bio-chembl 0.1.1

data/lib/bio-chembl/chembl.rb

@@ -0,0 +1,30 @@
+require 'cgi'
+require 'uri'
+require 'bio/version'
+require 'curb'
+require 'nokogiri'
+
+#
+module BioChEMBL
+
+  def self.website
+    "https://www.ebi.ac.uk/chembl/"
+  end
+  
+  # BioChEMBL.to_array(aTarget.synonyms) #=> []
+  def self.to_array(str)
+    str.to_s.split('; ')
+  end
+
+end
+
+
+require 'bio-chembl/chemblid.rb'
+require 'bio-chembl/rest_client.rb'
+
+require 'bio-chembl/datamodel.rb'
+require 'bio-chembl/compound.rb'
+require 'bio-chembl/target.rb'
+require 'bio-chembl/assay.rb'
+require 'bio-chembl/bioactivity.rb'
+

data/lib/bio-chembl/chemblid.rb

@@ -0,0 +1,102 @@
+require 'cgi'
+require 'uri'
+require 'bio/version'
+require 'curb'
+require 'nokogiri'
+
+
+
+
+module BioChEMBL
+      
+  # ChEMBL ID
+  #
+  # CHEMBL1
+  #
+  # cid = BioChEMBL::ChEMBLID.new("CHEMBL1")
+  # cid.is_compound? #=> true
+  # cid.resolve #=> aBioChEMBL::Compound
+  #
+  class ChEMBLID < String
+    
+    attr_accessor :data_type
+ 
+    def self.validate_chemblId(str)
+      unless str =~ /^CHEMBL\d+$/
+        raise Exception, "Invalid ChEMBL ID."
+      end
+    end
+       
+    def initialize(str)
+      @data_type = nil
+      self.validate_chemblId(str)
+      super(str)
+    end
+
+   
+    def resolve
+      case @data_type
+      when Compound
+        Compound.find(self.to_s)
+      when Target
+        Target.find(self.to_s)
+      when Assay
+        Assay.find(self.to_s)
+      else
+        begin
+          Compound.find(self.to_s)
+        rescue
+        end 
+        begin
+          Target.find(self.to_s)
+        rescue
+        end 
+        begin
+          Assay.find(self.to_s)
+        rescue
+        end 
+      end
+    end
+    
+    def is_compound?
+      if @data_type == Compound
+        return true
+      else
+        if Compound.find(self.to_s)
+          @data_type = Compound
+          return true
+        else
+          return false  
+        end 
+      end
+    end
+    
+    def is_target?
+      if @data_type == Assay
+        return true
+      else
+        if Assay.find(self.to_s)
+          @data_type = Assay
+          return true
+        else
+          return false  
+        end 
+      end    
+    end
+    
+    def is_assay?
+      if @data_type == Assay
+        return true
+      else
+        if Assay.find(self.to_s)
+          @data_type = Assay
+          return true
+        else
+          return false  
+        end 
+      end   
+    end
+
+  end
+
+end

data/lib/bio-chembl/compound.rb

@@ -0,0 +1,157 @@
+require 'nokogiri'
+require 'bio-chembl/datamodel.rb'
+
+module BioChEMBL
+  
+  # ChEMBL Compound Data Container and Parser
+  #
+  # XML Data string
+  # <compound>
+  #  <chemblId>CHEMBL1</chemblId>
+  #  <knownDrug>No</knownDrug>
+  #  <medChemFriendly>Yes</medChemFriendly>
+  #  <passesRuleOfThree>No</passesRuleOfThree>
+  #  <molecularFormula>C32H32O8</molecularFormula>
+  #  <smiles>COc1ccc2[C@@H]3[C@H](COc2c1)C(C)(C)OC4=C3C(=O)C(=O)C5=C4OC(C)(C)[C@@H]6COc7cc(OC)ccc7[C@H]56</smiles>
+  #  <stdInChiKey>GHBOEFUAGSHXPO-XZOTUCIWSA-N</stdInChiKey>
+  #  <numRo5Violations>1</numRo5Violations>
+  #  <rotatableBonds>2</rotatableBonds>
+  #  <molecularWeight>544.59167</molecularWeight>
+  #  <alogp>3.627</alogp>
+  #  <acdLogp>7.669</acdLogp>
+  #  <acdLogd>7.669</acdLogd>
+  # </compound>
+  #
+  # Usage
+  # ```cpd = BioChEMBL::Compound.find("CHEMBL1")
+  #    cpd.chemblId #=> "CHEMLB1"
+  #    cpd.smiles
+  #
+  #    cpd2 = BioChEMBL::Compound.find_all_by_smiles(cpd.smile)
+  #
+  #    cpd3 = BioChEMBL::Compound.parse(xml)
+  # ```
+  class Compound
+    extend BioChEMBL::DataModel
+
+    ATTRIBUTES =  [ 
+      :chemblId, 
+      :knownDrug,
+      :medChemFriendly,
+      :passesRuleOfThree,
+      :molecularFormula,
+      :smiles,
+      :stdInChiKey,
+      :species,
+      :numRo5Violations,
+      :rotatableBonds,
+      :molecularWeight,
+      :alogp,
+      :acdAcidicPka,
+      :acdLogp,
+      :acdLogd 
+    ]
+      
+    # aBioChEMBL::Compound instance have attribute accessors.
+    # Values of all attributes are in String.
+    set_attr_accessors(ATTRIBUTES)
+    
+    #
+    # BioChEMBL::Compound.parse(doc)
+    def self.parse(str)
+      case str
+      when /^</
+        format = 'xml'
+      when /^\{/
+        format = 'json'
+      else
+        raise ArgumentError, "Unexpected file format: #{str.inspect}" 
+      end
+      begin
+        eval "self.parse_#{format}(str)"
+      rescue 
+        raise NoMethodError
+      end    
+    end 
+    
+    # XML 
+    # <compound>
+    def self.parse_xml(str)
+      xml = Nokogiri::XML(str)
+      this = new  
+      eval set_attr_values(ATTRIBUTES)
+      this
+    end 
+    
+    # XML
+    # <list><compound> ...
+    def self.parse_list_xml(str)
+      xmls = Nokogiri::XML(str)
+      xmls.xpath("/list/compound").map do |cpd|
+        self.parse_xml(cpd.to_s)
+      end
+    end 
+
+    # JSON
+    def self.parse_json(str)
+      raise NotImplementedError
+    end
+
+    # RDF
+    def self.parse_rdf(str)
+      raise NotImplementedError
+    end   
+    
+    
+    # Compound.find(chemblId)
+    # Find a compound data by a ChEMBL ID
+    def self.find(chemblId)
+      self.parse_xml(REST.new.compounds(chemblId))
+    end 
+
+    # Compound.find_by_smiles(smiles)
+    # Find a compound data by a SMILES
+    def self.find_by_smiles(smiles)
+      self.find_all_by_smiles(smiles).first
+    end 
+
+    # Compound.find_all_by_smiles(smiles)
+    # Find compounds by a SMILES.
+    def self.find_all_by_smiles(smiles)
+      self.parse_list_xml(REST.new.compounds_smiles(smiles))
+    end 
+
+    # Compound.find_by_stdinchikey(stdinchikey)
+    # Find a compound data by a StdInChiKey
+    def self.find_by_stdinchikey(stdinchikey)
+      self.parse_xml(REST.new.compounds_stdinchikey(stdinchikey))
+    end 
+
+    # Compound.find_all_by_substructure(smiles)
+    # Substructure Search by a SMILES
+    def self.find_all_by_substructure(smiles)
+      self.parse_list_xml(REST.new.compounds_substructure(smiles))
+    end 
+    
+    # Compound.find_similarity(smiles_with_similarity)
+    # Search compounds by a SMILES with similarity
+    def self.find_all_by_similarity(smiles_with_similarity)
+      self.parse_list_xml(REST.new.compounds_similarity(smiles_with_similarity))
+    end 
+
+    
+    # new
+    def initialize(chemblId = nil)
+      @chemblId = chemblId
+    end 
+    
+    # Resolve the compound data by given ChEMBL ID
+    def resolve
+      resolved = self.class.find(@chemblId)
+      ATTRIBUTES.each do |attr|
+        eval "@#{attr} = resolved.#{attr}"
+      end
+    end
+  end
+  
+end

data/lib/bio-chembl/datamodel.rb

@@ -0,0 +1,25 @@
+require 'nokogiri'
+
+
+module BioChEMBL
+  
+  # BioChEMBL::DataModel
+  #
+  module DataModel
+    
+    def set_attr_accessors(attributes)
+      attributes.each do |attr|
+        eval "attr_accessor :#{attr}"
+      end
+    end
+    
+   
+    def set_attr_values(attributes)
+      attributes.map do |attr|
+        "this.#{attr} = xml.xpath('/#{self.to_s.split('::').last.downcase}/#{attr}').text"
+      end.join("\n")
+    end
+    
+  end
+
+end

data/lib/bio-chembl/rest_api_server.rb

@@ -0,0 +1,40 @@
+require 'cgi'
+require 'uri'
+require 'nokogiri'
+
+
+module BioChEMBL
+  class REST
+    # serv = BioChEMBL::REST::Server.new
+    #    
+    class Server
+      def initialize
+        
+      end
+      
+      # serv.query(:status)
+      # serv.query(:compounds, 'CHEML1')
+      def query(action, args, options)
+      end
+      
+      # /compounds/CHEMBL1
+      def compounds(chemblId)
+      end
+      # /compounds/CHEMBL1/image
+      def compounds_image(chemblId)
+      end
+      # /compounds/CHEMBL1/bioactivities
+      def comopunds_bioactivities(chemblId)
+      end
+      
+    end
+    
+  end
+  
+  class DB
+    class Query
+      def initialize
+      end
+    end
+  end
+end

data/lib/bio-chembl/rest_client.rb

@@ -0,0 +1,264 @@
+require 'cgi'
+require 'uri'
+require 'bio/version'
+require 'curb'
+require 'nokogiri'
+
+require 'bio-chembl/compound.rb'
+require 'bio-chembl/target.rb'
+require 'bio-chembl/assay.rb'
+require 'bio-chembl/bioactivity.rb'
+
+module BioChEMBL
+  
+  class REST
+    
+    HOST_NAME = "www.ebi.ac.uk"
+    API_ROOT = "chemblws"
+    BASE_URI = "https://" + HOST_NAME + "/" + API_ROOT
+    
+    
+    # BioChEMBL::REST::ChEMBL_URI module
+    #
+    module ChEMBL_URI
+      # 
+      def self.address(path)
+        "#{BASE_URI}/#{path}"
+      end
+     
+      
+      # BioChEMBL::REST::ChEMBL_URI.status 
+      def self.status
+        # Example URL: http://www.ebi.ac.uk/chemblws/status/
+        address("status/")
+      end
+      
+      # BioChEMBL::REST::ChEMBL_URI.compounds()       
+      # compounds("CHEMBL1")
+      def self.compounds(chemblId = nil, arg = nil, params = nil)
+        if chemblId and arg == nil and params == nil 
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/CHEMBL1
+          address("compounds/#{chemblId}")
+        elsif chemblId and arg == 'image' and params == nil
+        # Example URL: http://www.ebi.ac.uk/chemblws/compounds/CHEMBL192/image
+          address("compounds/#{chemblId}/#{arg}")
+        elsif chemblId and arg == 'image' and params 
+        # Example URL with dimensions parameter: http://www.ebi.ac.uk/chemblws/compounds/CHEMBL192/image?dimensions=200
+          address("compounds/#{chemblId}/#{arg}?" + params.map {|k,v| "#{k}=#{v}"}.join("&"))
+        elsif chemblId and arg == 'bioactivities' and params == nil
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/CHEMBL2/bioactivities
+          address("compounds/#{chemblId}/#{arg}")
+        else 
+          raise Exception, "Undefined address. ID: #{chemblId}, arg: #{arg}, params: #{params.inspect}"
+        end
+      end
+      # BioChEMBL::REST::ChEMBL_URI.compounds_stdinchikey()
+      def self.compounds_stdinchikey(stdinchikey)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/stdinchikey/QFFGVLORLPOAEC-SNVBAGLBSA-N
+        address("compounds/stdinchikey/#{stdinchikey}")
+      end 
+      # BioChEMBL::REST::ChEMBL_URI.compounds_smiles()
+      def self.compounds_smiles(smiles)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/smiles/COc1ccc2[C@@H]3[C@H](COc2c1)C(C)(C)OC4=C3C(=O)C(=O)C5=C4OC(C)(C)[C@@H]6COc7cc(OC)ccc7[C@H]56
+        address("compounds/smiles/#{smiles}")
+      end
+      # BioChEMBL::REST::ChEMBL_URI.compounds_substructure()
+      def self.compounds_substructure(smiles)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/substructure/COc1ccc2[C@@H]3[C@H](COc2c1)C(C)(C)OC4=C3C(=O)C(=O)C5=C4OC(C)(C)[C@@H]6COc7cc(OC)ccc7[C@H]56
+        address("compounds/substructure/#{smiles}")
+      end
+      # BioChEMBL::REST::ChEMBL_URI.compounds_similarity()
+      def self.compounds_similarity(smiles)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/compounds/similarity/COc1ccc2[C@@H]3[C@H](COc2c1)C(C)(C)OC4=C3C(=O)C(=O)C5=C4OC(C)(C)[C@@H]6COc7cc(OC)ccc7[C@H]56/70
+        address("compounds/similarity/#{smiles}")
+      end
+  
+      # BioChEMBL::REST::ChEMBL_URI.targets()
+      def self.targets(chemblId = nil, arg = nil)
+        if chemblId and arg == nil
+          # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/targets/CHEMBL2477
+          address("targets/#{chemblId}")
+        elsif chemblId and arg == 'bioactivities' 
+          # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/targets/CHEMBL240/bioactivities
+          address("targets/#{chemblId}/bioactivities")
+        elsif chemblId == nil and arg == nil 
+          # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/targets
+          address("targets")
+        else
+          raise Exception, "Undefined."
+        end
+      end
+      # BioChEMBL::REST::ChEMBL_URI.targets_uniprot()
+      def self.targets_uniprot(uniprot_id)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/targets/uniprot/Q13936
+        address("targets/uniprot/#{uniprot_id}")
+      end 
+      # BioChEMBL::REST::ChEMBL_URI.targets_refseq()
+      def self.targets_refseq(refseq_id)
+        # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/targets/refseq/NP_001128722
+        address("targets/refseq/#{refseq_id}")
+      end
+      
+      # BioChEMBL::REST::ChEMBL_URI.assays()
+      def self.assays(chemblId, arg = nil)
+        if chemblId and arg == nil
+          # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/assays/CHEMBL1217643
+          address("assays/#{chemblId}")
+        elsif chemblId and arg == 'bioactivities' 
+          # Example URL (XML Output): http://www.ebi.ac.uk/chemblws/assays/CHEMBL1217643/bioactivities
+          address("assays/#{chemblId}/bioactivities")
+        else
+          raise Exception, "Undefined. ChEMBL ID: #{chemblId}, arg: #{arg}"
+        end
+      end
+    end
+    
+
+    # BioChEMBL::REST.website
+    def self.website
+      "https://www.ebi.ac.uk/chembldb/index.php/ws"
+    end
+
+    # BioChEMBL::REST.usage
+    def self.usage    
+      ["a = Bio::DB::ChEMBL::REST.new",
+       "a.status => https://www.ebi.ac.uk/chembldb/index.php/ws#serviceStatus",
+       'a.compounds("CHEMBL1") => http://www.ebi.ac.uk/chemblws/compounds/CHEMBL1',
+       'a.compounds.stdinchikey("QFFGVLORLPOAEC-SNVBAGLBSA-N") => http://www.ebi.ac.uk/chemblws/compounds/stdinchikey/QFFGVLORLPOAEC-SNVBAGLBSA-N'
+       ].join("\n")
+    end
+    
+    # BioChEMBL::REST.up? #=> true/false
+    def self.up?
+      if new.status == "UP"
+        true
+      else
+        false
+      end
+    end
+
+    
+    # new
+    def initialize(uri = BASE_URI)
+      uri = URI.parse(uri) unless uri.kind_of?(URI)
+      @header = {
+        'User-Agent' => "BioChEMBL, BioRuby/#{Bio::BIORUBY_VERSION_ID}"
+      }
+      @debug = false
+    end
+
+    # If true, shows debug information to $stderr.
+    attr_accessor :debug
+
+    # get HTTP GET URL
+    def get(url)
+      easy = Curl::Easy.new(url) do |c|
+        @header.each do |k,v|
+          c.headers[k] = v
+        end
+      end 
+      easy.perform
+      easy
+    end
+    
+    # 
+    def prepare_return_value(response)
+      if @debug then
+        $stderr.puts "ChEMBL: #{response.inspect}"
+      end
+      case response.response_code 
+      when 200
+        response.body_str
+      when 400
+        raise Exception, "400 Bad request  #{response.inspect}"
+      when 404
+        raise Exception, "404 Not found   #{response.inspect}"
+      when 500
+        raise Exception, "500 Service unavailable"
+      else
+        nil
+      end
+    end
+    
+    # uri
+    def uri
+      ChEMBL_URI
+    end
+
+    # address
+    def address(path)
+      "#{BASE_URI}/#{path}"
+    end
+
+
+    
+    # API methods
+    
+    def get_body(method, args = [])
+      code = case args.size
+      when 0
+        "get(uri.#{method})"
+      when 1
+        "get(uri.#{method}(#{args[0].inspect}))"
+      when 2
+        "get(uri.#{method}(#{args[0].inspect}, #{args[1].inspect}))"
+      when 3
+        "get(uri.#{method}(#{args[0].inspect}, #{args[1].inspect}, #{args[2].inspect}))"
+      else
+        raise Exception, "method=#{method}, args=#{args.inspect}"
+      end  
+      
+      response = eval code
+      prepare_return_value(response)
+    end
+    private :get_body
+    
+    # 
+    def current_method_name
+      caller(1).first.scan(/`(.*)'/)[0][0].to_s
+    end 
+    private :current_method_name
+    
+    
+    def status
+      get_body(current_method_name)
+    end
+
+    def compounds(chemblId = nil, action = nil, params = nil)
+      get_body(current_method_name, [chemblId, action, params])
+    end
+    
+    def compounds_stdinchikey(stdinchikey)
+      get_body(current_method_name, [stdinchikey])
+    end
+     
+    def compounds_smiles(smiles)
+      get_body(current_method_name, [smiles])
+    end
+    
+    def compounds_substructure(smiles)
+      get_body(current_method_name, [smiles])
+    end
+    
+    def compounds_similarity(smiles)
+      get_body(current_method_name, [smiles])
+    end
+
+    def targets(chemblId = nil, action = nil)
+      get_body(current_method_name, [chemblId, action])
+    end
+    
+    def targets_uniprot(uniprot_id)
+      get_body(current_method_name, [uniprot_id])
+    end
+     
+    def targets_refseq(refseq_id)
+      get_body(current_method_name, [refseq_id])
+    end
+
+    def assays(chemblId = nil, action = nil)
+      get_body(current_method_name, [chemblId, action])
+    end
+  end
+
+end