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 = true
  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       puts "Input files:"
  62       input_files.each do | input |
  63         puts input
  64       end
  65       puts
  66
  67       counter = 1
  68       input_files.each do | input |
  69
  70         puts counter.to_s + "/" +  input_files.size.to_s + " " + input + ": "
  71
  72         counter += 1
  73
  74         hmm_name = ""
  75         id_norm = false
  76         orig_input = input
  77
  78         if input.downcase.end_with?( "_ni.fasta" )
  79           hmm_name = input[ 0 .. input.length - 10 ]
  80         elsif input.downcase.end_with?( ".fasta" )
  81           hmm_name = input[ 0 .. input.length - 7 ]
  82           unless File.exist? hmm_name
  83             id_norm = true
  84             puts
  85             puts "a. identifier normalization:"
  86             cmd = "#{TAP} #{input} #{hmm_name}_ni.fasta #{hmm_name}.nim"
  87             run_command( cmd )
  88             input = hmm_name + "_ni.fasta"
  89           else
  90             input = hmm_name + "/" + hmm_name + "_ni.fasta"
  91             unless File.exist? input
  92               error "expected to already exist: " + input
  93             end
  94             puts "a. identifier normalization already done:" + input
  95           end
  96         else
  97           error "illegal name: " + input
  98         end
  99
 100         unless File.exist? hmm_name
 101           Dir.mkdir( hmm_name )
 102         end
 103
 104         puts
 105         hmmscan_output = hmm_name + "/" + hmm_name + "_hmmscan_" + e_for_hmmscan.to_s
 106         unless File.exist? hmmscan_output
 107           puts "b. hmmscan:"
 108           cmd = "#{HMMSCAN} #{hmmscan_option} --domtblout #{hmmscan_output} -E #{e_for_hmmscan.to_s} #{PFAM}Pfam-A.hmm #{input}"
 109           run_command( cmd )
 110         else
 111           puts "b. hmmscan output already exists: " + hmmscan_output
 112         end
 113         puts
 114
 115
 116         hsp_output = hmm_name + "/" + hmm_name + "_hmmscan_#{e_for_hmmscan.to_s}_domain_table"
 117         unless File.exist? hsp_output
 118           puts "c. hmmscan to simple domain table:"
 119           cmd = "#{HSP} #{hmmscan_output} #{hsp_output}"
 120           run_command( cmd )
 121         else
 122           puts "c. hmmscan to simple domain table output already exists: " + hsp_output
 123         end
 124         puts
 125
 126         d2f_output = "#{hmm_name}/#{hmm_name}_hmmscan_#{e_for_hmmscan.to_s}.dff"
 127         unless File.exist? d2f_output
 128           puts "d. domain table to forester format:"
 129           cmd = "#{D2F} -e=10 #{hsp_output} #{input} #{d2f_output}"
 130           run_command( cmd )
 131         else
 132           puts "d. domain table to forester format output already exists: " + d2f_output
 133         end
 134         puts
 135
 136
 137         dsx_output = "#{hmm_name}/#{hmm_name}__#{hmm_name}__ee#{e_value_exp.to_s}_#{length}"
 138         unless File.exist? dsx_output
 139           puts "e. dsx:"
 140           cmd = "#{DSX} -d -e=1e-#{e_value_exp.to_s} -l=#{length} #{hmm_name} #{hmmscan_output} #{input} #{dsx_output}"
 141           run_command( cmd )
 142         else
 143           puts "e. dsx output already exists: " + dsx_output
 144         end
 145         puts
 146
 147         if id_norm
 148           FileUtils.mv "#{hmm_name}_ni.fasta", "#{hmm_name}/#{hmm_name}_ni.fasta"
 149           FileUtils.mv "#{hmm_name}.nim", "#{hmm_name}/#{hmm_name}.nim"
 150           FileUtils.cp orig_input, "#{hmm_name}/#{orig_input}"
 151         end
 152
 153         msa_dir = hmm_name + "/msa_ee#{e_value_exp.to_s}_#{length}"
 154         msa_100_dir =hmm_name + "/msa100_ee#{e_value_exp.to_s}_#{length}"
 155
 156         unless File.exist? msa_dir
 157           Dir.mkdir( msa_dir )
 158         end
 159         unless File.exist? msa_100_dir
 160           Dir.mkdir( msa_100_dir )
 161         end
 162
 163         run_1 = false
 164         run_100 = false
 165
 166         unless File.exist? "#{msa_dir}/#{dsx_output}"
 167           run_1 = true
 168           FileUtils.cp "#{dsx_output}.fasta", "#{msa_dir}/#{dsx_output}"
 169         end
 170
 171         unless File.exist? "#{msa_100_dir}/#{dsx_output}"
 172           run_100 = true
 173           FileUtils.cp "#{dsx_output}.fasta", "#{msa_100_dir}/#{dsx_output}"
 174         end
 175
 176         if File.exist?( TEMPLATE_FILE )
 177           if run_1
 178             FileUtils.cp TEMPLATE_FILE, msa_dir
 179           end
 180           if run_100
 181             FileUtils.cp TEMPLATE_FILE, msa_100_dir
 182           end
 183
 184           if LAUNCH_ANALYSIS
 185             puts "f. analysis:"
 186             if run_1
 187               Dir.chdir msa_dir
 188               run_command "#{PF} -b=1 -s"
 189               Dir.chdir "../.."
 190             end
 191             if run_100
 192               Dir.chdir msa_100_dir
 193               run_command "#{PF} -b=100 -s"
 194               Dir.chdir "../.."
 195             end
 196             puts
 197           end
 198         end
 199
 200       end
 201
 202     end
 203
 204     def run_command cmd
 205       puts cmd
 206       `#{cmd}`
 207     end
 208
 209     def get_base_name n
 210       if n.downcase.end_with?( "_ni.fasta" )
 211         n[ 0 .. n.length - 10 ]
 212       elsif n.downcase.end_with?( ".fasta" )
 213         n[ 0 .. n.length - 7 ]
 214       else
 215         error "illegal name: " + n
 216       end
 217     end
 218
 219     def error msg
 220       puts
 221       puts msg
 222       puts
 223       exit
 224     end
 225
 226   end
 227
 228   p = RunPhyloPipeline.new()
 229
 230   p.run()
 231
 232 end
 233
 234
 235