macroape 4.0.1 → 4.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4499602694103f438e53430766a1b5bad85baf61
-  data.tar.gz: e95adc086488900be4e0f0654da3543a8ea6a8c2
+  metadata.gz: fe425b463f7ed42c64027a15d68ee85a38334058
+  data.tar.gz: f9789653cf1708f00eb9255a24c4c009c352f3a6
 SHA512:
-  metadata.gz: 42fa7ba9a07625432176831623549395695aa936f19efcfd62e9282f1dc6bb5b93359103b7e84de6d3fcda4686dc5f8f1264756a050e0b1f2fb433c4b96809bb
-  data.tar.gz: d4e94f5384e9e1cf4d34d777349b2d4e64e780f7b8ad9a5c71228c4b83b0b791165499c7bdc8c6c84a0a8d384bd617e9c312e68cb27e32713cd0be79a34cf879
+  metadata.gz: a09caa08fe447ef672ad91793151f0d1a2fef245141e86f5f54e467d881e32863a0e76502e7345297d702c9115d1b721e365a1c6e314e5cd5b55920e0f04a585
+  data.tar.gz: fcdf7a5724d1a8ad1d589a5d1d9ec892d780df3d05234af09e6ee546a486a95f4c833317ec2392fae65c55dd406b30d7e71e0eba9c09833edd86c4d489817679

data/TODO.txt

@@ -1,3 +1,10 @@
+Why  Helpers.find_threshold_output('SP1_f1.pwm 0.8 -d 10')  on java works only for pvalues less or equal than 0.5, but 0.55 or 0.8 breaks the program (it doesn't stop). My hypothesis is that gauss threshold estimation fails
+
+
+Make --same-strand mode which disallows revcomp transformation of motif - so it'll be possible to compare RNA-PWMs.
+
+Fix align_motifs so that keys can go after arguments (use OptionParser in other words)
+
 ToDo:
 6)
 # TODO:  FIX: this test fails due to floating point precision error: estimated threshold is -19.0418 but '-19.0418'.to_f * 10000 = -190417.99999999997

data/lib/macroape/cli/align_motifs.rb

@@ -77,11 +77,11 @@ module Macroape
 
         shifts = []
         shifts << [leader_pwm_file, 0, :direct]
-        pwm_first = data_model.new(File.read(leader_pwm_file)).to_pwm
+        pwm_first = data_model.new(File.read(leader_pwm_file)).set_parameters(background: leader_background).to_pwm
         pwm_first.set_parameters(background: leader_background, max_hash_size: max_hash_size).discrete!(discretization)
 
         rest_pwm_files.each do |motif_name|
-          pwm_second = data_model.new(File.read(motif_name)).to_pwm
+          pwm_second = data_model.new(File.read(motif_name)).set_parameters(background: rest_motifs_background).to_pwm
           pwm_second.set_parameters(background: rest_motifs_background, max_hash_size: max_hash_size).discrete!(discretization)
           cmp = Macroape::PWMCompare.new(pwm_first, pwm_second).set_parameters(max_pair_hash_size: max_pair_hash_size)
           info = cmp.jaccard_by_pvalue(pvalue)

data/lib/macroape/cli/eval_alignment.rb

@@ -101,7 +101,7 @@ module Macroape
           raise "Error! File #{first_file} don't exist" unless File.exist?(first_file)
           input_first = File.read(first_file)
         end
-        pwm_first = data_model.new(input_first).to_pwm
+        pwm_first = data_model.new(input_first).set_parameters(background: first_background).to_pwm
 
         if second_file == '.stdin'
           input_second = parser.parse
@@ -109,7 +109,7 @@ module Macroape
           raise "Error! File #{second_file} don't exist" unless File.exist?(second_file)
           input_second = File.read(second_file)
         end
-        pwm_second = data_model.new(input_second).to_pwm
+        pwm_second = data_model.new(input_second).set_parameters(background: second_background).to_pwm
 
         pwm_first.set_parameters(background: first_background, max_hash_size: max_hash_size).discrete!(discretization)
         pwm_second.set_parameters(background: second_background, max_hash_size: max_hash_size).discrete!(discretization)

data/lib/macroape/cli/eval_similarity.rb

@@ -81,7 +81,7 @@ module Macroape
           raise "Error! File #{first_file} don't exist" unless File.exist?(first_file)
           input_first = File.read(first_file)
         end
-        pwm_first = data_model.new(input_first).to_pwm
+        pwm_first = data_model.new(input_first).set_parameters(background: first_background).to_pwm
 
         if second_file == '.stdin'
           input_second = parser.parse
@@ -89,7 +89,7 @@ module Macroape
           raise "Error! File #{second_file} don't exist" unless File.exist?(second_file)
           input_second = File.read(second_file)
         end
-        pwm_second = data_model.new(input_second).to_pwm
+        pwm_second = data_model.new(input_second).set_parameters(background: second_background).to_pwm
 
         pwm_first.set_parameters(background: first_background, max_hash_size: max_hash_size).discrete!(discretization)
         pwm_second.set_parameters(background: second_background, max_hash_size: max_hash_size).discrete!(discretization)

data/lib/macroape/cli/find_pvalue.rb

@@ -62,7 +62,7 @@ module Macroape
           raise "Error! File #{filename} doesn't exist" unless File.exist?(filename)
           input = File.read(filename)
         end
-        pwm = data_model.new(input).to_pwm
+        pwm = data_model.new(input).set_parameters(background: background).to_pwm
         pwm.set_parameters(background: background, max_hash_size: max_hash_size).discrete!(discretization)
 
         counts = pwm.counts_by_thresholds(* thresholds.map{|count| count * discretization})

data/lib/macroape/cli/find_threshold.rb

@@ -68,7 +68,7 @@ module Macroape
           raise "Error! File #{filename} doesn't exist" unless File.exist?(filename)
           input = File.read(filename)
         end
-        pwm = data_model.new(input).to_pwm
+        pwm = data_model.new(input).set_parameters(background: background).to_pwm
         pwm.set_parameters(background: background, max_hash_size: max_hash_size).discrete!(discretization)
 
         infos = []

data/lib/macroape/cli/preprocess_collection.rb

@@ -90,6 +90,7 @@ module Macroape
           filelist.each do |filename|
             motif = data_model.new(File.read(filename))
             motif.name ||= File.basename(filename, File.extname(filename))
+            motif.set_parameters(background: background)
             motifs << motif
           end
         else

data/lib/macroape/cli/scan_collection.rb

@@ -94,7 +94,7 @@ module Macroape
           query_input = File.read(filename)
         end
 
-        query_pwm = data_model.new(query_input).to_pwm
+        query_pwm = data_model.new(query_input).set_parameters(background: query_background).to_pwm
         query_pwm.set_parameters(background: query_background, max_hash_size: max_hash_size)
 
         query_pwm_rough = query_pwm.discrete(rough_discretization)

data/lib/macroape/version.rb

@@ -1,4 +1,4 @@
 module Macroape
-  VERSION = "4.0.1"
+  VERSION = "4.0.2"
   STANDALONE = false
 end

data/spec/count_distribution_spec.rb

@@ -6,6 +6,9 @@ describe Bioinform::PWM do
   let :matrix_second do [[1,2,3,4],[2,3,4,5]] end
   let :pwm_first do Bioinform::PWM.new(matrix_first) end
   let :pwm_second do Bioinform::PWM.new(matrix_second) end
+  let :background do [0.1,0.4,0.4,0.1] end
+  let :pwm_first_on_background do pwm_first.tap{|pwm| pwm.set_parameters(background: background)} end
+  let :pwm_second_on_background do pwm_second.tap{|pwm| pwm.set_parameters(background: background)} end
 
   context '#count_distribution_after_threshold' do
 
@@ -21,6 +24,19 @@ describe Bioinform::PWM do
       distribution_second.should == { 5=>3, 6=>4, 7=>3, 8=>2, 9=>1 }
     end
 
+    it 'for PWMs on different background it should contain the same scores (keys of hash)' do
+      pwm_first.count_distribution_after_threshold(0).keys.sort.should == pwm_first_on_background.count_distribution_after_threshold(0).keys.sort
+      pwm_first.count_distribution_after_threshold(13).keys.sort.should == pwm_first_on_background.count_distribution_after_threshold(13).keys.sort
+    end
+
+    it 'should return hash of score => count for all scores >= threshold  when calculated on background' do
+      distribution_second = pwm_second_on_background.count_distribution_after_threshold(0)
+      distribution_second.should have_nearly_the_same_values({ 3=>0.01, 4=>0.08, 5=>0.24, 6=>0.34, 7=>0.24, 8=>0.08, 9=>0.01 }, 1e-7 )
+
+      distribution_second = pwm_second_on_background.count_distribution_after_threshold(5)
+      distribution_second.should have_nearly_the_same_values({ 5=>0.24, 6=>0.34, 7=>0.24, 8=>0.08, 9=>0.01 }, 1e-7 )
+    end
+
     it 'should use existing precalculated hash @count_distribution if it exists' do
       pwm = pwm_second;
       pwm.instance_variable_set :@count_distribution, { 3=>10, 4=>20, 5=>30, 6=>40, 7=>30, 8=>20, 9=>10 }
@@ -48,4 +64,46 @@ describe Bioinform::PWM do
     end
   end
 
-end
+  context '#pvalue_by_threshold' do
+    it 'should return probability to be >= than threshold' do
+      pwm_second.pvalue_by_threshold(7).should be_within(1e-7).of(6.0/16)
+    end
+    it 'should return probability to be >= than threshold when calculated on background' do
+      pwm_second_on_background.pvalue_by_threshold(7).should be_within(1e-7).of(0.33)
+    end
+  end
+  context '#threshold' do
+    it 'should return threshold such that according pvalue doesn\'t exceed requested value' do
+      requested_pvalue = 6.0/16
+      threshold = pwm_second.threshold(requested_pvalue)
+      pwm_second.pvalue_by_threshold(threshold).should <= requested_pvalue
+    end
+    it 'should return threshold such that according pvalue doesn\'t exceed requested value when calculated on background' do
+      requested_pvalue = 0.33
+      threshold = pwm_second_on_background.threshold(requested_pvalue)
+      pwm_second_on_background.pvalue_by_threshold(threshold).should <= requested_pvalue
+    end
+    it 'should return threshold such that according pvalue doesn\'t exceed requested value when actual pvalue isn\'t exact equal to requested' do
+      requested_pvalue = 0.335
+      threshold = pwm_second_on_background.threshold(requested_pvalue)
+      pwm_second_on_background.pvalue_by_threshold(threshold).should <= requested_pvalue
+    end
+  end
+  context '#weak_threshold' do
+  it 'should return threshold such that according pvalue exceed requested value' do
+      requested_pvalue = 6.0/16
+      threshold = pwm_second.weak_threshold(requested_pvalue)
+      pwm_second.pvalue_by_threshold(threshold).should >= requested_pvalue
+    end
+    it 'should return threshold such that according pvalue exceed requested value when calculated on background' do
+      requested_pvalue = 0.33
+      threshold = pwm_second_on_background.weak_threshold(requested_pvalue)
+      pwm_second_on_background.pvalue_by_threshold(threshold).should >= requested_pvalue
+    end
+    it 'should return threshold such that according pvalue exceed requested value when actual pvalue isn\'t exact equal to requested' do
+      requested_pvalue = 0.335
+      threshold = pwm_second_on_background.weak_threshold(requested_pvalue)
+      pwm_second_on_background.pvalue_by_threshold(threshold).should >= requested_pvalue
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -1,4 +1,11 @@
 $bioinform_folder = File.expand_path(File.join(File.dirname(__FILE__), '..', '..', 'bioinform', 'lib'))
 $LOAD_PATH.unshift $bioinform_folder
 
-require 'rspec'
+require 'rspec'
+
+# comparing hashes with float values
+RSpec::Matchers.define :have_nearly_the_same_values do |expected, vicinity|
+  match do |actual|
+    expected.all?{|key, _| actual.has_key?(key)} && actual.all?{|key, _| expected.has_key?(key)} && expected.all?{|key, value| (actual[key] - value).abs <= vicinity }
+  end
+end

metadata

@@ -1,27 +1,27 @@
 --- !ruby/object:Gem::Specification
 name: macroape
 version: !ruby/object:Gem::Version
-  version: 4.0.1
+  version: 4.0.2
 platform: ruby
 authors:
 - Ilya Vorontsov
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-01 00:00:00.000000000 Z
+date: 2013-09-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bioinform
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
         version: 0.1.10
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
         version: 0.1.10
 description: Macroape is an abbreviation for MAtrix CompaRisOn by Approximate P-value
@@ -42,7 +42,7 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
-- .gitignore
+- ".gitignore"
 - Gemfile
 - LICENSE
 - README.md
@@ -118,17 +118,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.3
+rubygems_version: 2.1.5
 signing_key: 
 specification_version: 4
 summary: PWM comparison tool using MACROAPE approach
@@ -170,4 +170,3 @@ test_files:
 - test/preprocess_collection_test.rb
 - test/scan_collection_test.rb
 - test/test_helper.rb
-has_rdoc: