending #{postdict}",:info)
 
-search_response=RestClient.post "#{mascot_cgi}/nph-mascot.exe?1",  postdict, {:cookies=>cookie}
+#site = RestClient::Resource.new(mascot_cgi, timeout=300)
+#search_response=site['/nph-mascot.exe?1'].post , postdict, {:cookies=>cookie}
+
+search_response=RestClient::Request.execute(:method => :post, :url => "#{mascot_cgi}/nph-mascot.exe?1", :payload => postdict,:headers=>{:cookies=>cookie},:timeout => search_tool.options.timeout, :open_timeout => 10)
+
+
+#search_response=RestClient.post "#{mascot_cgi}/nph-mascot.exe?1", postdict, {:cookies=>cookie}
 
 $genv.log("Mascot search response was #{search_response}",:info)