From 5d4ab5553e1928e573aa6671ba6a273308b24e21 Mon Sep 17 00:00:00 2001
From: "helma@in-silico.ch" <helma@in-silico.ch>
Date: Wed, 10 Oct 2018 14:58:44 +0200
Subject: more differentiated feature classes

---
 lib/feature.rb  | 62 ++++++++++++++++++++++++++++++++++++++-------------------
 lib/lazar.rb    |  1 -
 lib/physchem.rb |  8 ++++----
 test/feature.rb |  6 +++---
 4 files changed, 48 insertions(+), 29 deletions(-)

diff --git a/lib/feature.rb b/lib/feature.rb
index 1d18a00..e6fede6 100644
--- a/lib/feature.rb
+++ b/lib/feature.rb
@@ -2,23 +2,11 @@ module OpenTox
 
   # Basic feature class
   class Feature
-    field :measured, type: Boolean
-    field :calculated, type: Boolean
-    field :category, type: String
-    field :unit, type: String
-    field :conditions, type: Hash
-
-    # Is it a nominal feature
-    # @return [TrueClass,FalseClass]
-    def nominal?
-      self.class == NominalFeature
-    end
+  end
 
-    # Is it a numeric feature
-    # @return [TrueClass,FalseClass]
-    def numeric?
-      self.class == NumericFeature
-    end
+  # Original ID (e.g. from CSV input)
+  class OriginalId < Feature
+    field :dataset_id, type: BSON::ObjectId
   end
 
   # Feature for categorical variables
@@ -28,10 +16,46 @@ module OpenTox
 
   # Feature for quantitative variables
   class NumericFeature < Feature
+    field :unit, type: String
+  end
+
+  # Nominal biological activity
+  class NominalBioActivity < NominalFeature
+    field :original_feature_id, type: BSON::ObjectId
+    field :transformation, type: Hash
+  end
+
+  # Numeric biological activity
+  class NumericBioActivity < NumericFeature
+    field :original_feature_id, type: BSON::ObjectId
+    field :transformation, type: String
+  end
+
+  # Nominal lazar prediction
+  class NominalLazarPrediction < NominalFeature
+    field :model_id, type: BSON::ObjectId
+    field :training_feature_id, type: BSON::ObjectId
+  end
+
+  # Numeric lazar prediction
+  class NumericLazarPrediction < NumericFeature
+    field :model_id, type: BSON::ObjectId
+    field :training_feature_id, type: BSON::ObjectId
+  end
+
+  class NominalSubstanceProperty < NominalFeature
+  end
+
+  class NumericSubstanceProperty < NumericFeature
+  end
+
+  class NanoParticleProperty < NumericSubstanceProperty
+    field :category, type: String
+    field :conditions, type: Hash
   end
 
   # Feature for SMARTS fragments
-  class Smarts < NominalFeature
+  class Smarts < Feature
     field :smarts, type: String 
     index "smarts" => 1
     # Create feature from SMARTS string
@@ -42,8 +66,4 @@ module OpenTox
     end
   end
 
-  class OriginalId < Feature
-    field :dataset_id, type: BSON::ObjectId
-  end
-
 end
diff --git a/lib/lazar.rb b/lib/lazar.rb
index 32f0317..d032282 100644
--- a/lib/lazar.rb
+++ b/lib/lazar.rb
@@ -94,6 +94,5 @@ CLASSES = ["Feature","Substance","Dataset","LazarPrediction","CrossValidation","
   "train-test-validation.rb",
   "leave-one-out-validation.rb",
   "crossvalidation.rb",
-  #"experiment.rb",
   "import.rb",
 ].each{ |f| require_relative f }
diff --git a/lib/physchem.rb b/lib/physchem.rb
index 07df867..2af043b 100644
--- a/lib/physchem.rb
+++ b/lib/physchem.rb
@@ -1,7 +1,7 @@
 module OpenTox
 
   # Feature for physico-chemical descriptors
-  class PhysChem < NumericFeature
+  class PhysChem < NumericSubstanceProperty
 
     field :library, type: String
     field :descriptor, type: String
@@ -45,7 +45,7 @@ module OpenTox
     def self.descriptors desc=DESCRIPTORS
       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)
+        self.find_or_create_by(:name => name, :library => lib, :descriptor => desc, :description => description)
       end
     end
 
@@ -59,11 +59,11 @@ module OpenTox
           CDK_DESCRIPTIONS.select{|d| desc == d[:java_class].split('.').last.sub('Descriptor','') }.first[:names].each do |n|
             dname = "#{name}.#{n}"
             description = DESCRIPTORS[dname]
-            udesc << self.find_or_create_by(:name => dname, :library => lib, :descriptor => desc, :description => description, :measured => false, :calculated => true)
+            udesc << self.find_or_create_by(:name => dname, :library => lib, :descriptor => desc, :description => description)
           end
         else
           description = DESCRIPTORS[name]
-          udesc << self.find_or_create_by(:name => name, :library => lib, :descriptor => desc, :description => description, :measured => false, :calculated => true)
+          udesc << self.find_or_create_by(:name => name, :library => lib, :descriptor => desc, :description => description)
         end
       end
       udesc
diff --git a/test/feature.rb b/test/feature.rb
index 85ce588..c2bdbb5 100644
--- a/test/feature.rb
+++ b/test/feature.rb
@@ -55,15 +55,15 @@ class FeatureTest < MiniTest::Test
   end
 
   def test_physchem_description
-    assert_equal 346, PhysChem.descriptors.size
-    assert_equal 15, PhysChem.openbabel_descriptors.size
+    assert_equal 347, PhysChem.descriptors.size
+    assert_equal 16, PhysChem.openbabel_descriptors.size
     assert_equal 286, PhysChem.cdk_descriptors.size
     assert_equal 45, PhysChem.joelib_descriptors.size
     assert_equal 309, PhysChem.unique_descriptors.size
   end
 
   def test_physchem
-    assert_equal 346, PhysChem.descriptors.size
+    assert_equal 347, PhysChem.descriptors.size
     c = Compound.from_smiles "CC(=O)CC(C)C"
     logP = PhysChem.find_or_create_by :name => "Openbabel.logP"
     assert_equal 1.6215, c.calculate_properties([logP]).first
-- 
cgit v1.2.3