bio-gadget 0.4.8 → 0.5.0

data/lib/bio-gadget/femrg.rb

@@ -1,61 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    desc 'femrg [GTF...]', 'Extract and merge overlapping first exons.'
-
-    def femrg(*gtf_files)
-
-      exon1s = Hash.new
-      fp = open("| grep -h '\texon\t' #{gtf_files.nil? ? '-' : gtf_files.join(' ')} | cut -f 1,4,5,7,9")
-      fp.each { |line|
-        chr, sstart, sstop, str, attr = line.rstrip.split(/\t/)
-        exon1s[chr] = Hash.new unless exon1s.key?(chr)
-        exon1s[chr][str] = Hash.new unless exon1s[chr].key?(str)
-        id = attr.match(/transcript_id \"([^\"]+)\"/).to_a[1]
-        en = attr.match(/exon_number \"(\d+)\"/).to_a[1].to_i
-        if (!exon1s[chr][str].key?(id) ||
-            (str == "+" && en < exon1s[chr][str][id][2]) ||
-            (str == "-" && exon1s[chr][str][id][2] < en))
-          exon1s[chr][str][id] = [sstart.to_i, sstop.to_i, en]
-        end
-      }
-      fp.close
-
-      idx = 0
-      exon1s.each { |chr, exon1schr|
-        exon1schr.each { |str, exon1schrstr|
-          ids = exon1schrstr.keys.sort { |a, b|
-            exon1schrstr[a][0] <=> exon1schrstr[b][0]
-          }
-          #
-          clusters = Array.new
-          members = [id = ids.shift]
-          start, stop, en = exon1schrstr[id]
-          while ids.length > 0
-            nid = ids.shift
-            nstart, nstop, nen = exon1schrstr[nid]
-            if stop < nstart
-              clusters.push([members, start, stop])
-              members = [nid]
-              start = nstart
-              stop = nstop
-            else
-              members.push(nid)
-              start, stop = [start, stop, nstart, nstop].sort.values_at(0, 3)
-            end
-          end
-          clusters.push([members, start, stop])
-          #
-          clusters.each { |members, start, stop|
-            attr = "gene_id \"FE#{idx}\"; transcript_id \"FE#{idx}\"; "
-            puts [chr, 'bio-gadget:femrg', 'transcript', start, stop, 1000, str, '.', attr].join("\t")
-            puts [chr, 'bio-gadget:femrg', 'exon', start, stop, 1000, str, '.', "#{attr}exon_number \"1\"; member_ids \"#{members.join('|')}\""].join("\t")
-            idx += 1
-          }
-        }
-      }
-
-    end
-
-  end
-end

data/lib/bio-gadget/fqxz.rb

@@ -1,30 +0,0 @@
-require 'parallel'
-require 'pathname'
-
-module Bio
-  class Gadget < Thor
-
-    namespace :bio
-
-    desc 'fqxz', 'automatic (re)compression of *.fq(.gz|.bz2) files'
-    def fqxz
-      Parallel.map(Pathname.glob('*.fq{.gz,.bz2,}')) { |fqfilename|
-        xzfilename = fqfilename.sub(/\.fq(\.(gz|bz2))*$/, '.fq.xz')
-        if !xzfilename.exist?
-          case fqfilename.extname
-          when '.gz'
-            decompressor = 'gunzip -c'
-          when '.bz2'
-            decompressor = 'bunzip2 -c'
-          else
-            decompressor = 'cat'
-          end
-          puts "compressing #{xzfilename}..."
-          system "#{decompressor} #{fqfilename} | xz -z -e -c > #{xzfilename} 2> #{xzfilename}.log"
-          system "xz -t #{xzfilename} >> #{xzfilename}.log 2>&1"
-        end
-      }
-    end
-
-  end
-end

data/lib/bio-gadget/peak.rb

@@ -1,94 +0,0 @@
-require 'mkfifo'
-require 'parallel'
-
-module Bio
-  class Gadget < Thor
-
-    desc 'peak WIG1,WIG2,... [GTF]', <<DESC
-Find peak within each exon from (gzipped) variableStep wigs by a majority vote. It will read from a standard input if no GTF option.
-DESC
-    def peak(wigs, gtf="/dev/stdin")
-
-      nchrpos2val = Hash.new
-      wigs.split(/,/).each { |wig|
-        n = wigs.split(/,/).index(wig)
-        nchrpos2val[n] = Hash.new unless nchrpos2val.key?(n)
-        chr = ''
-        fp = open("| zcat #{wig} | grep -v ^track")
-        fp.each { |line|
-          cols = line.rstrip.split(/\s+/)
-          if cols[0] == 'variableStep'
-            chr = cols[1].match(/chrom=(\S+)$/).to_a[1]
-            nchrpos2val[n][chr] = Hash.new unless nchrpos2val[n].key?(chr)
-          else
-            nchrpos2val[n][chr][cols[0].to_i] = cols[1].to_f
-          end
-        }
-        fp.close
-      }
-
-      chr2exon = Hash.new
-      open("| grep exon #{gtf}").each { |line|
-        cols = line.rstrip.split(/\t/)
-        oid = cols[8].match(/oId \"([^\"]+)/).to_a[1]
-        oid = "#{cols[0]}.#{oid}" if cols[0] =~ /^RNA_SPIKE_/
-        exn = cols[8].match(/exon_number \"(\d+)\"/).to_a[1]
-        chr = cols[0]
-        str = cols[6]
-        start = cols[3].to_i
-        stop = cols[4].to_i
-        chr2exon[chr] = Array.new unless chr2exon.key?(chr)
-        chr2exon[chr].push([str, oid, exn, start, stop])
-      }
-
-      fifopath = mytemppath('fifo-')
-      File.mkfifo(fifopath)
-
-      pid = Kernel.fork {
-        exec "cat #{fifopath}"
-      }
-
-      fp = open(fifopath, 'w')
-      Parallel.each(chr2exon.keys,
-                    :in_processes => Parallel.processor_count) { |chr|
-        #
-        chr2exon[chr].each { |str, oid, exn, start, stop|
-          #
-          peak = ''
-          poss = Hash.new
-          nchrpos2val.each { |n, chrpos2val|
-            if chrpos2val.key?(chr)
-              pos2val = chrpos2val[chr]
-              tmppos2val = Hash.new
-              pos2val.each { |pos, val|
-                tmppos2val[pos] = val if start <= pos && pos <= stop
-              }
-              if tmppos2val.size > 0
-                tmpposs = tmppos2val.keys.sort { |a, b|
-                  tmppos2val[b] == tmppos2val[a] ? (str == '+' ? (a <=> b) : (b <=> a)) : (tmppos2val[b] <=> tmppos2val[a])
-                }
-                tmppos = tmpposs[0]
-                # puts "#{n} | #{chr}:#{start}-#{stop} #{str} | #{tmpposs}"
-                poss[tmppos] = poss.key?(tmppos) ? poss[tmppos]+1 : 1
-              end
-            end
-          }
-          if poss.size > 0
-            peaks = poss.keys.sort { |a, b|
-              poss[b] == poss[a] ? (str == '+' ? (a <=> b) : (b <=> a)) : (poss[b] <=> poss[a])
-            }
-            peak = peaks[0]
-          end
-          #
-          fp.syswrite([oid, exn, peak].join("\t") + "\n")
-        }
-        #
-      }
-      fp.close
-
-      Process.waitall
-
-    end
-
-  end
-end

data/lib/bio-gadget/qvstat.rb

@@ -1,34 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    namespace :bio
-
-    desc 'qvstat QUAL', 'statistics of quality values in *.qual file'
-    def qvstat(qualfile)
-      stat = Hash.new
-      myopen(qualfile) { |fp|
-        fp.each { |line|
-          next if /^[\#\>]/ =~ line
-          qvs = line.rstrip.split
-          qvs.each_index { |i|
-            qv = qvs[i]
-            stat[qv] = Array.new unless stat.key?(qv)
-            if stat[qv][i].nil?
-              stat[qv][i] = 1
-            else
-              stat[qv][i] += 1
-            end
-          }
-        }
-      }
-      statfile = qualfile.sub(/.qual(.gz)?$/, '.qvstat')
-      open(statfile, 'w') { |out|
-        qvs = stat.keys.sort { |a, b| a.to_i <=> b.to_i }
-        qvs.each { |qv|
-          out.puts "#{qv} #{stat[qv].join(' ')}"
-        }
-      }
-    end
-
-  end
-end

data/lib/bio-gadget/rgt2mtx.rb

@@ -1,60 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    desc 'rgt2mtx [RGT]', <<DESC
-Convert cuffdiff read group tracking file (*.read_group_tracking) into tab-separated matrix. If no given name of tracking file, it reads from standard input.
-DESC
-    option 'gtf', :aliases => '-g', :type => :string, :desc => 'GTF to revert old transcript_id (oId) renamed by cuffcompare. Moreover, "RNA_SPIKE_*" chromosome names will be inserted for the transcripts aligned on the forward strand of spike-in sequences.'
-    option 'sample', :aliases => '-s', :type => :string, :desc => 'Mapping from condition/replicate to sample ID for the column names of output matrix. Tab-separated text with three columns of condition, replicate and the sample ID.'
-    def rgt2mtx(rgt="/dev/stdin")
-
-      tid2oid = Hash.new
-      unless options['gtf'].nil?
-        open(options['gtf']).each { |line|
-          cols = line.rstrip.split(/\t/)
-          tid = cols[8].match(/transcript_id \"([^\"]+)/).to_a[1]
-          oid = cols[8].match(/oId \"([^\"]+)/).to_a[1]
-          oid = "#{cols[0]}.#{oid}" if cols[0] =~ /^RNA_SPIKE_/
-          tid2oid[tid] = oid
-        }
-      end
-
-      cr2sid = Hash.new
-      unless options['sample'].nil?
-        open(options['sample']).each { |line|
-          c, r, sid = line.rstrip.split(/\t/)
-          cr2sid["#{c}|#{r}"] = sid
-        }
-      end
-
-      id = nil
-      header = true
-      raws = Hash.new
-      open("| tail -n +2 #{rgt} | sort -k 1").each { |line|
-        cols = line.rstrip.split(/\t/)
-        id = cols[0] if id.nil?
-        if id != cols[0]
-          if header
-            if options['sample'].nil?
-              puts ([''] + raws.keys.sort).join("\t")
-            else
-              tmp = ['']
-              raws.keys.sort.each { |cr| tmp.push(cr2sid[cr]) }
-              puts tmp.join("\t")
-            end
-            header = false
-          end
-          tmp = [tid2oid.key?(id) ? tid2oid[id] : id]
-          raws.keys.sort.each { |k| tmp.push(raws[k]) }
-          puts tmp.join("\t")
-          id = cols[0]
-        end
-        raws["#{cols[1]}|#{cols[2]}"] = cols[3]
-      }
-      tmp = [tid2oid.key?(id) ? tid2oid[id] : id]
-      raws.keys.sort.each { |k| tmp.push(raws[k]) }
-      puts tmp.join("\t")
-    end
-
-  end
-end

data/lib/bio-gadget/version.rb

@@ -1,9 +0,0 @@
-require 'thor'
-
-module Bio
-  class Gadget < Thor
-
-    VERSION = "0.4.8"
-
-  end
-end

data/lib/bio-gadget/wig5p.rb

@@ -1,51 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    desc 'wig5p BAM', <<DESC
-Convert bam-format alignments into wig-format table. Count is at 5'-end of alignments. Moreover the counts are normalized when there are alignments to references named 'RNA_SPIKE_*'. This procedure requires samtools and bamToBed in BEDTools.
-DESC
-    option 'reverse', :aliases => '-r', :type => :boolean, :default => false
-    option 'name', :aliases => '-n', :type => :string, :default => ' '
-    def wig5p(bam)
-
-      str = options['reverse'] ? '-' : '+'
-
-      tmp = 0.0
-      open("| samtools view #{bam} | grep RNA_SPIKE_ | grep 'XS:A:+' | cut -f 1 | sort -u").each { |line| tmp += 1.0 }
-      abort "No spike-in reads." if tmp == 0.0
-      spike = 2234.8/tmp
-
-      acc2dups = Hash.new
-      open("| samtools view #{bam} | cut -f 1 | sort | uniq -c").each { |line|
-        dups, acc = line.strip.split(/\s/)
-        acc2dups[acc] = dups.to_f
-      }
-
-      cnts = Hash.new
-      tmpbam = mytemppath('.bam')
-      abort unless system("samtools view -h #{bam} | grep -E 'XS:A:\\#{str}|^@SQ' | samtools view -S -b - > #{tmpbam}")
-
-      open("| bamToBed -i #{tmpbam} | cut -f 1,#{options['reverse'] ? 3 : 2},4").each { |line|
-        chr, poss, acc = line.rstrip.split(/\t/)
-        cnts[chr] = Hash.new if !cnts.key?(chr)
-        pos = poss.to_i
-        if !cnts[chr].key?(pos)
-          cnts[chr][pos] = 1.0/acc2dups[acc]
-        else
-          cnts[chr][pos] = cnts[chr][pos]+1.0/acc2dups[acc]
-        end
-      }
-
-      puts "track type=wiggle_0 name=\"#{options['name']}\" description=\" \" alwaysZero=on visivility=full maxHeightPixels=128:64:16 color=#{options['reverse'] ? '0,128,255' : '255,128,0'}"
-
-      offset = options['reverse'] ? 0 : 1
-      signal = options['reverse'] ? -spike : spike
-      cnts.each { |chr, posvals|
-        puts "variableStep chrom=#{chr}"
-        posvals.keys.sort.each { |pos| puts [pos+offset, signal*posvals[pos]].join("\t") }
-      }
-
-    end
-
-  end
-end

data/lib/bio-gadget/wigchr.rb

@@ -1,28 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    namespace :bio
-
-    desc 'wigchr WIG CHR', 'extract wiggle track on specified chromosome'
-    def wigchr(wigfile, chr)
-      target = false
-      myopen(wigfile) { |fp|
-        fp.each { |line|
-          if (/^(fixed|variable)Step/ =~ line)
-            if (/chrom=#{chr}\s/ =~ line)
-              target = true
-              puts line
-            else
-              target = false
-            end
-          elsif (/^\d/ =~ line)
-            puts line if target
-          else
-            puts line
-          end
-        }
-      }
-    end
-
-  end
-end