rbbt-phgx 0.3.1 → 1.0.0

data/lib/rbbt/mutation/mutation_assessor.rb

@@ -28,13 +28,23 @@ module MutationAssessor
 
     Log.debug "Querying Mutation Assessor for: #{vars.split(/\n/).length}"
     tries = 0
+    nocache = false
     begin
       doc = nil
       TmpFile.with_file(post_data) do |post_file|
-        doc = Nokogiri::HTML(Open.read(URL, :wget_options => {"--post-file" => post_file }, :nocache => false))
+        Log.medium "Updating cache:" if nocache == :update
+        doc = Nokogiri::HTML(Open.read(URL, :wget_options => {"--post-file" => post_file }, :nocache => nocache))
       end
 
       textareas = doc.css('textarea')
+
+      if textareas.empty?
+        puts "No text area"
+        puts doc
+        puts
+        raise NotDone, "No text aread found in response HTML"
+      end
+
       result = textareas.last.content
 
       if result =~ /Cannot parse variant/
@@ -44,14 +54,16 @@ module MutationAssessor
         raise "Cannot parse variants. Variants in file #{ variants }"
       end
 
-      raise NotDone, "Not done" if result =~ /\t.sent\t./
+      raise NotDone, "Not done" if result =~ /\t\[sent\]\t/
     rescue NotDone
-      Log.debug "Mutation Assessor not done, waiting:"
-      Log.debug result
-      
-      sleep 30
       tries += 1
+      nocache = :update
+
+      Log.medium "Mutation Assessor not done, waiting:"
+      sleep 30
+
       if tries < 10
+        Log.medium "Retrying mutation assessor"
         retry
       else
         raise "Error processing Mutation Assessor response"
@@ -80,7 +92,7 @@ module MutationAssessor
   def self.chunked_predict(mutations)
     chunks = mutations.length.to_f / 1000
     chunks = chunks.ceil
-    Misc.divide(mutations.sort, chunks).inject(nil) do |acc, list|
+    Misc.divide(mutations.sort_by{|m| m * ":"}, chunks).inject(nil) do |acc, list|
       if acc.nil?
         acc = predict(list)
       else
@@ -145,14 +157,9 @@ module MutationAssessor
 
         next if uni_accs.compact.reject{|v| v.nil? or v.empty?}.empty?
 
-        ddd uni_accs
-
         mutations = values[protein_field]
-        ddd mutations
 
         uni_accs.zip(mutations).collect do |uni_acc,mutation|
-          ddd uni_acc
-          ddd mutation
           res = case
                 when (mutation.nil? or mutation.empty?)
                   "No Prediction"

data/lib/rbbt/mutation/sift.rb

@@ -6,7 +6,8 @@ module SIFT
   URL_ENSP="http://sift.jcvi.org/sift-bin/retrieve_enst.pl"
 
   def self.predict(mutations)
-    doc = Nokogiri::HTML(Open.read(URL_ENSP, :wget_options => {"--post-data" => "'ENSP=#{mutations.collect{|p| p * ","} * "\n"}'"}, :nocache => false))
+    data_str = mutations.collect{|mut| mut.sub(':', ',')}.uniq * "\n"
+    doc = Nokogiri::HTML(Open.read(URL_ENSP, :wget_options => {"--post-data=" => "'ENSP=#{data_str}'"}))
 
     rows = []
     doc.css('tr').each do |row|
@@ -15,10 +16,20 @@ module SIFT
 
     rows.shift
 
-    if Array === mutations
-      rows
+    if rows.any?
+      TSV.open StringIO.new(rows.collect{|row| row.collect{|v| v.sub(/(ENSP\d+),/,'\1:')} * "\t"} * "\n"), :list,
+        :key_field => "Mutated Isoform", :fields =>["Ensembl Protein ID", "Amino Acid Position", "Wildtype Amino Acid", "Mutant Amino Acid", "Prediction", "Score 1", "Score 2", "Score 3"]
     else
-      rows.first
+      TSV.setup({}, :type => :list, :key_field => "Mutated Isoform", :fields =>["Ensembl Protein ID", "Amino Acid Position", "Wildtype Amino Acid", "Mutant Amino Acid", "Prediction", "Score 1", "Score 2", "Score 3"])
+    end
+  end
+
+  def self.chunked_predict(mutations)
+    chunks = mutations.length.to_f / 100
+    chunks = chunks.ceil
+    tsv = TSV.setup({}, :type => :list, :key_field => "Mutated Isoform", :fields =>["Ensembl Protein ID", "Amino Acid Position", "Wildtype Amino Acid", "Mutant Amino Acid", "Prediction", "Score 1", "Score 2", "Score 3"])
+    Misc.divide(mutations.uniq.sort, chunks).inject(tsv) do |acc, list|
+        acc = TSV.setup(acc.merge(predict(list)))
     end
   end
 

data/lib/rbbt/sources/kegg.rb

@@ -6,6 +6,7 @@ module KEGG
   self.pkgdir = "phgx"
   self.subdir = "share/kegg"
 
+
   KEGG.claim KEGG.root.find, :rake, Rbbt.share.install.KEGG.Rakefile.find(:lib)
 
   def self.names
@@ -13,16 +14,20 @@ module KEGG
   end
 
   def self.descriptions
-    @@descriptions ||= KEGG.pathways.tsv :fields => ["Pathway Description"], :persist => true, :type => :single
+    @@descriptions ||= KEGG.pathways.tsv(:fields => ["Pathway Description"], :persist => true, :type => :single).tap{|o| o.unnamed = true}
   end
 
 
+  def self.index2genes
+    @@index2genes ||= KEGG.gene_pathway.tsv(:key_field => "KEGG Pathway ID", :fields => ["KEGG Gene ID"], :persist => true, :type => :flat, :merge => true).tap{|o| o.unnamed = true}
+  end
+
   def self.index2ens
-    @@index2ens ||= KEGG.identifiers.index :persist => true
+    @@index2ens ||= KEGG.identifiers.index(:persist => true).tap{|o| o.unnamed = true}
   end
 
   def self.index2kegg
-    @@index2kegg ||= KEGG.identifiers.index :target => "KEGG Gene ID", :persist => true
+    @@index2kegg ||= KEGG.identifiers.index(:target => "KEGG Gene ID", :persist => true).tap{|o| o.unnamed = true}
   end
 
   def self.id2name(id)
@@ -34,24 +39,98 @@ module KEGG
   end
 end
 
-module Gene
+if defined? Entity 
+
+  module KeggPathway
+    extend Entity
+    self.format = "KEGG Pathway ID"
+
+    self.annotation :organism
+
+    def self.filter(query, field = nil, options = nil, entity = nil)
+      return true if query == entity
 
-  def to_kegg
-    if Array === self
-      Gene.setup(KEGG.index2kegg.values_at(*to("Ensembl Gene ID")), "KEGG Gene ID", organism)
-    else
-      Gene.setup(KEGG.index2kegg[to("Ensembl Gene ID")], "KEGG Gene ID", organism)
+      return true if KeggPathway.setup(entity.dup, options.merge(:format => field)).name.index query
+
+      false
+    end
+
+    property :name => :single2array do
+      return nil if self.nil?
+      name = KEGG.id2name(self)
+      name.sub(/ - Homo.*/,'') unless name.nil?
+    end
+
+    property :description => :single2array do
+      KEGG.description(self)
     end
-  end
 
-  def from_kegg
-    if Array === self
-      Gene.setup(KEGG.index2ens.values_at(*self), "Ensembl Gene ID", organism)
-    else
-      Gene.setup(KEGG.index2ens[self], "Ensembl Gene ID", organism)
+    property :genes => :array2single do |*args|
+      organism = args.first || self.organism
+      @genes ||= KEGG.index2genes.values_at(*self).
+        each{|pth| pth.organism = organism if pth.respond_to? :organism }
     end
   end
 
+  if defined? Gene and Entity === Gene
+    module Gene
+      self.format = "KEGG Gene ID"
+
+      def to_kegg
+        return self if format == "KEGG Gene ID"
+        if Array === self
+          Gene.setup(KEGG.index2kegg.values_at(*to("Ensembl Gene ID")), "KEGG Gene ID", organism)
+        else
+          Gene.setup(KEGG.index2kegg[to("Ensembl Gene ID")], "KEGG Gene ID", organism)
+        end
+      end
+
+      def _from_kegg
+        return self.clean_annotations unless format == "KEGG Gene ID"
+        if Array === self
+          KEGG.index2ens.values_at(*self)
+        else
+          KEGG.index2ens[self]
+        end
+      end
+
+      def from_kegg
+        return self unless format == "KEGG Gene ID"
+        Gene.setup(_from_kegg, "Ensembl Gene ID", organism)
+      end
+
+      property :_to => :array2single do |new_format|
+        return self if format == new_format
+        list = self._from_kegg 
+
+        tsv = Translation.job(:tsv_translate, "", :organism => organism, :genes => list, :format => new_format).exec.tap{|o| o.unnamed = true}
 
+        tsv.values_at(*list)
+      end
 
+      property :to! => :array2single do |new_format|
+        return self if format == new_format
+
+        new = _to(new_format)
+        new.each_with_index do |n,i|
+          c = self.annotated_array_clean_get_brackets(i)
+          if c.nil? or n.nil?
+            self[i] = nil
+          else
+            c.replace n
+          end
+        end
+      end
+
+      property :to => :array2single do |new_format|
+        return self if format == new_format
+        Gene.setup(_to(new_format), new_format, organism)
+      end
+
+      property :kegg_pathways => :array2single do
+        @kegg_pathways ||= KEGG.gene_pathway.tsv(:persist => true, :key_field => "KEGG Gene ID", :fields => ["KEGG Pathway ID"], :type => :flat, :merge => true).values_at(*self.to_kegg).
+          each{|pth| pth.organism = organism if pth.respond_to? :organism }.tap{|o| KeggPathway.setup(o, organism)}
+      end
+    end
+  end
 end

data/lib/rbbt/sources/matador.rb

@@ -6,5 +6,4 @@ module Matador
   self.subdir = "share/matador"
 
   Matador.claim Matador.root.find, :rake, Rbbt.share.install.Matador.Rakefile.find(:lib)
-
 end

data/lib/rbbt/sources/string.rb

@@ -6,5 +6,4 @@ module STRING
   self.subdir = "share/string"
 
   STRING.claim STRING.root.find, :rake, Rbbt.share.install.STRING.Rakefile.find(:lib)
-
 end

data/share/Cancer/cancer_genes.tsv

@@ -1,4 +1,4 @@
-#Symbol	Name	GeneID	Chr	Chr Band	Cancer Somatic Mut	Cancer Germline Mut	Tumour Types  (Somatic Mutations)	Tumour Types (Germline Mutations)	Cancer Syndrome	Tissue Type	Cancer Molecular Genetics	Mutation Type 	Translocation Partner	Other Germline Mut	Other Syndrome/Disease
+#Associated Gene Name	Name	GeneID	Chr	Chr Band	Cancer Somatic Mut	Cancer Germline Mut	Tumour Types (Somatic Mutations)	Tumour Types (Germline Mutations)	Cancer Syndrome	Tissue Type	Cancer Molecular Genetics	Mutation Type 	Translocation Partner	Other Germline Mut	Other Syndrome/Disease
 ABL1	v-abl Abelson murine leukemia viral oncogene homolog 1	25	9	9q34.1 	yes		CML, ALL, T-ALL			L	Dom	T, Mis	BCR, ETV6, NUP214		
 ABL2	v-abl Abelson murine leukemia viral oncogene homolog 2	27	1	1q24-q25 	yes		AML			L	Dom	T	ETV6		
 ACSL3	acyl-CoA synthetase long-chain family member 3	2181	2	2q36	yes		prostate			E	Dom	T	ETV1		

data/share/install/Matador/Rakefile

@@ -2,9 +2,8 @@ require File.join(File.dirname(__FILE__),'../lib/rake_helper')
 
 define_source_tasks "matador" => "http://matador.embl.de/media/download/matador.tsv.gz"
 
-
 process_tsv :protein_drug, 'matador',
-  :key  => 3, 
+  :key_field  => 3, 
   :fix     => lambda{|l| l.sub(/9606./,'')}, 
   :fields   => [1,0,7,8,9,10,11,12], 
   :header_hash => "",

data/test/rbbt/mutation/test_mutation_assessor.rb

@@ -3,16 +3,15 @@ require 'rbbt/mutation/mutation_assessor'
 
 class TestMutationAssessor < Test::Unit::TestCase
 
-  def _test_predict_aminoacid_mutation
+  def test_predict_aminoacid_mutation
     mutations = {
       "EGFR_HUMAN" => %w(R521K)
     }
 
-    puts MutationAssessor.predict(mutations)
     assert_equal 1, MutationAssessor.predict(mutations).length
   end
 
-  def _test_predict_aminoacid_mutation_tsv
+  def test_predict_aminoacid_mutation_tsv
     tsv = TSV.setup({"EGFR_HUMAN" => [%w(R521K)]}, :key_field => "UniProt/SwissProt ID", :fields => ["Protein Mutation"], :type => :double)
 
     assert_equal "neutral", MutationAssessor.add_predictions(tsv).slice("MutationAssessor:Prediction").values.first.flatten.first
@@ -26,7 +25,7 @@ class TestMutationAssessor < Test::Unit::TestCase
       "P53_HUMAN" => %w(R21K),
     }
 
-    puts MutationAssessor.chunked_predict(mutations)
+    assert(MutationAssessor.chunked_predict(mutations).include? "EGFR_HUMAN R521K")
   end
 
 

data/test/rbbt/mutation/test_sift.rb

@@ -16,10 +16,18 @@ class TestSIFT < Test::Unit::TestCase
 
     assert_equal "TOLERATED", SIFT.predict_aminoacid_mutation_batch( [[accession, mutation]]).first[3]
   end
-  def test_predict_aminoacid_mutation_batch
+
+  def test_predict
     ensp = "ENSP00000224605"
     mutation = "A63T"
-    assert_equal "TOLERATED", SIFT.predict( [[ensp, mutation]]).first[5]
+    assert_equal "TOLERATED", SIFT.predict( [[ensp, mutation] * ":"]).values.first["Prediction"]
   end
 
+  def test_chunked_predict
+    ensp = "ENSP00000224605"
+    mutation = "A63T"
+    assert_equal "TOLERATED", SIFT.chunked_predict( [[ensp, mutation] * ":"] * 2000).values.first["Prediction"]
+  end
+
+
 end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: rbbt-phgx
 version: !ruby/object:Gem::Version 
-  hash: 17
+  hash: 23
   prerelease: 
   segments: 
-  - 0
-  - 3
   - 1
-  version: 0.3.1
+  - 0
+  - 0
+  version: 1.0.0
 platform: ruby
 authors: 
 - Miguel Vazquez
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-10-03 00:00:00 +02:00
+date: 2012-01-13 00:00:00 +01:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -56,11 +56,9 @@ files:
 - lib/rbbt/sources/hprd.rb
 - lib/rbbt/sources/kegg.rb
 - lib/rbbt/sources/matador.rb
-- lib/rbbt/sources/nci.rb
 - lib/rbbt/sources/pharmagkb.rb
 - lib/rbbt/sources/pina.rb
 - lib/rbbt/sources/reactome.rb
-- lib/rbbt/sources/sift.rb
 - lib/rbbt/sources/stitch.rb
 - lib/rbbt/sources/string.rb
 - share/Cancer/anais_annotations

data/lib/rbbt/sources/nci.rb

@@ -1,6 +0,0 @@
-require 'phgx'
-
-module NCI
-  extend Resource
-  data_module PhGx
-end

data/lib/rbbt/sources/sift.rb

@@ -1,5 +0,0 @@
-require 'phgx'
-
-module SIFT
-  #PhGx.software.opt.SIFT.define_as_install Rbbt.share.install.software.SIFT.find
-end