rbbt-sources 0.4.0 → 1.0.0

data/etc/biomart/missing_in_archive

@@ -0,0 +1,15 @@
+may2009:
+    - agilent_wholegenome
+    - agilent_cgh_44b
+    - illumina_humanwg_6_v2
+    - illumina_humanwg_6_v3
+dec2007:
+    - protein_id
+    - affy_hc_g110
+    - affy_hg_u133a_2
+    - affy_huex_1_0_st_v2
+    - affy_hugene_1_0_st_v1
+    - agilent_wholegenome
+    - agilent_cgh_44b
+    - illumina_humanwg_6_v2
+    - illumina_humanwg_6_v3

data/lib/rbbt/sources/COSMIC.rb

@@ -0,0 +1,14 @@
+require 'rbbt'
+require 'rbbt/resource'
+module COSMIC
+  extend Resource
+  self.subdir = "share/databases/COSMIC"
+
+  COSMIC.claim COSMIC.Mutations, :proc do 
+    url = "ftp://ftp.sanger.ac.uk/pub/CGP/cosmic/data_export/CosmicMutantExport_v54_120711.tsv"
+
+    TSV.open(Open.open(url), :header_hash => "", :key_field => "Mutation GRCh37 genome position", :merge => true).to_s
+  end
+end
+
+puts COSMIC.Mutations.produce

data/lib/rbbt/sources/COSTART.rb

@@ -2,7 +2,7 @@ require 'rbbt-util'
 
 module COSTART
 
-  Rbbt.share.databases.COSTART.COSTART.define_as_proc do 
+  Rbbt.claim Rbbt.share.databases.COSTART.COSTART, :proc do 
       terms = ["#COSTART Terms"]
       Open.open('http://hedwig.mgh.harvard.edu/biostatistics/files/costart.html').lines.each do |line|
         puts line

data/lib/rbbt/sources/CTCAE.rb

@@ -2,5 +2,5 @@ require 'rbbt-util'
 require 'rbbt/util/excel2tsv'
 
 module CTCAE
-  Rbbt.share.CTCAE.CTCAE.define_as_url TSV.excel2tsv('http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14.xls')
+  Rbbt.claim Rbbt.share.databases.CTCAE.CTCAE, :proc do TSV.excel2tsv('http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14.xls').to_s end
 end

data/lib/rbbt/sources/InterPro.rb

@@ -0,0 +1,17 @@
+require 'rbbt-util'
+module InterPro
+  extend Resource
+  self.subdir = "share/databases/InterPro"
+
+  InterPro.claim InterPro.root.find, :rake, Rbbt.share.install.InterPro.Rakefile.find(:lib)
+
+  def self.tsv(*args)
+    old_url = BioMart::BIOMART_URL
+    begin
+      BioMart::BIOMART_URL.replace "http://www.ebi.ac.uk/interpro/biomart/martservice?query="
+      BioMart.tsv(*args)
+    ensure
+      BioMart::BIOMART_URL.replace old_url
+    end
+  end
+end

data/lib/rbbt/sources/NCI.rb

@@ -0,0 +1,7 @@
+require 'rbbt-util'
+module NCI
+  extend Resource
+  self.subdir = "share/databases/NCI"
+
+  NCI.claim NCI.root.find, :rake, Rbbt.share.install.NCI.Rakefile.find(:lib)
+end

data/lib/rbbt/sources/biomart.rb

@@ -1,5 +1,6 @@
 require 'rbbt'
-require 'rbbt/util/tsv'
+require 'rbbt/tsv'
+require 'rbbt/tsv/attach'
 require 'rbbt/util/log'
 require 'cgi'
 
@@ -14,7 +15,7 @@ module BioMart
 
   BIOMART_URL = 'http://biomart.org/biomart/martservice?query='
 
-  MISSING_IN_ARCHIVE = Rbbt.etc.biomart.missing_in_archive.yaml
+  MISSING_IN_ARCHIVE = Rbbt.etc.biomart.missing_in_archive.exists? ? Rbbt.etc.biomart.missing_in_archive.yaml : {}
 
   private
 
@@ -68,10 +69,10 @@ module BioMart
 
     new_datafile = TmpFile.tmp_file
     if data.nil?
-      TSV.merge_rows Open.open(result_file), new_datafile
+      TSV.merge_row_fields Open.open(result_file), new_datafile
       data = new_datafile
     else
-      TSV.paste_merge data, result_file, new_datafile
+      TSV.merge_different_fields data, result_file, new_datafile
       FileUtils.rm data
       data = new_datafile
     end
@@ -117,7 +118,6 @@ module BioMart
     }
 
     chunks << chunk if chunk.any?
-    
 
     Log.low "Chunks: #{chunks.length}"
     chunks.each_with_index{|chunk,i|
@@ -125,15 +125,15 @@ module BioMart
       data = get(database, main, chunk, filters, data, open_options)
     }
 
-    open_options[:filename] ||= "BioMart: '#{main}' [#{(attrs || []) * ', '}] [#{(filters || []) * ', '}" 
+    open_options[:filename] ||= "BioMart[#{main}+#{attrs.length}]"
     if filename.nil?
-      results = TSV.new data, open_options
+      results = TSV.open data, open_options
       results.key_field = main
       results.fields = attrs
       results
     else
       Open.write(filename) do |f|
-        f.puts "#: " << Misc.hash2string(TSV::EXTRA_ACCESSORS.collect{|key| [key, open_options[key]]})
+        f.puts "#: " << Misc.hash2string(TSV::ENTRIES.collect{|key| [key, open_options[key]]})
         if field_names.nil?
           f.puts "#" << [main, attrs].flatten * "\t"
         else
@@ -148,7 +148,7 @@ module BioMart
 
   def self.tsv(database, main, attrs = nil, filters = nil, data = nil, open_options = {})
     if @archive_url 
-      attrs = attrs.reject{|attr| MISSING_IN_ARCHIVE[@archive].include? attr[1]}
+      attrs = attrs.reject{|attr| (MISSING_IN_ARCHIVE[@archive] || []).include? attr[1]}
     end
 
     codes = attrs.collect{|attr| attr[1]}

data/lib/rbbt/sources/entrez.rb

@@ -1,18 +1,19 @@
-require 'rbbt-util'
-require 'rbbt/util/tsv'
+require 'rbbt'
+require 'rbbt/tsv'
+require 'rbbt/resource'
 require 'rbbt/bow/bow'
 require 'set'
 
 module Entrez
 
-  Rbbt.share.databases.entrez.gene_info.define_as_url 'ftp://ftp.ncbi.nih.gov/gene/DATA/gene_info.gz'
-  Rbbt.share.databases.entrez.gene2pubmed.define_as_url 'ftp://ftp.ncbi.nih.gov/gene/DATA/gene2pubmed.gz'
+  Rbbt.claim Rbbt.share.databases.entrez.gene_info, :url, 'ftp://ftp.ncbi.nih.gov/gene/DATA/gene_info.gz'
+  Rbbt.claim Rbbt.share.databases.entrez.gene2pubmed, :url, 'ftp://ftp.ncbi.nih.gov/gene/DATA/gene2pubmed.gz'
 
   def self.entrez2native(taxs, options = {})
-    options = Misc.add_defaults options, :key => 1, :fields => 5, :persistence => true, :merge => true
+    options = Misc.add_defaults options, :key_field => 1, :fields => 5, :persist => true, :merge => true
 
     taxs = [taxs] unless Array === taxs
-    options.merge! :grep => taxs.collect{|t| "^#{ t }\t"}
+    options.merge! :grep => taxs.collect{|t| "^" + t.to_s}
     
     tsv = Rbbt.share.databases.entrez.gene_info.tsv :flat, options
     tsv.key_field = "Entrez Gene ID"
@@ -20,12 +21,24 @@ module Entrez
     tsv
   end
 
+  def self.entrez2name(taxs, options = {})
+    options = Misc.add_defaults options, :key_field => 1, :fields => 2, :persist => true, :merge => true
+
+    taxs = [taxs] unless Array === taxs
+    options.merge! :grep => taxs.collect{|t| "^" + t.to_s}
+    
+    tsv = Rbbt.share.databases.entrez.gene_info.tsv :flat, options
+    tsv.key_field = "Entrez Gene ID"
+    tsv.fields    = ["Associated Gene Name"]
+    tsv
+  end
+
+
   def self.entrez2pubmed(taxs)
-    options = {:key => 1, :fields => 2, :persistence => true, :merge => true}
+    options = {:key_field => 1, :fields => 2, :persist => true, :merge => true}
 
     taxs = [taxs] unless taxs.is_a?(Array)
-    taxs = taxs.collect{|t| t.to_s}
-    options.merge! :grep => taxs.collect{|t| "^#{ t }\t"}
+    options.merge! :grep => taxs.collect{|t| "^" + t.to_s}
 
     Rbbt.share.databases.entrez.gene2pubmed.tsv :flat, options
   end
@@ -58,18 +71,31 @@ module Entrez
   private 
 
   def self.get_online(geneids)
-    geneids_list = ( geneids.is_a?(Array) ? geneids.join(',') : geneids.to_s )
-    url = "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=gene&retmode=xml&id=#{geneids_list}" 
 
-    xml = Open.read(url, :wget_options => {:quiet => true}, :nocache => true)
+    genes_complete =  geneids.is_a?(Array) ? geneids : [geneids]
 
-    genes = xml.scan(/(<Entrezgene>.*?<\/Entrezgene>)/sm).flatten
+    genes = []
+    Misc.divide(genes_complete, (genes_complete.length / 100) + 1).each do |geneids_list|
+      begin
+        Misc.try3times do
+          url = "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=gene&retmode=xml&id=#{geneids_list * ","}" 
+
+          xml = Open.read(url, :wget_options => {:quiet => true}, :nocache => true)
+
+          genes += xml.scan(/(<Entrezgene>.*?<\/Entrezgene>)/sm).flatten
+        end
+      rescue
+        puts $!.message
+        genes += geneids_list.collect{|g| nil}
+      end
+    end
 
     if geneids.is_a? Array
-      list = {}
-      genes.each_with_index{|gene,i|
-        geneid = geneids[i]
-        list[geneid ] = gene
+      list = Hash[*genes_complete.zip([nil]).flatten]
+      genes.each{|gene|
+        geneid = gene.match(/<Gene-track_geneid>(\d+)/)[1]
+        geneid = geneid.to_i unless list.include? geneid
+        list[geneid] = gene
       }
       return list
     else
@@ -99,6 +125,7 @@ module Entrez
         end
       }
 
+
       return list unless missing.any?
       genes = get_online(missing)
 

data/lib/rbbt/sources/go.rb

@@ -1,22 +1,24 @@
-require 'rbbt-util'
+require 'rbbt'
+require 'rbbt/resource'
+require 'rbbt/persist/tsv'
 
 # This module holds helper methods to deal with the Gene Ontology files. Right
 # now all it does is provide a translation form id to the actual names.
 module GO
 
-  Rbbt.share.databases.GO.gene_ontology.define_as_url 'ftp://ftp.geneontology.org/pub/go/ontology/gene_ontology.obo'
-  Rbbt.share.databases.GO.gslim_generic.define_as_url 'http://www.geneontology.org/GO_slims/goslim_generic.obo'
+  Rbbt.claim Rbbt.share.databases.GO.gene_ontology, :url, 'ftp://ftp.geneontology.org/pub/go/ontology/gene_ontology.obo'
+  Rbbt.claim Rbbt.share.databases.GO.gslim_generic, :url, 'http://www.geneontology.org/GO_slims/goslim_generic.obo'
 
   MULTIPLE_VALUE_FIELDS = %w(is_a)
-  TSV_GENE_ONTOLOGY = File.join(Persistence.cachedir, 'gene_ontology')
+  TSV_GENE_ONTOLOGY = File.join(Persist.cachedir, 'gene_ontology')
 
   # This method needs to be called before any translations can be made, it is
   # called automatically the first time the id2name method is called. It loads
   # the gene_ontology.obo file and extracts all the fields, although right now,
   # only the name field is used.
   def self.init
-    init = Persistence.persist_tsv('gene_ontology', :Misc) do 
-      info = {}
+    Persist.persist_tsv(nil, 'gene_ontology', {}, :persist => true) do |info|
+      info.serializer = :marshal if info.respond_to? :serializer and info.serializer == :type
       Rbbt.share.databases.GO.gene_ontology.read.split(/\[Term\]/).each{|term| 
         term_info = {}
 
@@ -33,12 +35,13 @@ module GO
         next if term_info["id"].nil?
         info[term_info["id"]] = term_info
       }
+
       info
     end
   end
 
   def self.info
-    self.init
+    @info ||= self.init
   end
 
   def self.goterms

data/lib/rbbt/sources/jochem.rb

@@ -0,0 +1,4 @@
+require 'rbbt-util'
+module JoChem
+  Rbbt.claim Rbbt.share.databases.JoChem, :rake, Rbbt.share.install.JoChem.Rakefile.find 
+end

data/lib/rbbt/sources/organism.rb

@@ -1,32 +1,35 @@
-require 'rbbt-util'
-require 'rbbt/util/resource'
-
+require 'rbbt'
+require 'rbbt/resource'
+require 'rbbt/resource/with_key'
 
 module Organism
   extend Resource
-  relative_to Rbbt, "share/organisms"
+  self.pkgdir = "rbbt"
+  self.subdir = "share/organisms"
 
-  class OrganismNotProcessedError < StandardError; end
+  ["Hsa", "Rno", "Sce"].each do |organism|
+    claim Organism[organism], :rake, Rbbt.share.install.Organism[organism].Rakefile.find
 
-  def self.datadir(org)
-    File.join(Rbbt.datadir, 'organisms', org)
-  end 
+    module_eval "#{ organism } = with_key '#{organism}'"
+  end
+
+  class OrganismNotProcessedError < StandardError; end
 
   def self.attach_translations(org, tsv, target = nil, fields = nil, options = {})
     Log.high "Attaching Translations for #{ org.inspect }, target #{target.inspect}, fields #{fields.inspect}"
-    options = Misc.add_defaults options, :persistence => true, :case_insensitive => false
+    options = Misc.add_defaults options, :persist => true, :case_insensitive => false
 
-    options.merge! :key    => target unless target.nil?
+    options.merge! :key_field    => target unless target.nil?
     options.merge! :fields => fields unless fields.nil?
 
     index = identifiers(org).tsv options
 
-    tsv.attach index, [:key]
+    tsv.attach index, :fields => [:key], :persist_input => true
   end
 
   def self.normalize(org, list, target = nil, fields = nil, options = {})
     return [] if list.nil? or list.empty?
-    options = Misc.add_defaults options, :persistence => true, :case_insensitive => true, :double => false
+    options = Misc.add_defaults options, :persist => true, :case_insensitive => true, :double => false
     double = Misc.process_options options, :double
 
 
@@ -50,14 +53,20 @@ module Organism
     end
   end
 
-  def self.guess_id(org, values)
-    identifiers = TSV.new(Organism.identifiers(org), :persistence => true)
+  def self.guess_id(org, values, identifiers = nil)
+    identifiers ||= TSV.setup(Organism.identifiers(org), :persist => true)
     field_matches = identifiers.field_matches(values)
     field_matches.sort_by{|field, matches| matches.uniq.length}.last
   end
 
+  def self.guess_id(org, values)
+    field_matches = TSV.field_match_counts(Organism.identifiers(org).find, values)
+    field_matches.sort_by{|field, count| count.to_i}.last
+  end
+
+
   def self.organisms
-    Dir.glob(File.join(Rbbt.share.organisms.find, '*')).collect{|f| File.basename(f)}
+    Dir.glob(File.join(Organism.root.find, '*')).collect{|f| File.basename(f)}
   end
 
   def self.name(organism)
@@ -70,14 +79,4 @@ module Organism
     }.first
   end
 
-  ["Hsa", "Rno", "Sce"].each do |organism|
-    rakefile = Rbbt["share/install/Organism/#{ organism }/Rakefile"]
-    rakefile.lib_dir = Resource.caller_lib_dir __FILE__
-    rakefile.pkgdir = 'phgx'
-    Organism[organism].define_as_rake rakefile
-    module_eval "#{ organism } = with_key '#{organism}'"
-  end
-
 end
-
-

data/lib/rbbt/sources/organism/sequence.rb

@@ -11,7 +11,12 @@ module Organism
     exon_transcripts ||= Organism.transcript_exons(org).tsv(:double, :key => "Ensembl Exon ID", :fields => ["Ensembl Transcript ID"], :merge => true, :persistence => true )
     transcript_info  ||= Organism.transcripts.tsv(org).tsv(:list, :persistence => true )
 
-    transcripts = exon_transcripts[exon].first
+    transcripts = begin
+                    exon_transcripts[exon].first
+                  rescue
+                    []
+                  end
+
     transcripts.select{|transcript| transcript_info[transcript]["Ensembl Protein ID"].any?}
   end
 
@@ -156,6 +161,8 @@ module Organism
     transcript_offsets = {}
     exons.each do |exon|
       transcript_offsets[exon] ||= {}
+      offsets = nil
+      next unless exon_offsets.include? exon
       offsets = exon_offsets[exon].zip_fields
 
       offsets.collect do |transcript, offset|
@@ -173,7 +180,7 @@ module Organism
     exon_end     ||= Organism.exons(org).tsv(:single, :persistence => true, :fields => ["Exon Chr End"], :cast => :to_i)
     exon_strand  ||= Organism.exons(org).tsv(:single, :persistence => true, :fields => ["Exon Strand"], :cast => :to_i)
 
-    exons = exons_at_genomic_positions(org, positions)
+    exons          = exons_at_genomic_positions(org, positions)
     offsets        = Organism.exon_transcript_offsets(org, exons.flatten.uniq, exon_offsets, exon_info)
 
     position_exons = {}
@@ -203,8 +210,135 @@ module Organism
     position_offsets
   end
 
+  def self.exon_junctures_at_chromosome_positions(org, chromosome, positions)
+    chromosome = chromosome.to_s
+    chromosome_start = Persistence.persist(Organism.exons(org), "Exon_start[#{chromosome}]", :fwt, :chromosome => chromosome, :range => false) do |file, options|
+      tsv = file.tsv(:persistence => true, :type => :list)
+      tsv.select("Chromosome Name" => chromosome).collect do |exon, values|
+        [exon, values["Exon Chr Start"].to_i]
+      end
+    end
+    chromosome_end = Persistence.persist(Organism.exons(org), "Exon_end[#{chromosome}]", :fwt, :chromosome => chromosome, :range => false) do |file, options|
+      tsv = file.tsv(:persistence => true, :type => :list)
+      tsv.select("Chromosome Name" => chromosome).collect do |exon, values|
+        [exon, values["Exon Chr End"].to_i]
+      end
+    end
+
+    if Array === positions
+      positions.collect{|position| 
+        position = position.to_i
+        chromosome_start[(position - 2)..(position + 2)] + chromosome_end[(position - 2)..(position + 2)];
+      }
+    else
+      position = positions.to_i
+      chromosome_start[(position - 2)..(position + 2)] + chromosome_end[(position - 2)..(position + 2)];
+    end
+ 
+  end
+
+  def self.exon_junctures_at_genomic_positions(org, positions)
+    positions = [positions] unless Array === positions.first
+
+    exons = []
+    chromosomes = {}
+    indices     = {}
+    positions.each_with_index do |info,i|
+      chr, pos = info
+      chromosomes[chr] ||= []
+      indices[chr] ||= []
+      chromosomes[chr] << pos
+      indices[chr] << i
+    end
+
+    chromosomes.each do |chr, pos_list|
+      chr_exons = exon_junctures_at_chromosome_positions(org, chr, pos_list)
+      chr_exons.zip(indices[chr]).each do |exon, index| exons[index] = exon end
+    end
+
+    exons
+  end
+
+  def self.identify_variations_at_chromosome_positions(org, chromosome, positions, variations)
+    chromosome = chromosome.to_s
+
+    chromosome_bed = Persistence.persist(variations, "Variation_positions[#{chromosome}]", :fwt, :chromosome => chromosome, :range => false) do |file, options|
+      rows = []
+      chromosome = options[:chromosome]
+      f = CMD.cmd("grep '[[:space:]]#{chromosome}[[:space:]]' #{ file }", :pipe => true)
+      while not f.eof?
+        line = f.gets.chomp
+        id, chr, pos = line.split "\t"
+        rows << [id, pos.to_i]
+      end
+
+      rows
+    end
+
+    if Array === positions
+      positions.collect{|position| 
+        chromosome_bed[position]; 
+      }
+    else
+      chromosome_bed[positions];  
+    end
+  end
+
+
+  def self.identify_variations_at_genomic_positions(org, positions, variations_file)
+    positions = [positions] unless Array === positions.first
+
+    variations = []
+    chromosomes = {}
+    indices     = {}
+    positions.each_with_index do |info,i|
+      chr, pos = info
+      chromosomes[chr] ||= []
+      indices[chr] ||= []
+      chromosomes[chr] << pos
+      indices[chr] << i
+    end
+
+    chromosomes.each do |chr, pos_list|
+      chr_variations = identify_variations_at_chromosome_positions(org, chr, pos_list, variations_file)
+      chr_variations.zip(indices[chr]).each do |variation, index| variations[index] = variation end
+    end
+
+    variations
+  end
+
+  task_option :organism, "Organism", :string, "Hsa"
+  task_option :genomic_mutations, "Position (chr:position)\tMutation", :tsv
+  task_dependencies nil
+  task :genomic_mutations_in_exon_junctions => :tsv do |org,genomic_mutations|
+    genomic_mutations = case
+                        when TSV === genomic_mutations
+                          genomic_mutations
+                        else
+                          TSV.new StringIO.new(genomic_mutations), :list
+                        end
+    genomic_mutations.key_field ||= "Position"
+    genomic_mutations.fields    ||= ["Mutation"]
+
+    positions = genomic_mutations.keys.collect{|l| l.split(":")}
+
+    step(:resources, "Load Resources")
+
+    exon_junctures = {}
+    genomic_mutations.keys.zip(Organism.exon_junctures_at_genomic_positions(org, positions)).each do |position, exons|
+      exon_junctures[position] = exons 
+    end
+
+    genomic_mutations.add_field "Exon Junctions" do |position, values|
+      exon_junctures[position] * "|"
+    end
+
+    genomic_mutations.to_s :sort, true
+  end
+
+
   task_option :organism, "Organism", :string, "Hsa"
-  task_option :genomic_mutations, "Position (chr:position), Allele", :tsv
+  task_option :genomic_mutations, "Position (chr:position)\tMutation", :tsv
   task_dependencies nil
   task :genomic_mutations_to_genes => :tsv do |org,genomic_mutations|
     genomic_mutations = case
@@ -213,8 +347,8 @@ module Organism
                         else
                           TSV.new StringIO.new(genomic_mutations), :list
                         end
-    genomic_mutations.key_field = "Position"
-    genomic_mutations.fields = ["Mutation"]
+    genomic_mutations.key_field ||= "Position"
+    genomic_mutations.fields    ||= ["Mutation"]
 
     positions = genomic_mutations.keys.collect{|l| l.split(":")}
 
@@ -234,7 +368,7 @@ Translates a collection of mutations in genomic coordinates into mutations in am
 protein products of transcripts including those positions.
   EOF
   task_option :organism, "Organism", :string, "Hsa"
-  task_option :genomic_mutations, "Position (chr:position), Allele", :tsv
+  task_option :genomic_mutations, "Position (chr:position)\tMutation", :tsv
   task_dependencies nil
   task :genomic_mutations_to_protein_mutations => :tsv do |org,genomic_mutations|
     genomic_mutations = case
@@ -244,8 +378,8 @@ protein products of transcripts including those positions.
                           TSV.new StringIO.new(genomic_mutations), :list
                         end
 
-    genomic_mutations.key_field = "Position"
-    genomic_mutations.fields = ["Mutation"]
+    genomic_mutations.key_field ||= "Position"
+    genomic_mutations.fields    ||= ["Mutation"]
 
     positions = genomic_mutations.keys.collect{|l| l.split(":")}
 
@@ -256,7 +390,6 @@ protein products of transcripts including those positions.
     results.type = :double
     results.filename = path
 
-
     step(:resources, "Load Resources")
     transcript_sequence = Organism.transcript_sequence(org).tsv(:single, :persistence => true)
     transcript_5utr     = Organism.transcript_5utr(org).tsv(:single, :persistence => true, :cast => 'to_i')
@@ -264,26 +397,31 @@ protein products of transcripts including those positions.
     exon_start          = Organism.exons(org).tsv(:single, :persistence => true, :fields => ["Exon Chr Start"], :cast => :to_i)
     exon_end            = Organism.exons(org).tsv(:single, :persistence => true, :fields => ["Exon Chr End"], :cast => :to_i)
     exon_strand         = Organism.exons(org).tsv(:single, :persistence => true, :fields => ["Exon Strand"], :cast => :to_i)
+    transcript_to_protein = Organism.transcripts(org).tsv(:single, :fields => "Ensembl Protein ID", :persistence => true)
 
     step(:offsets, "Find transcripts and offsets for mutations")
     offsets = Organism.genomic_position_transcript_offsets(org, positions, exon_offsets, exon_start, exon_end, exon_strand)
 
     step(:aminoacid, "Translate mutation to amino acid substitutions")
     offsets.each do |position, transcripts|
-      alleles = genomic_mutations[position * ":"].collect{|allele| Misc.IUPAC_to_base(allele)}.flatten
+      if genomic_mutations.type === :double
+        alleles = genomic_mutations[position * ":"]["Mutation"].collect{|mutation| Misc.IUPAC_to_base(mutation)}.compact.flatten
+      else
+        alleles = Misc.IUPAC_to_base(genomic_mutations[position * ":"]["Mutation"]) || []
+      end
 
       transcripts.each do |transcript, offset_info|
         offset, strand = offset_info
-        begin
-          codon = Organism.codon_at_transcript_position(org, transcript, offset, transcript_sequence, transcript_5utr)
-        rescue
-          Log.medium $!.message
-          next
-        end
-
-        if not codon.nil?
+        codon = begin
+                  Organism.codon_at_transcript_position(org, transcript, offset, transcript_sequence, transcript_5utr)
+                rescue
+                  Log.medium $!.message
+                  next
+                end
+
+        if not codon.nil? and not codon.empty?
           alleles.each do |allele|
-            allele = Misc::BASE2COMPLEMENT[allele] if strand == -1
+            allele = Misc::BASE2COMPLEMENT[allele] if strand == "-1"
             change = Organism.codon_change(allele, *codon.values_at(0,1))
             pos_code = position * ":"
             mutation = [change.first, codon.last + 1, change.last] * ""
@@ -298,8 +436,93 @@ protein products of transcripts including those positions.
 
     end
 
+    step(:identify_proteins, "Identify Proteins for Transcripts")
+    transcript_field = results.identify_field "Ensembl Transcript ID"
+    results.add_field "#{org.sub(/\/.*/,'')}:Ensembl Protein ID" do |key,values|
+      values[transcript_field].collect do |transcript| transcript_to_protein[transcript] end
+    end
+
+
     results
   end
+
+
+  task_option :organism, "Organism", :string, "Hsa"
+  task_option :genomic_mutations, "Position (chr:position)\tMutation", :tsv
+  task_dependencies nil
+  task :identify_germline_variations => :tsv do |org,genomic_mutations|
+    genomic_mutations = case
+                        when TSV === genomic_mutations
+                          genomic_mutations
+                        else
+                          TSV.new StringIO.new(genomic_mutations), :list
+                        end
+
+    genomic_mutations.key_field ||= "Position"
+    genomic_mutations.fields    ||= ["Mutation"]
+
+    positions = genomic_mutations.keys.collect{|l| l.split(":")}
+
+
+    step(:prepare, "Prepare Results")
+    results = TSV.new({})
+    results.key_field = "Position"
+    results.fields = ["SNP Id"]
+    results.type = :double
+    results.filename = path
+
+
+    step(:resources, "Load Resources")
+
+    snp_ids = Organism.identify_variations_at_genomic_positions(org, positions, Organism.germline_variations(org).produce).collect{|ids| ids * "|"}
+    snps_for_positions = Hash[*genomic_mutations.keys.zip(snp_ids).flatten]
+
+    genomic_mutations.add_field "Germline SNP Id" do |position, values|
+      snps_for_positions[position]
+    end
+
+    genomic_mutations
+  end
+
+
+  task_option :organism, "Organism", :string, "Hsa"
+  task_option :genomic_mutations, "Position (chr:position)\tMutation", :tsv
+  task_dependencies nil
+  task :identify_somatic_variations => :tsv do |org,genomic_mutations|
+    genomic_mutations = case
+                        when TSV === genomic_mutations
+                          genomic_mutations
+                        else
+                          TSV.new StringIO.new(genomic_mutations), :list
+                        end
+
+    genomic_mutations.key_field ||= "Position"
+    genomic_mutations.fields    ||= ["Mutation"]
+
+    positions = genomic_mutations.keys.collect{|l| l.split(":")}
+
+
+    step(:prepare, "Prepare Results")
+    results = TSV.new({})
+    results.key_field = "Position"
+    results.fields = ["SNP Id"]
+    results.type = :double
+    results.filename = path
+
+
+    step(:resources, "Load Resources")
+
+    snp_ids = Organism.identify_variations_at_genomic_positions(org, positions, Organism.somatic_variations(org).produce).collect{|ids| ids * "|"}
+    snps_for_positions = Hash[*genomic_mutations.keys.zip(snp_ids).flatten]
+
+    genomic_mutations.add_field "Germline SNP Id" do |position, values|
+      snps_for_positions[position]
+    end
+
+    genomic_mutations
+  end
+
+
 end
 
 if __FILE__ == $0
@@ -333,6 +556,17 @@ X	10085674	C	T
 21 19638426 G T
   EOF
 
+  exon_juncture_test = <<-EOF
+#Position Mutation
+7:150753996 T
+  EOF
+
+
+  job =  Organism.job :genomic_mutations_in_exon_junctures, "Test1", TSV.new(StringIO.new(exon_juncture_test), :list, :sep => " "), :organism => "Hsa"
+  job.run
+  job.clean if job.error?
+  puts job.messages
+  puts job.read
 
 #  # Build 36
 #  picmi_test = <<-EOF