physchem calculation and storage in compouds

3 files changed, 34 insertions, 29 deletions

diff --git a/lib/compound.rb b/lib/compound.rb
index 4ea4db4..8c11831 100644
--- a/lib/compound.rb
+++ b/lib/compound.rb
@@ -17,7 +17,6 @@ module OpenTox
     field :smiles, type: String
     field :inchikey, type: String
     field :names, type: Array
-    #field :warnings, type: Array, default: []
     field :cid, type: String
     field :chemblid, type: String
     field :png_id, type: BSON::ObjectId
@@ -88,17 +87,26 @@ module OpenTox
       fingerprints[type]
     end
 
-    def physchem descriptor_ids
-      calculated_descriptor_ids = self[:physchem_descriptors].keys
-      p names
-      new = UNIQUEDESCRIPTORS-names
-      p new
-      d = self.physchem(self, new)
-      #p d
-      #self[:physchem_descriptors].merge! d
-      self.update_attribute(:physchem_descriptors, self[:physchem_descriptors].merge(d))
+    def physchem descriptors=PhysChem.openbabel_descriptors
+      # TODO: speedup java descriptors
+      calculated_ids = physchem_descriptors.keys
+      # BSON::ObjectId instances are not allowed as keys in a BSON document.
+      new_ids = descriptors.collect{|d| d.id.to_s} - calculated_ids
+      descs = {}
+      algos = {}
+      new_ids.each do |id|
+        descriptor = PhysChem.find id
+        descs[[descriptor.library, descriptor.descriptor]]  = descriptor
+        algos[descriptor.name] = descriptor
+      end
+      # avoid recalculating Cdk features with multiple values
+      descs.keys.uniq.each do |k|
+        descs[k].send(k[0].downcase,k[1],self).each do |n,v|
+          physchem_descriptors[algos[n].id.to_s] = v # BSON::ObjectId instances are not allowed as keys in a BSON document.
+        end
+      end
       save
-      self[:physchem_descriptors]
+      physchem_descriptors
     end
 
     # Create a compound from smiles string
diff --git a/lib/physchem.rb b/lib/physchem.rb
index 1126e69..64018ad 100644
--- a/lib/physchem.rb
+++ b/lib/physchem.rb
@@ -37,10 +37,12 @@ module OpenTox
 
     DESCRIPTORS = OBDESCRIPTORS.merge(CDKDESCRIPTORS.merge(JOELIBDESCRIPTORS))
 
+
     require_relative "unique_descriptors.rb"
 
-    def self.descriptors
-      DESCRIPTORS.collect do |name,description|
+    def self.descriptors desc=DESCRIPTORS
+      # TODO create PhysChem features @startup
+      desc.collect do |name,description|
         lib,desc = name.split('.',2)
         self.find_or_create_by(:name => name, :library => lib, :descriptor => desc, :description => description, :measured => false, :calculated => true, :numeric => true, :nominal => false)
       end
@@ -64,25 +66,20 @@ module OpenTox
       udesc
     end
 
-    # Description of available descriptors
-    def self.description descriptor
-      lib = descriptor.split('_').first
-      case lib
-      when "Openbabel"
-        OBDESCRIPTORS[descriptor]
-      when "Cdk"
-        name = descriptor.split('_')[0..-2].join('_')
-        CDKDESCRIPTORS[name]
-      when "Joelib"
-        JOELIBDESCRIPTORS[descriptor]
-      when "lookup"
-        "Read feature values from a dataset"
-      end
+    def self.openbabel_descriptors
+      descriptors OBDESCRIPTORS
+    end
+
+    def self.cdk_descriptors
+      descriptors CDKDESCRIPTORS
+    end
+
+    def self.joelib_descriptors
+      descriptors JOELIBDESCRIPTORS
     end
 
     def calculate compound
       result = send library.downcase,descriptor,compound
-      p result
       result[self.name]
     end
 
diff --git a/lib/unique_descriptors.rb b/lib/unique_descriptors.rb
index 03a9b08..8341a67 100644
--- a/lib/unique_descriptors.rb
+++ b/lib/unique_descriptors.rb
@@ -24,7 +24,7 @@ UNIQUEDESCRIPTORS = [
   "Cdk.ALOGP", #Calculates atom additive logP and molar refractivity values as described by Ghose and Crippen and
   "Cdk.APol", #Descriptor that calculates the sum of the atomic polarizabilities (including implicit hydrogens).
   "Cdk.AcidicGroupCount", #Returns the number of acidic groups.
-  "Cdk.AminoAcidCount", #Returns the number of amino acids found in the system
+  #"Cdk.AminoAcidCount", #Returns the number of amino acids found in the system
   #"Cdk.AromaticAtomsCount", #Descriptor based on the number of aromatic atoms of a molecule.
   #"Cdk.AromaticBondsCount", #Descriptor based on the number of aromatic bonds of a molecule.
   #"Cdk.AtomCount", #Descriptor based on the number of atoms of a certain element type.