bio-gadget 0.3.1 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: aaeb43dbf9adba74b5741250dab6e1764421a4c4
-  data.tar.gz: f45ff460eeee4aee51b7888f39592dac24075898
+  metadata.gz: a8b08f5a256d452f6399fe4484d6a18811bc3b58
+  data.tar.gz: 532b1903529b32c549eb321e363d0dc8e8efaf27
 SHA512:
-  metadata.gz: 0f7dbebd237bb629a0e1d758b6e024427586d54a7cf4ccac59270b34e9c2dec374ee0d5bf81c88465eaabf531a520c712876349d8e72db3793fd9560c45cceee
-  data.tar.gz: 4d39fb3ee8a65513bbea18750e435b5f5d54166c1e3f92920f6905e83248c7e0a39361382660ca80253572b466b0458d60b6cf4e3891971f02e34ef83292bed8
+  metadata.gz: 23f150133d857f16a9ff0a326201fb044ec167d51fcf5addc23cb445ef219b75ff84038d457c3325fe076ce024b91d3afa2ffa33dcbba3c33401b9c75dfaf0bb
+  data.tar.gz: edbfc69fcfa8f7db2003236fc8d7b01046a051955907ad45bb2939096b389d6c3a6f19c79d95dbf3212ed320a20c6f9bf477a5f57e3ce9a542ccedd52ab4e4e5

data/README.org

@@ -8,20 +8,23 @@
 
 * Usage
 
-To check all commands in this package,
+To check all commands and the usages in this package
 
 : bio-gadget help
-: bio-gadget list
 
 Currently available commands are
 
-: bio-gadget dedup                                 # deduplicate fastq (via STDIN)
-: bio-gadget demlt BARCODE [FASTQ]                 # demultiplex fastq by barcodes
-: bio-gadget ensann GENENAMEGZ [ENSGENE]           # create ensGene.gtf
-: bio-gadget fqxz                                  # automatic (re)compression of *.fq(.gz|.bz2) files
-: bio-gadget qvstat QUAL                           # statistics of quality values in *.qual file
-: bio-gadget ucscann ISOFORMSGZ XREFGZ [KNOWNGENE] # create knownGene.gtf
-: bio-gadget wigchr WIG CHR                        # extract wiggle track on specified chromosome
+- dedup :: Deduplicate fastq (via STDIN)
+- demlt :: Demultiplex fastq by barcodes
+- fqxz :: automatic (re)compression of *.fq(.gz|.bz2) files
+- qvstat :: Statistics of quality values in *.qual file
+- rgt2mtx :: Convert cuffdiff read group tracking file into tab-separated matrix
+- wig5p :: Convert bam-format alignments into wig-format table
+- wigchr :: Extract wiggle track on specified chromosome
+
+** Memo                                                            :noexport:
+- gtfensembl :: Create gtf of ENSEMBL Genes (a.k.a. ensGene.gtf) from XX files of UCSC annotation database; .txt.gz of 
+- gtfucsc :: Create gtf of UCSC Genes from four (knownIsoforms.txt.gz, kgXref.txt.gz, kgTxInfo.txt.gz and knownGene.txt.gz) files of UCSC annotation database. The output gtf contains gene_id, transcript_id and gene_name attributes, so it convenients to use as a gene/transcript annotation file for tophat/cufflinks etc.
 
 * Contributing
 

data/lib/bio-gadget.rb

@@ -2,8 +2,9 @@ require 'bio-gadget/version'
 require 'bio-gadget/dedup'
 require 'bio-gadget/demlt'
 require 'bio-gadget/fqxz'
-require 'bio-gadget/gtfann'
 require 'bio-gadget/qvstat'
+require 'bio-gadget/rgt2mtx'
+require 'bio-gadget/wig5p'
 require 'bio-gadget/wigchr'
 
 require 'tempfile'
@@ -11,8 +12,6 @@ require 'tempfile'
 module Bio
   class Gadget < Thor
 
-    namespace :bio
-
     private
 
     def myopen(file, &block)
@@ -38,5 +37,6 @@ module Bio
     END {
       @@mytemppaths.each { |path| File.unlink(path) if File.exist?(path) }
     }
+
   end
 end

data/lib/bio-gadget/rgt2mtx.rb

@@ -0,0 +1,33 @@
+module Bio
+  class Gadget < Thor
+
+    desc 'rgt2mtx [RGT]', <<DESC
+Convert cuffdiff read group tracking file into tab-separated matrix. If no given name of tracking file, it reads from standard input.
+DESC
+    def rgt2mtx(rgt="/dev/stdin")
+
+      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
+            puts (['tracking_id'] + raws.keys.sort).join("\t")
+            header = false
+          end
+          tmp = [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 = [id]
+      raws.keys.sort.each { |k| tmp.push(raws[k]) }
+      puts tmp.join("\t")
+    end
+
+  end
+end

data/lib/bio-gadget/version.rb

@@ -3,7 +3,7 @@ require 'thor'
 module Bio
   class Gadget < Thor
 
-    VERSION = "0.3.1"
+    VERSION = "0.4.0"
 
   end
 end

data/lib/bio-gadget/wig5p.rb

@@ -0,0 +1,51 @@
+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

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bio-gadget
 version: !ruby/object:Gem::Version
-  version: 0.3.1
+  version: 0.4.0
 platform: ruby
 authors:
 - Shintaro Katayama
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-03-17 00:00:00.000000000 Z
+date: 2013-04-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thor
@@ -100,9 +100,10 @@ files:
 - lib/bio-gadget/dedup.rb
 - lib/bio-gadget/demlt.rb
 - lib/bio-gadget/fqxz.rb
-- lib/bio-gadget/gtfann.rb
 - lib/bio-gadget/qvstat.rb
+- lib/bio-gadget/rgt2mtx.rb
 - lib/bio-gadget/version.rb
+- lib/bio-gadget/wig5p.rb
 - lib/bio-gadget/wigchr.rb
 homepage: https://github.com/shka/ruby-bio-gadget
 licenses: []

data/lib/bio-gadget/gtfann.rb

@@ -1,93 +0,0 @@
-module Bio
-  class Gadget < Thor
-
-    desc 'ucscann ISOFORMSGZ XREFGZ [KNOWNGENE]', 'create knownGene.gtf'
-    def ucscann(isoforms, xref, knowngene='/dev/stdin')
-
-      acc2id = Hash.new
-      open("| gunzip -c #{isoforms}").each { |line|
-        id, acc = line.rstrip.split(/\t/)
-        acc2id[acc] = id
-      }
-
-      acc2sym = Hash.new
-      open("| gunzip -c #{xref} | cut -f 1,5").each { |line|
-        acc, sym = line.rstrip.split(/\t/)
-        acc2sym[acc] = sym
-      }
-
-      open("| sort -k1,1 -k4,4n", 'w') { |fp|
-        open(knowngene).each { |line|
-          cols = line.rstrip.split(/\t/)
-          acc, chr, str = cols.values_at(0, 1, 2)
-          cs = cols[5].to_i
-          ce = cols[6].to_i
-          lefts = cols[8].split(/,/)
-          rights = cols[9].split(/,/)
-          prop = "gene_id \"#{acc2id[acc]}\"; transcript_id \"#{acc}\""
-          prop += "; gene_name \"#{acc2sym[acc]}\"" if acc2sym.has_key?(acc)
-          lefts.each_index { |i|
-            next if lefts[i].nil?
-            l = lefts[i].to_i
-            r = rights[i].to_i
-            fp.puts [chr, 'knownGene', 'exon', l+1, r, 0, str, '.', prop].join("\t")
-          }
-          if cs != ce
-            fp.puts [chr, 'knownGene', 'start_codon', (str == '+' ? cs+1 : ce-2), (str == '+' ? cs+3 : ce), 0, str, '.', prop].join("\t")
-            lefts.each_index { |i|
-              next if lefts[i].nil?
-              l = lefts[i].to_i
-              next if ce-1 < l
-              r = rights[i].to_i
-              next if r-1 < cs
-              fp.puts [chr, 'knownGene', 'CDS', (cs < l ? l : cs), (r < ce ? r : ce), 0, str, '.', prop].join("\t")
-            }
-          end
-        }
-      }
-
-    end
-
-    desc 'ensann GENENAMEGZ [ENSGENE]', 'create ensGene.gtf'
-    def ensann(genename, ensgene='/dev/stdin')
-
-      acc2sym = Hash.new
-      open("| gunzip -c #{genename}").each { |line|
-        acc, sym = line.rstrip.split(/\t/)
-        acc2sym[acc] = sym
-      }
-
-      open("| sort -k1,1 -k4,4n", 'w') { |fp|
-        open(ensgene).each { |line|
-          cols = line.rstrip.split(/\t/)
-          acc, chr, str, id = cols.values_at(1, 2, 3, 12)
-          cs = cols[6].to_i
-          ce = cols[7].to_i
-          lefts = cols[9].split(/,/)
-          rights = cols[10].split(/,/)
-          prop = "gene_id \"#{id}\"; transcript_id \"#{acc}\""
-          prop += "; gene_name \"#{acc2sym[acc]}\"" if acc2sym.has_key?(acc)
-          lefts.each_index { |i|
-            next if lefts[i].nil?
-            l = lefts[i].to_i
-            r = rights[i].to_i
-            fp.puts [chr, 'ensGene', 'exon', l+1, r, 0, str, '.', prop].join("\t")
-          }
-          if cs != ce
-            fp.puts [chr, 'ensGene', 'start_codon', (str == '+' ? cs+1 : ce-2), (str == '+' ? cs+3 : ce), 0, str, '.', prop].join("\t")
-            lefts.each_index { |i|
-              next if lefts[i].nil?
-              l = lefts[i].to_i
-              next if ce-1 < l
-              r = rights[i].to_i
-              next if r-1 < cs
-              fp.puts [chr, 'ensGene', 'CDS', (cs < l ? l : cs), (r < ce ? r : ce), 0, str, '.', prop].join("\t")
-            }
-          end
-        }
-      }
-
-    end
-
-  end
-end