forester/ruby/evoruby/exe/run_phylo_pipeline_x.rb

   1 #!/usr/local/bin/ruby -w
   2 #
   3 # = run_phylo_pipeline
   4 #
   5 # Copyright::  Copyright (C) 2010 Christian M. Zmasek
   6 # License::    GNU Lesser General Public License (LGPL)
   7 #
   8 # $Id Exp $
   9 #
  10 #
  11
  12 require 'fileutils'
  13
  14 module Evoruby
  15
  16   class RunPhyloPipeline
  17
  18     LAUNCH_ANALYSIS = false
  19     HOME          = "/home/czmasek/"
  20     FORESTER_RUBY = "#{HOME}SOFTWARE/FORESTER/DEV/forester/forester/ruby/evoruby/exe/"
  21     PFAM          = "#{HOME}DATA/PFAM/PFAM270X/"
  22     HMMSCAN       = "#{HOME}SOFTWARE/HMMER/hmmer-3.0/src/hmmscan"
  23     HSP           = "#{FORESTER_RUBY}hsp.rb"
  24     D2F           = "#{FORESTER_RUBY}d2f.rb"
  25     DSX           = "#{FORESTER_RUBY}dsx.rb"
  26     TAP           = "#{FORESTER_RUBY}tap.rb"
  27     PF            = "#{FORESTER_RUBY}phylogeny_factory.rb"
  28     TEMPLATE_FILE = '00_phylogeny_factory.template'
  29
  30     def run
  31       unless ARGV.length >= 2 && ARGV.length <= 4
  32         error "arguments are:  <min-length> " +
  33          "<neg E-value exponent for domain extraction> [E-value for hmmscan, default is 10] [hmmscan option, default is --nobias, --max for no heuristics]"
  34       end
  35
  36       length      = ARGV[ 0 ].to_i
  37       e_value_exp = ARGV[ 1 ].to_i
  38
  39       e_for_hmmscan = 10
  40       hmmscan_option = "--nobias"
  41
  42       if ARGV.length == 4
  43         hmmscan_option = ARGV[ 3 ]
  44       end
  45       if ARGV.length == 3 || ARGV.length == 4
  46         e_for_hmmscan = ARGV[ 2 ].to_i
  47       end
  48
  49       if e_value_exp < 0
  50         error "E-value exponent for domain extraction cannot be negative"
  51       end
  52       if length <= 1
  53         error "length cannot be smaller than or equal to 1"
  54       end
  55       if e_for_hmmscan < 1
  56         error "E-value for hmmscan cannot be smaller than 1"
  57       end
  58
  59       input_files = Dir.entries(".").select { |f| !File.directory?( f ) && f.downcase.end_with?( ".fasta" ) }
  60
  61       input_files.sort!
  62
  63       puts "Input files:"
  64       input_files.each do | input |
  65         puts input
  66       end
  67       puts
  68
  69       counter = 1
  70       input_files.each do | input |
  71
  72         puts counter.to_s + "/" +  input_files.size.to_s + " " + input + ": "
  73
  74         counter += 1
  75
  76         hmm_name = ""
  77         id_norm = false
  78         orig_input = input
  79
  80         if input.downcase.end_with?( "_ni.fasta" )
  81           hmm_name = input[ 0 .. input.length - 10 ]
  82         elsif input.downcase.end_with?( ".fasta" )
  83           hmm_name = input[ 0 .. input.length - 7 ]
  84           unless File.exist? hmm_name
  85             id_norm = true
  86             puts
  87             puts "a. identifier normalization:"
  88             cmd = "#{TAP} #{input} #{hmm_name}_ni.fasta #{hmm_name}.nim"
  89             run_command( cmd )
  90             input = hmm_name + "_ni.fasta"
  91           else
  92             input = hmm_name + "/" + hmm_name + "_ni.fasta"
  93             unless File.exist? input
  94               error "expected to already exist: " + input
  95             end
  96             puts "a. identifier normalization already done: " + input
  97           end
  98         else
  99           error "illegal name: " + input
 100         end
 101
 102         unless File.exist? hmm_name
 103           Dir.mkdir( hmm_name )
 104         end
 105
 106         puts
 107         hmmscan_output = hmm_name + "/" + hmm_name + "_hmmscan_" + e_for_hmmscan.to_s
 108         unless File.exist? hmmscan_output
 109           puts "b. hmmscan:"
 110           cmd = "#{HMMSCAN} #{hmmscan_option} --domtblout #{hmmscan_output} -E #{e_for_hmmscan.to_s} #{PFAM}Pfam-A.hmm #{input}"
 111           run_command( cmd )
 112         else
 113           puts "b. hmmscan output already exists: " + hmmscan_output
 114         end
 115         puts
 116
 117
 118         hsp_output = hmm_name + "/" + hmm_name + "_hmmscan_#{e_for_hmmscan.to_s}_domain_table"
 119         unless File.exist? hsp_output
 120           puts "c. hmmscan to simple domain table:"
 121           cmd = "#{HSP} #{hmmscan_output} #{hsp_output}"
 122           run_command( cmd )
 123         else
 124           puts "c. hmmscan to simple domain table output already exists: " + hsp_output
 125         end
 126         puts
 127
 128         d2f_output = "#{hmm_name}/#{hmm_name}_hmmscan_#{e_for_hmmscan.to_s}.dff"
 129         unless File.exist? d2f_output
 130           puts "d. domain table to forester format:"
 131           cmd = "#{D2F} -e=10 #{hsp_output} #{input} #{d2f_output}"
 132           run_command( cmd )
 133         else
 134           puts "d. domain table to forester format output already exists: " + d2f_output
 135         end
 136         puts
 137
 138         dsx_output_base = "#{hmm_name}__#{hmm_name}__ee#{e_value_exp.to_s}_#{length}"
 139         dsx_output = hmm_name +"/" + dsx_output_base
 140         unless File.exist? dsx_output + ".fasta"
 141           puts "e. dsx:"
 142           cmd = "#{DSX} -d -e=1e-#{e_value_exp.to_s} -l=#{length} #{hmm_name} #{hmmscan_output} #{input} #{dsx_output}"
 143           run_command( cmd )
 144         else
 145           puts "e. dsx output already exists: " + dsx_output + ".fasta"
 146         end
 147         puts
 148
 149         if id_norm
 150           FileUtils.mv "#{hmm_name}_ni.fasta", "#{hmm_name}/#{hmm_name}_ni.fasta"
 151           FileUtils.mv "#{hmm_name}.nim", "#{hmm_name}/#{hmm_name}.nim"
 152           FileUtils.cp orig_input, "#{hmm_name}/#{orig_input}"
 153         end
 154
 155         msa_dir = hmm_name + "/msa_ee#{e_value_exp.to_s}_#{length}"
 156         msa_100_dir =hmm_name + "/msa100_ee#{e_value_exp.to_s}_#{length}"
 157
 158         unless File.exist? msa_dir
 159           Dir.mkdir( msa_dir )
 160         end
 161         unless File.exist? msa_100_dir
 162           Dir.mkdir( msa_100_dir )
 163         end
 164
 165         run_1 = false
 166         run_100 = false
 167
 168         unless File.exist? "#{msa_dir}/#{dsx_output_base}"
 169           run_1 = true
 170           FileUtils.cp "#{dsx_output}.fasta", "#{msa_dir}/#{dsx_output_base}"
 171         end
 172
 173         unless File.exist? "#{msa_100_dir}/#{dsx_output_base}"
 174           run_100 = true
 175           FileUtils.cp "#{dsx_output}.fasta", "#{msa_100_dir}/#{dsx_output_base}"
 176         end
 177
 178         if File.exist?( TEMPLATE_FILE )
 179           if run_1
 180             FileUtils.cp TEMPLATE_FILE, msa_dir
 181           end
 182           if run_100
 183             FileUtils.cp TEMPLATE_FILE, msa_100_dir
 184           end
 185
 186           if LAUNCH_ANALYSIS && ( run_1 || run_100 )
 187             puts "f. analysis:"
 188             if run_1
 189               Dir.chdir msa_dir
 190               run_command "#{PF} -b=1 -s"
 191               Dir.chdir "../.."
 192             end
 193             if run_100
 194               Dir.chdir msa_100_dir
 195               run_command "#{PF} -b=100 -s"
 196               Dir.chdir "../.."
 197             end
 198             puts
 199           end
 200         end
 201
 202       end
 203
 204     end
 205
 206     def run_command cmd
 207       puts cmd
 208       `#{cmd}`
 209     end
 210
 211     def get_base_name n
 212       if n.downcase.end_with?( "_ni.fasta" )
 213         n[ 0 .. n.length - 10 ]
 214       elsif n.downcase.end_with?( ".fasta" )
 215         n[ 0 .. n.length - 7 ]
 216       else
 217         error "illegal name: " + n
 218       end
 219     end
 220
 221     def error msg
 222       puts
 223       puts msg
 224       puts
 225       exit
 226     end
 227
 228   end
 229
 230   p = RunPhyloPipeline.new()
 231
 232   p.run()
 233
 234 end
 235
 236
 237