cwords 0.1.10 → 0.1.11

data/bin/cwords_mkdb

@@ -1,16 +1,6 @@
 #!/usr/bin/env ruby
 scriptdir = File.dirname(__FILE__) + "/../scripts/"
 
-helps = ARGV.join(" ").match(/--help/)
-if (helps) 
-  # start lightweight
-  exec "jruby --client " + scriptdir + "cwords_mkdb.rb " + ARGV.join(" ")
-end
-
-mems = ARGV.join(" ").match(/M=(\w+)/)
-mem = mems ? mems[1] : '4096m'
-argv = ARGV.select{|x| not x=~ /^M=\w+$/}.join(' ')
-puts "Starting cwords_mkdb with max heap size " + mem + " ...\n"
-
-cmd = "jruby --server --fast -J-Xmx#{mem} " + scriptdir + "cwords_mkdb.rb " + argv
+argv = ARGV.join(" ")
+cmd = "ruby " + scriptdir + "cwords_mkdb.rb " + argv
 exec cmd

data/lib/wordRS-lib.rb

@@ -5,7 +5,7 @@ class Array
   def threach(n = 1, &b)
     return [] if n == 0 or size == 0
     result = Array.new(size)
-    self.send(:each,&b) if n == 1
+    return self.send(:each,&b) if n == 1 # trying return here
     
     n = [n,size].min
     

data/scripts/cwords_mkdb.rb

@@ -1,4 +1,4 @@
-#!/usr/bin/env jruby
+#!/usr/bin/env ruby
 
 srcdir = File.dirname(__FILE__)
 basedir = srcdir + "/../"
@@ -25,13 +25,13 @@ options[:seqfile] = nil
 options[:partitions] = 1
 options[:stats] = ['p'] # p=p
 options[:ruby]='jruby --fast -J-Xmx1024m'
-options[:shuffles]=5000
+options[:shuffles]=500
 options[:bg]=1 #mononucleotide shuffling
 
 $coptions = OptionParser.new do |opts|
   opts.on("-w", "--wordsize ARG", "wordsize") { |o| options[:wordsize] = o.split(",").map{|x| x.to_i}}
   opts.on("-s", "--seqfile ARG", "sequence file") {|o| options[:seqfile] = o}
-  opts.on("-p", "--partitions ARG", "number of sequence partitions") {|o| options[:partitions] = o.to_i}
+  opts.on("-t", "--threads ARG", "number of concurrent processes") {|o| options[:partitions] = o.to_i}
   opts.on("-a", "--stats ARG", "sequence file") {|o| options[:stats] = o.split('')}
   opts.on("-u", "--shuffle ARG", "number of shuffles") {|o| options[:shuffles] = o.to_i}
   opts.on("--ruby ARG", "ruby interpreter") {|o| options[:ruby] = o}
@@ -72,7 +72,7 @@ end
       
 puts "starting #{n} processes ..."
 
-cmd = "#{options[:ruby]} #{basedir}/scripts/wordsrus_mkdb.rb"
+cmd = "#{options[:ruby]} #{basedir}/scripts/cwords_mkdb_worker.rb"
 cmd += " -w #{options[:wordsize].join(',')} -s #{options[:seqfile]} -a #{options[:stats].join(",")} -u #{options[:shuffles]} --bg #{options[:bg]}"
 
 stamp = Time.now.to_i

data/scripts/cwords_mkdb_worker.rb

@@ -0,0 +1,99 @@
+
+srcdir = File.dirname(__FILE__)
+basedir = srcdir + "/../"
+libdir = basedir + 'lib/'
+$LOAD_PATH << libdir
+
+require 'wordRS-lib.rb'
+require 'rubygems'
+require 'progressbar'
+require 'optparse'
+require 'fileutils'
+require 'java'
+require libdir + 'ushuffle.jar'
+java_import 'UShuffle'
+us = UShuffle.new
+
+###
+### Main
+###
+
+
+#default options
+options = Hash.new
+options[:wordsize] = [7]
+options[:seqfile] = nil
+options[:partition] = nil
+options[:stats] = ['p'] # p=p-value, z=z-score
+options[:shuffles]=1000
+options[:bg]=1 #mononucleotide shuffling
+
+$coptions = OptionParser.new do |opts|
+  opts.on("-w", "--wordsize ARG", "wordsize") { |o| options[:wordsize] = o.split(",").map{|x| x.to_i}.sort}
+  opts.on("-s", "--seqfile ARG", "sequence file (FASTA format)") {|o| options[:seqfile] = o}
+  opts.on("-p", "--partition ARG", "only process a partition (i.e. 5-10) of sequences") {|o| options[:partition] = o}
+  opts.on("-a", "--stats ARG", "sequence file") {|o| options[:stats] = o.split(',')}
+  opts.on("-u", "--shuffle ARG", "number of shuffles") {|o| options[:shuffles] = o.to_i}
+  opts.on("-b", "--bg ARG", "background nucleotide model") {|o| options[:bg] = o.to_i}
+end
+
+def show_help(msg="", code=0, io=STDOUT)
+  io.puts "#{msg}\n#{$coptions}"
+  exit(code)
+end
+
+$coptions.parse!(ARGV)
+#mandatory parameters
+[:seqfile].each{ |p| show_help("option '#{p}' mandatory") if options[p].nil?}
+
+exit("seqfile must have fasta-format") if !options[:seqfile].match(/.fa$/)
+dbdir = basedir + "/db/" + File.basename(options[:seqfile],'.fa') + "_bg#{options[:bg]}"
+FileUtils.mkdir_p dbdir # create dir if it does not exist
+
+decimals = 6
+bg = options[:bg]
+
+# word id's
+@wid = Hash.new
+i = 0
+options[:wordsize].each do |ws|
+  ['a','g','c','t'].rep_perm(ws) {|seqa| @wid[seqa.join('')]=i ; i+=1 }
+end
+
+@seqs = IO.readlines(options[:seqfile],">")[1..-1]
+if options[:partition]
+  puts "partition #{options[:partition]}"
+  (pstart,pstop) = options[:partition].split('-')
+  @seqs = @seqs[pstart.to_i-1..pstop.to_i-1]
+end
+
+puts "computing statistics for #{@seqs.size} sequences"
+pbar = ProgressBar.new("seqs",@seqs.size)
+
+@seqs.each do |s|
+  ff = s.split("\n")
+  id = ff.shift
+  seq = ff[0..-2].join('').downcase.gsub('u','t') # last field is ">"
+  # next if not nucleotide sequence, i.e. "unavailable"
+  next if (seq.split('').uniq - ['a','c','g','t']).size > 0
+  
+  #observed word counts
+  @observed = Array.new(@wid.size,0)
+  options[:wordsize].each{|ws| (0..seq.size-ws).each{|i| wid = @wid[seq[i, ws]]; @observed[wid] += 1 if not wid.nil?}}
+  
+  #expected word counts
+  @expected = Array.new(@wid.size) {Array.new(options[:shuffles],0).to_statarray}
+  us.init_shuffle(seq,bg)
+  options[:shuffles].times do |si|
+    seqsh = us.shuffle
+    options[:wordsize].each{|ws| (0..seq.size-ws).each{|i| wid = @wid[seqsh[i, ws]]; @expected[wid][si] += 1 if not wid.nil?}}
+  end
+  
+  #store results
+  @wid.each do |w,wid|
+   obs = @observed[wid]
+   File.open("#{dbdir}/#{w}.rnk", 'a') {|f| f.puts [id,obs,@expected[wid].select{|x| x>=obs}.size,@expected[wid].to_statarray.mean].join("\t")}
+  end
+  pbar.inc
+end
+pbar.finish

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 1
-  - 10
-  version: 0.1.10
+  - 11
+  version: 0.1.11
 platform: ruby
 authors: 
 - Anders Jacobsen
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-03-12 00:00:00 +01:00
+date: 2010-04-09 00:00:00 +02:00
 default_executable: 
 - bin/cwords
 dependencies: 
@@ -52,6 +52,7 @@ files:
 - resources/word_annotation.tsv
 - scripts/cwords.rb
 - scripts/cwords_mkdb.rb
+- scripts/cwords_mkdb_worker.rb
 - scripts/cluster_words.rb
 - scripts/complementary_words.rb
 has_rdoc: true