semtools 0.1.1

data/lib/semtools/sim_handler.rb

@@ -0,0 +1,113 @@
+# @author Fernando Moreno Jabato <jabato(at)uma(dot)es>
+# @description functionalities to handle similitude features
+
+
+
+# Applies the WhiteSimilarity from 'text' package over two given texts
+# Param:
+# +textA+:: text to be compared with textB
+# +textB+:: text to be compared with textA
+# Returns the similarity percentage between [0,1]
+def text_similitude(textA, textB)
+  # Check special cases
+  return -1.0 if (textA.nil?) | (textB.nil?)
+  return -1.0 if (!textA.is_a? String) | (!textB.is_a? String)
+  return -1.0 if (textA.length <= 0) | (textB.length <= 0)
+  # Calculate similitude
+  require 'text'
+  white = Text::WhiteSimilarity.new
+  return white.similarity(textA.lstrip, textB.lstrip)
+end  
+
+# Applies the WhiteSimilarity from 'text' package over two given text sets and returns the similitudes
+# of the each element of the first set over the second set 
+# Param:
+# +textsA+:: text set to be compared with textsB
+# +textsB+:: text set to be compared with textsA
+# Returns the maximum similarity percentage between [0,1] for each element of textsA against all elements of textsB
+def ctext_AtoB(textsA, textsB)
+  # Check special cases
+  return [-1.0] if (textsA.nil?) | (textsB.nil?)
+  return [-1.0] if (!textsA.is_a? Array) | (!textsB.is_a? Array)
+  return [-1.0] if (textsA.length <= 0) | (textsB.length <= 0)
+  # Calculate similitude
+  similitudesA = []
+  textsA.each do |fragA|
+    frag_A_similitudes = []
+    textsB.each do |fragB|
+      frag_A_similitudes << text_similitude(fragA, fragB)
+    end
+    begin 
+      similitudesA << frag_A_similitudes.max
+    rescue => e
+      STDERR.puts frag_A_similitudes.inspect
+      STDERR.puts textsA.inspect , textsB.inspect
+      STDERR.puts e.message
+      STDERR.puts e.backtrace
+      Process.exit
+    end 
+  end
+  return similitudesA
+end
+
+# Applies the WhiteSimilarity from 'text' package over two given complex texts.
+# Complex texts will be splitted and compared one by one from A to B and B to A
+# Param:
+# +textA+:: text to be compared with textB
+# +textB+:: text to be compared with textA
+# +splitChar+:: char to split text* complex names
+# +charsToRemove+:: char (or chars set) to be removed from text to be compared
+# Returns the similarity percentage between [0,1] obtained by bidirectional all Vs all similarity
+def complex_text_similitude(textA, textB, splitChar = ";", charsToRemove = "")
+  # Check special cases
+  return -1.0 if (textA.nil?) | (textB.nil?)
+  return -1.0 if (!textA.is_a? String) | (!textB.is_a? String)
+  return -1.0 if (textA.length <= 0) | (textB.length <= 0)
+  # Split&Clean both sets
+  textA_splitted = textA.split(splitChar)
+  textB_splitted = textB.split(splitChar)
+  if !charsToRemove.empty?
+    textA_splitted.map! {|str| str.gsub(/[#{charsToRemove}]/,'')}
+    textA_splitted.select! {|str| str.length > 0}
+    textB_splitted.map! {|str| str.gsub(/[#{charsToRemove}]/,'')}
+    textB_splitted.select! {|str| str.length > 0}
+  end
+  # Per each X elemnt, compare against all Y elements
+  similitudesA = ctext_AtoB(textA_splitted, textB_splitted)
+  similitudesB = ctext_AtoB(textB_splitted, textA_splitted)
+  # Obtain bidirectional similitude
+  similitudesA = similitudesA.inject{ |sum, el| sum + el }.to_f / similitudesA.size
+  similitudesB = similitudesB.inject{ |sum, el| sum + el }.to_f / similitudesB.size
+  # Obtain bidirectional similitude
+  bidirectional_sim = (similitudesA + similitudesB) / 2
+  # Return info
+  return bidirectional_sim
+end
+
+# Applies the WhiteSimilarity from 'text' package over all complex text stored into an array.
+# Complex texts will be splitted and compared one by one from A to B and B to A
+# Param:
+# +items_array+:: text elements to be compared all against others
+# +splitChar+:: char to split text* complex names
+# +charsToRemove+:: char (or chars set) to be removed from texts to be compared
+# +unique+:: boolean flag which indicates if repeated elements must be removed
+# Returns the similarity percentage for all elements into array
+def similitude_network(items_array, splitChar = ";", charsToRemove = "", unique = false)
+  # Special cases
+  return nil if items_array.nil?
+  return nil if !items_array.is_a? Array
+  return nil if items_array.length <= 0
+  # Remove repeated elements
+  items_array.uniq! if unique
+  # Define hash to be filled
+  sims = {}
+  # Per each item into array => Calculate similitude
+  while(items_array.length > 1)
+    current = items_array.shift
+    sims[current] = {}
+    items_array.each do |item|
+      sims[current][item] = complex_text_similitude(current,item,splitChar,charsToRemove)
+    end
+  end 
+  return sims
+end

data/lib/semtools/version.rb

@@ -0,0 +1,3 @@
+module Semtools
+  VERSION = "0.1.1"
+end

data/semtools.gemspec

@@ -0,0 +1,37 @@
+
+lib = File.expand_path("../lib", __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require "semtools/version"
+
+Gem::Specification.new do |spec|
+  spec.name          = "semtools"
+  spec.version       = Semtools::VERSION
+  spec.authors       = ["seoanezonjic", "fmjabato"]
+  spec.email         = ["seoanezonjic@hotmail.com", "fmjabato@gmail.com"]
+
+  spec.summary       = %q{Gem to handle semantic based calculations in text and defined ontologies as GO or HPO.}
+  spec.description   = %q{This gem allows to perform ontology based operations and calculation of Semantic similarity and information coefficient using different implementations.}
+  spec.homepage      = "https://github.com/seoanezonjic/semtools"
+  spec.license       = "MIT"
+
+  # Prevent pushing this gem to RubyGems.org. To allow pushes either set the 'allowed_push_host'
+  # to allow pushing to a single host or delete this section to allow pushing to any host.
+#  if spec.respond_to?(:metadata)
+ #   spec.metadata["allowed_push_host"] = "TODO: Set to 'http://mygemserver.com'"
+ # else
+ #   raise "RubyGems 2.0 or newer is required to protect against " \
+ #     "public gem pushes."
+ # end
+
+  spec.files         = `git ls-files -z`.split("\x0").reject do |f|
+    f.match(%r{^(test|spec|features)/})
+  end
+  spec.bindir        = "bin"
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.require_paths = ["lib"]
+  
+  spec.add_dependency "text"
+
+  spec.add_development_dependency "rake"
+  spec.add_development_dependency "rspec"
+end

metadata

@@ -0,0 +1,113 @@
+--- !ruby/object:Gem::Specification
+name: semtools
+version: !ruby/object:Gem::Version
+  version: 0.1.1
+platform: ruby
+authors:
+- seoanezonjic
+- fmjabato
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2021-02-17 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: text
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description: This gem allows to perform ontology based operations and calculation
+  of Semantic similarity and information coefficient using different implementations.
+email:
+- seoanezonjic@hotmail.com
+- fmjabato@gmail.com
+executables:
+- console
+- onto2json.rb
+- setup
+- strsimnet.rb
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".rspec"
+- ".travis.yml"
+- CODE_OF_CONDUCT.md
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- bin/console
+- bin/onto2json.rb
+- bin/setup
+- bin/strsimnet.rb
+- lib/data/hp.obo
+- lib/data/phenotype_annotation.tab
+- lib/semtools.rb
+- lib/semtools/math_methods.rb
+- lib/semtools/ontology.rb
+- lib/semtools/sim_handler.rb
+- lib/semtools/version.rb
+- semtools.gemspec
+homepage: https://github.com/seoanezonjic/semtools
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+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: []
+rubygems_version: 3.2.3
+signing_key:
+specification_version: 4
+summary: Gem to handle semantic based calculations in text and defined ontologies
+  as GO or HPO.
+test_files: []