X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=bp_bin%2Fassemble_seq_velvet;h=3b023b213caca2664ce071c075af00a396ba7f6e;hb=c59cc231c38eb62f8acd0bd604cadca057583802;hp=71fc5c129c9972320fdc3d8d79976eca32fcfc58;hpb=01cf17366cc275b6e388af00f1a981902345122f;p=biopieces.git diff --git a/bp_bin/assemble_seq_velvet b/bp_bin/assemble_seq_velvet index 71fc5c1..3b023b2 100755 --- a/bp_bin/assemble_seq_velvet +++ b/bp_bin/assemble_seq_velvet @@ -38,68 +38,33 @@ class Velvet @verbose = verbose end - def run_velveth(kmer_min, kmer_max, type) - @kmer_min = kmer_min - @kmer_max = kmer_max - - kmer = @kmer_min - - while kmer <= @kmer_max - dir_velveth = File.join(@directory, "Kmer_#{kmer}") - - Dir.mkdir(dir_velveth) - - commands = [] - commands << "velveth" - commands << dir_velveth - commands << kmer - commands << "-#{type}" - commands << "-fasta" - commands << @sequence_file - - execute(commands) - - kmer += 2 - end + def run_velveth(kmer_min, kmer_max, kmer_step, type) + commands = [] + commands << "velveth" + commands << @directory + commands << "#{kmer_min},#{kmer_max},#{kmer_step}" + commands << "-#{type}" + commands << "-fasta" + commands << @sequence_file + + execute(commands) end - def run_velvetg(cov_cutoffs, exp_cov) - Dir.glob("#{@directory}/Kmer_*").each do |dir_velveth| - files_velveth = Dir.glob("#{dir_velveth}/*") - - cov_cutoffs.each do |cov_cutoff| - dir_velvetg = File.join(dir_velveth, "Cov_cutoff_#{cov_cutoff}") + def run_velvetg(cov_cutoff, exp_cov, paired) + commands = [] + commands << "velvetg" + commands << @directory + commands << "-cov_cutoff #{cov_cutoff}" + commands << "-exp_cov #{exp_cov}" + commands << "-ins_length #{ins_length}" if paired + commands << "-clean yes" - Dir.mkdir(dir_velvetg) - FileUtils.cp_r files_velveth, dir_velvetg - - commands = [] - commands << "velvetg" - commands << dir_velvetg - commands << "-cov_cutoff #{cov_cutoff}" - commands << "-exp_cov #{exp_cov}" - commands << "-clean yes" - - execute(commands) - end - end - end - - def pick_best_assembly - list = [] - - Dir.glob("#{@directory}/Kmer_*/Cov_cutoff_*/contigs.fa").each do |file| - n50 = fasta_n50(file) - list << [file, n50] - end - - list.sort_by { |e| e.last }.last.first + execute(commands) end private def execute(commands) - commands.unshift "nice -n 19" commands.push "> /dev/null 2>&1" unless @verbose command = commands.join(" ") @@ -111,49 +76,26 @@ class Velvet $stderr.puts "Command failed: #{command}" end end - - def fasta_n50(file) - total = 0 - lengths = [] - count = 0 - n50 = 0 - - Fasta.open(file, mode="r") do |fasta_io| - fasta_io.each do |entry| - total += entry.length - lengths << entry.length - end - end - - lengths.sort.reverse.each do |length| - count += length - - if count >= total * 0.50 - n50 = length - break - end - end - - n50 - end end -types = 'short,shortPaired,long,longPaired' -cov_cutoffs = "2,4,8,16" +types = 'short,shortPaired,long,longPaired' casts = [] -casts << {:long=>'directory', :short=>'d', :type=>'dir', :mandatory=>true, :default=>nil, :allowed=>nil, :disallowed=>nil} -casts << {:long=>'type', :short=>'t', :type=>'string', :mandatory=>true, :default=>'short', :allowed=>types, :disallowed=>nil} -casts << {:long=>'kmer_min', :short=>'k', :type=>'uint', :mandatory=>true, :default=>19, :allowed=>nil, :disallowed=>nil} -casts << {:long=>'kmer_max', :short=>'K', :type=>'uint', :mandatory=>true, :default=>31, :allowed=>nil, :disallowed=>nil} -casts << {:long=>'cov_cutoff', :short=>'c', :type=>'list', :mandatory=>true, :default=>cov_cutoffs, :allowed=>nil, :disallowed=>nil} -casts << {:long=>'exp_cov', :short=>'e', :type=>'float', :mandatory=>true, :default=>'auto', :allowed=>nil, :disallowed=>nil} -casts << {:long=>'clean', :short=>'X', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'directory', :short=>'d', :type=>'dir', :mandatory=>true, :default=>nil, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'type', :short=>'t', :type=>'string', :mandatory=>true, :default=>'short', :allowed=>types, :disallowed=>nil} +casts << {:long=>'kmer_min', :short=>'k', :type=>'uint', :mandatory=>true, :default=>19, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'kmer_max', :short=>'K', :type=>'uint', :mandatory=>true, :default=>31, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'kmer_step', :short=>'s', :type=>'uint', :mandatory=>true, :default=>2, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'cov_cutoff', :short=>'c', :type=>'float', :mandatory=>true, :default=>'auto', :allowed=>nil, :disallowed=>nil} +casts << {:long=>'exp_cov', :short=>'e', :type=>'float', :mandatory=>true, :default=>'auto', :allowed=>nil, :disallowed=>nil} +casts << {:long=>'ins_length', :short=>'i', :type=>'uint', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil} +casts << {:long=>'clean', :short=>'X', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil} options = Biopieces.options_parse(ARGV, casts) raise ArgumentError, "kmer_min #{options[:kmer_min]} must be uneven." if options[:kmer_min].even? raise ArgumentError, "kmer_max #{options[:kmer_max]} must be uneven." if options[:kmer_max].even? +raise ArgumentError, "kmer_step #{options[:kmer_step]} must be even." unless options[:kmer_step].even? raise ArgumentError, "kmer_min >= kmer_max: #{options[:kmer_min]} >= #{options[:kmer_max]}" unless options[:kmer_max] > options[:kmer_min] Dir.mkdir(options[:directory]) unless Dir.exists?(options[:directory]) @@ -161,17 +103,20 @@ Dir.mkdir(options[:directory]) unless Dir.exists?(options[:directory]) file_fasta = File.join(options[:directory], "sequence_in.fna") Biopieces.open(options[:stream_in], options[:stream_out]) do |input, output| - Fasta.open(file_fasta, mode="w") do |fasta_io| - input.each_record do |record| - fasta_io.puts record - end + Fasta.open(file_fasta, mode="w") do |fasta_io| + input.each_record do |record| + if record[:SEQ_NAME] and record[:SEQ] + seq = Seq.new_bp(record) + fasta_io.puts seq.to_fasta + end + end end unless File.size(file_fasta) == 0 velvet = Velvet.new(options[:directory], file_fasta, options[:verbose]) - velvet.run_velveth(options[:kmer_min], options[:kmer_max], options[:type]) - velvet.run_velvetg(options[:cov_cutoff], options[:exp_cov]) - file_contigs = velvet.pick_best_assembly + velvet.run_velveth(options[:kmer_min], options[:kmer_max], options[:kmer_step], options[:type]) + velvet.run_velvetg(options[:cov_cutoff], options[:exp_cov], options[:type].match("Paired")) + file_contigs = File.join(options[:directory], "contigs.fa") Fasta.open(file_contigs, mode="r") do |fasta_io| fasta_io.each do |entry|