bioinform 0.1.15 → 0.1.16

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8a22c28659b2ef2205e84ba7dcd9778d392ec54b
-  data.tar.gz: f7280fc53cd3dbdd195278b9df06d91367d5baa0
+  metadata.gz: 58c71a402e76e3edcd2105b21dd19a84b37b4c80
+  data.tar.gz: f300eba436a2b06c5b5feef4bb3fb95c0397b0a7
 SHA512:
-  metadata.gz: 08999aa322ad4c0119a92b2e41087312afcc2747a01e2cf11a772a59e6d47d2eb30f3bb232a1dfc0400ffd363359a01b61235abafcd6e2561aba7742aba4ca56
-  data.tar.gz: 9a2662d63282b3362340006dea59e2da0f95869f673d4b80a9aa29433e49d0f7f85ec1464123a280b7053a6f502adeea9325f7aef9a2d73ad7b31d3c0931b26a
+  metadata.gz: 1ab6af7c582ca7de30a7b08c73c5979bda3b727cfd810d52114bbd9b657311dc7d634d0f991fd375b2efb86b730badbc48cc41ea512d05197f46f49b673b0ee5
+  data.tar.gz: f2162a51dec39a7c500474ac4739c40869c00be3b9f837224e35f21137436ea9a827b0d99d5e82d617f0aff30fa30da3dbfed3066d4684fe4cc629f96f3e8443

data/lib/bioinform/data_models/collection.rb

@@ -72,4 +72,4 @@ module Bioinform
     end
 
   end
-end
+end

data/lib/bioinform/data_models/motif.rb

@@ -53,4 +53,4 @@ module Bioinform
       parameters.to_s
     end
   end
-end
+end

data/lib/bioinform/data_models/pcm.rb

@@ -5,6 +5,8 @@ require_relative '../conversion_algorithms/pcm2pwm_converter'
 
 module Bioinform
   class PCM < PM
+    make_parameters :pseudocount
+
     def count
       matrix.first.inject(&:+)
     end
@@ -31,4 +33,4 @@ module Bioinform
       super + validation_errors
     end
   end
-end
+end

data/lib/bioinform/data_models/pm.rb

@@ -123,6 +123,13 @@ module Bioinform
       end
     end
 
+    def consensus
+      each_position.map{|pos|
+        pos.each_with_index.max_by{|el, letter_index| el}
+      }.map{|el, letter_index| letter_index}.map{|letter_index| %w{A C G T}[letter_index] }.join
+    end
+
+
     def to_hash
       hsh = %w{A C G T}.each_with_index.collect_hash do |letter, letter_index|
         [ letter, @matrix.map{|pos| pos[letter_index]} ]
@@ -187,4 +194,4 @@ module Bioinform
       PWM.new(get_parameters.merge(matrix: matrix))
     end
   end
-end
+end

data/lib/bioinform/data_models/ppm.rb

@@ -3,9 +3,19 @@ require_relative '../data_models'
 
 module Bioinform
   class PPM < PM
+    make_parameters :effective_count, :pseudocount
     def to_ppm
       self
     end
+
+    def to_pcm
+      PCM.new(matrix.map{|pos| pos.map{|el| el * effective_count} }).tap{|pcm| pcm.name = name}
+    end
+
+    def to_pwm
+      pseudocount ? to_pcm.to_pwm(pseudocount) : to_pcm.to_pwm
+    end
+
     def self.valid_matrix?(matrix, options = {})
       precision = options[:precision] || 0.01
       super && matrix.all?{|pos| ((pos.inject(0, &:+) - 1.0).abs <= precision)  && pos.all?{|el| el >=0 } }
@@ -18,4 +28,4 @@ module Bioinform
       super + validation_errors
     end
   end
-end
+end

data/lib/bioinform/data_models/pwm.rb

@@ -52,5 +52,15 @@ module Bioinform
     def worst_suffix(i)
       @matrix[i...length].map(&:min).inject(0.0, &:+)
     end
+
+
+    def matrix_rounded(n)
+      matrix.map{|pos| pos.map{|x| x.round(n) } }
+    end
+    private :matrix_rounded
+
+    def round(n)
+      PWM.new(matrix_rounded(n)).tap{|pm| pm.name = name}
+    end
   end
-end
+end

data/lib/bioinform/parsers/matrix_parser.rb

@@ -0,0 +1,40 @@
+module Bioinform  
+  class MatrixParser
+    def initialize(options = {})
+      @has_name = options.fetch(:has_name, true)
+      @name_pattern = options.fetch(:name_pattern, /^>?\s*(?<name>[^\t\r\n]+).*$/)
+      @has_header_row = options.fetch(:has_header_row, false)
+      @has_header_column = options.fetch(:has_header_column, false)
+      @nucleotides_in = options.fetch(:nucleotides_in, :columns)
+
+      raise ':nucleotides_in option should be either :rows or :columns' unless  [:rows, :columns].include?(@nucleotides_in)
+    end
+
+    def parse!(input)
+      lines = input.lines
+      if @has_name
+        match = lines.shift.match(@name_pattern)
+        raise 'Name pattern doesn\'t match'  unless match
+        name = match[:name]
+      end
+      lines.shift  if @has_header_row
+      matrix = lines.map(&:strip).reject(&:empty?).map{|line| line.split }
+      matrix = matrix.map{|row| row.drop(1) }  if @has_header_column
+      matrix = matrix.map{|row| row.map{|el| Float(el) } }
+
+      matrix = matrix.transpose  if @nucleotides_in == :rows
+      # raise 'Matrix not valid' unless ! matrix.empty? && matrix.all?{|pos| pos.size == 4 }
+      {matrix: matrix, name: name}
+    end
+
+    def parse(input)
+      parse!(input) rescue nil
+    end
+
+    def valid?(input)
+      result = parse!(input)
+    rescue
+      false
+    end
+  end
+end

data/lib/bioinform/parsers.rb

@@ -4,4 +4,6 @@ require_relative 'parsers/yaml_parser'
 require_relative 'parsers/string_parser'
 require_relative 'parsers/string_fantom_parser'
 require_relative 'parsers/splittable_parser'
-require_relative 'parsers/jaspar_parser'
+require_relative 'parsers/jaspar_parser'
+
+require_relative 'parsers/matrix_parser'

data/lib/bioinform/version.rb

@@ -1,3 +1,3 @@
 module Bioinform
-  VERSION = "0.1.15"
+  VERSION = "0.1.16"
 end

data/lib/bioinform.rb

@@ -6,5 +6,32 @@ require_relative 'bioinform/data_models'
 require_relative 'bioinform/cli'
 
 module Bioinform
-  # Your code goes here...
+  class Error < StandardError
+  end
+
+  def self.get_pwm(data_model, matrix, background, pseudocount, effective_count)
+    pm = Bioinform.const_get(data_model).new(matrix)
+    pm.set_parameters(background: background)
+    if pseudocount && ! pseudocount.blank? && [:PCM,:PPM].include?(data_model.to_sym)
+      pm.set_parameters(pseudocount: pseudocount)
+    end
+    if effective_count && [:PPM].include?(data_model.to_sym)
+      pm.set_parameters(effective_count: effective_count)
+    end
+    pm.to_pwm
+  rescue => e
+    raise "PWM creation failed (#{e})"
+  end
+
+  def self.get_pcm(data_model, matrix, effective_count)
+    pm = Bioinform.const_get(data_model).new(matrix)
+    if effective_count && [:PPM].include?(data_model.to_sym)
+      pm.set_parameters(effective_count: effective_count)
+    end
+    pm.to_pcm
+  end
+
+  def self.get_ppm(data_model, matrix)
+    Bioinform.const_get(data_model).new(matrix).to_ppm
+  end
 end

data/spec/data_models/pm_spec.rb

@@ -356,5 +356,12 @@ module Bioinform
         end
       end
     end
+
+    describe '#consensus' do
+      let(:pm) { PM.new( [[1.3, 2.0, 4.9, 3.2], [7.13, 6.5, 3.25, 4.633], [-1.0, -0.5, -1.5, -1.0]] ) }
+      it 'is a string of nucleotides of maximal weights' do
+        pm.consensus.should == 'GAC'
+      end
+    end
   end
 end

data/spec/data_models/ppm_spec.rb

@@ -3,6 +3,52 @@ require_relative '../../lib/bioinform/data_models/pcm'
 
 module Bioinform
   describe PPM do
+    describe '#to_ppm' do
+      let (:ppm_motif) { Fabricate(:ppm) }
+      it 'returns self' do
+         ppm_motif.to_ppm.should eq ppm_motif
+      end
+    end
 
+    describe '#to_pcm' do
+      let (:ppm_motif) { Fabricate(:ppm_by_pcm) }
+      let (:pcm_motif) { Fabricate(:pcm) }
+
+      it 'returns pcm using given effective_count' do
+        ppm = ppm_motif.tap{|ppm| ppm.effective_count = pcm_motif.count }
+        ppm.to_pcm.should == pcm_motif
+      end
+      it 'without given count it raises an error' do
+        expect{ ppm_motif.to_pcm }.to raise_error
+      end
+      it 'returns pcm with the same name' do
+        ppm = ppm_motif.tap{|ppm| ppm.effective_count = pcm_motif.count }
+        ppm.to_pcm.name.should == ppm_motif.name
+      end
+    end
+
+    describe '#to_pwm' do
+      let (:ppm_motif_without_count) { Fabricate(:ppm_by_pcm) }
+      let (:ppm_motif) { ppm_motif_without_count.tap{|ppm| ppm.effective_count = 137 } }
+      let (:ppm_motif_with_log_pseudocount) { ppm_motif.tap{|ppm| ppm.effective_count = 137 } }
+      let (:pcm_motif) { ppm_motif.to_pcm }
+
+      it 'returns pwm the same as pwm of according pcm' do
+        ppm_motif.to_pwm.should == pcm_motif.to_pwm
+      end
+      it 'uses pseudocount to transform according pcm to pwm' do
+        ppm = ppm_motif.tap{|ppm| ppm.pseudocount = 10}
+        ppm_motif.to_pwm.should == pcm_motif.to_pwm(10)
+      end
+      it 'by default uses pseudocount equal to log of count' do
+        ppm_motif.to_pwm.should == ppm_motif.to_pcm.to_pwm(Math.log(137))
+      end
+      it 'without given count it raises an error' do
+        expect{ ppm_motif_without_count.to_pwm }.to raise_error
+      end
+      it 'returns pwm with the same name' do
+        ppm_motif.to_pwm.name.should == ppm_motif.name
+      end
+    end
   end
-end
+end

data/spec/data_models/pwm_spec.rb

@@ -79,5 +79,18 @@ module Bioinform
       end
     end
 
+    describe '#round' do
+      let(:matrix) { [[1.3, 2.0, 4.9, 3.2], [7.13, 6.5, 3.25, 4.633], [-1.0, -0.5, -1.5, -1.0]] }
+      let(:pm) { PWM.new( matrix ).tap{|pm| pm.name = 'motif name'} }
+      it 'gives model with matrix elements rounded' do
+        pm.round(1).matrix.should == [[1.3, 2.0, 4.9, 3.2], [7.1, 6.5, 3.3, 4.6], [-1.0, -0.5, -1.5, -1.0]]
+      end
+      it 'gives PWM model' do
+        pm.round(1).should be_kind_of(PWM)
+      end
+      it 'gives model with the same name' do
+        pm.round(1).name.should == 'motif name'
+      end
+    end
   end
-end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bioinform
 version: !ruby/object:Gem::Version
-  version: 0.1.15
+  version: 0.1.16
 platform: ruby
 authors:
 - Ilya Vorontsov
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-18 00:00:00.000000000 Z
+date: 2014-04-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: docopt
@@ -68,6 +68,7 @@ files:
 - lib/bioinform/formatters/transfac_formatter.rb
 - lib/bioinform/parsers.rb
 - lib/bioinform/parsers/jaspar_parser.rb
+- lib/bioinform/parsers/matrix_parser.rb
 - lib/bioinform/parsers/parser.rb
 - lib/bioinform/parsers/splittable_parser.rb
 - lib/bioinform/parsers/string_fantom_parser.rb
@@ -180,7 +181,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.2.1
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Classes for work with different input formats of positional matrices and