fminer.rb

4 files changed, 96 insertions, 57 deletions

diff --git a/fminer.rb b/fminer.rb
index 79b0d06..b9c8b5b 100644
--- a/fminer.rb
+++ b/fminer.rb
@@ -70,8 +70,7 @@ post '/fminer/bbrc/?' do
 
   halt 404, "Please submit a dataset_uri." unless params[:dataset_uri] and  !params[:dataset_uri].nil?
   halt 404, "Please submit a prediction_feature." unless params[:prediction_feature] and  !params[:prediction_feature].nil?
-  prediction_feature = params[:prediction_feature]
-
+  prediction_feature = OpenTox::Feature.find params[:prediction_feature], @subjectid
   training_dataset = OpenTox::Dataset.find "#{params[:dataset_uri]}", @subjectid
   halt 404, "No feature #{params[:prediction_feature]} in dataset #{params[:dataset_uri]}" unless training_dataset.features and training_dataset.features.include?(params[:prediction_feature])
 
@@ -89,8 +88,13 @@ post '/fminer/bbrc/?' do
     @@bbrc.SetMinfreq(minfreq)
     @@bbrc.SetType(1) if params[:feature_type] == "paths"
     @@bbrc.SetBackbone(eval params[:backbone]) if params[:backbone] and ( params[:backbone] == "true" or params[:backbone] == "false" ) # convert string to boolean
-    @@bbrc.SetChisqSig(params[:min_chisq_significance]) if params[:min_chisq_significance]
+    @@bbrc.SetChisqSig(params[:min_chisq_significance].to_f) if params[:min_chisq_significance]
     @@bbrc.SetConsoleOut(false)
+    if prediction_feature.feature_type == "regression"
+      @@bbrc.SetRegression(true)
+    else
+      @training_classes = training_dataset.feature_classes(prediction_feature.uri, @subjectid)
+    end
 
     feature_dataset = OpenTox::Dataset.new(nil, @subjectid)
     feature_dataset.add_metadata({
@@ -122,21 +126,30 @@ post '/fminer/bbrc/?' do
         next
       end
       entry.each do |feature,values|
-        if feature == prediction_feature
+        if feature == prediction_feature.uri
           values.each do |value|
             if value.nil? 
               LOGGER.warn "No #{feature} activiity for #{compound.to_s}."
             else
-              case value.to_s
-              when "true"
-                nr_active += 1
-                activity = 1
-              when "false"
-                nr_inactive += 1
-                activity = 0
-              else
-                activity = value.to_f
-                @@bbrc.SetRegression(true)
+              if prediction_feature.feature_type == "classification"
+                case value.to_s
+                when "true"
+                  nr_active += 1
+                  activity = 1
+                when "false"
+                  nr_inactive += 1
+                  activity = 0
+                when /#{@training_classes.last}/
+                  nr_active += 1
+                  activity = 1
+                when /#{@training_classes.first}/
+                  nr_inactive += 1
+                  activity = 0
+                else
+                  LOGGER.warn "Unknown class \"#{value.to_s}\"."
+                end
+              elsif prediction_feature.feature_type == "regression"
+                activity = Math.log10(value.to_f)
               end
               begin
                 @@bbrc.AddCompound(smiles,id)
@@ -232,7 +245,7 @@ post '/fminer/last/?' do
 
   halt 404, "Please submit a dataset_uri." unless params[:dataset_uri] and  !params[:dataset_uri].nil?
   halt 404, "Please submit a prediction_feature." unless params[:prediction_feature] and  !params[:prediction_feature].nil?
-  prediction_feature = params[:prediction_feature]
+  prediction_feature = OpenTox::Feature.find params[:prediction_feature], @subjectid
   training_dataset = OpenTox::Dataset.new "#{params[:dataset_uri]}", @subjectid
   training_dataset.load_all(@subjectid)
   halt 404, "No feature #{params[:prediction_feature]} in dataset #{params[:dataset_uri]}" unless training_dataset.features and training_dataset.features.include?(params[:prediction_feature])
@@ -252,8 +265,13 @@ post '/fminer/last/?' do
     @@last.SetType(1) if params[:feature_type] == "paths"
     @@last.SetMaxHops(params[:hops]) if params[:hops]
     @@last.SetConsoleOut(false)
+    if prediction_feature.feature_type == "regression"
+      @@last.SetRegression(true)
+    else
+      @training_classes = training_dataset.feature_classes(prediction_feature.uri)
+    end
 
-    feature_dataset = OpenTox::Dataset.new
+    feature_dataset = OpenTox::Dataset.new(nil, @subjectid)
     feature_dataset.add_metadata({
       DC.title => "LAST representatives for " + training_dataset.metadata[DC.title].to_s,
       DC.creator => url_for('/fminer/last',:full),
@@ -284,28 +302,37 @@ post '/fminer/last/?' do
         next
       end
       entry.each do |feature,values|
-        if feature == prediction_feature
+        if feature == prediction_feature.uri
           values.each do |value|
             if value.nil? 
               LOGGER.warn "No #{feature} activiity for #{compound.to_s}."
             else
-              case value.to_s
-              when "true"
-                nr_active += 1
-                activity = 1
-              when "false"
-                nr_inactive += 1
-                activity = 0
-              else
+              if prediction_feature.feature_type == "classification"
+                case value.to_s
+                when "true"
+                  nr_active += 1
+                  activity = 1
+                when "false"
+                  nr_inactive += 1
+                  activity = 0
+                when /#{@training_classes.last}/
+                  nr_active += 1
+                  activity = 1
+                when /#{@training_classes.first}/
+                  nr_inactive += 1
+                  activity = 0
+                else
+                  LOGGER.warn "Unknown class \"#{value.to_s}."
+                end
+              elsif prediction_feature.feature_type == "regression"
                 activity = value.to_f
-                @@last.SetRegression(true)
               end
               begin
                 @@last.AddCompound(smiles,id)
                 @@last.AddActivity(activity, id)
                 all_activities[id]=activity # DV: insert global information
                 compounds[id] = compound
-                smi[id] = smiles # AM LAST: changed this to store SMILES.
+		smi[id] = smiles # AM LAST: changed this to store SMILES.
                 id += 1
               rescue
                 LOGGER.warn "Could not add " + smiles + "\t" + value.to_s + " to fminer"
@@ -330,7 +357,7 @@ post '/fminer/last/?' do
     end
 
     lu = LU.new                             # AM LAST: uses last-utils here
-    dom=lu.read(xml)                        # AM LAST: parse GraphML (needs hpricot, @ch: to be included in wrapper!)
+    dom=lu.read(xml)                        # AM LAST: parse GraphML 
     smarts=lu.smarts_rb(dom,'nls')          # AM LAST: converts patterns to LAST-SMARTS using msa variant (see last-pm.maunz.de)
     instances=lu.match_rb(smi,smarts)       # AM LAST: creates instantiations
     instances.each do |smarts, ids|
diff --git a/last-utils b/last-utils
-Subproject 04bd1b73f54bb7422d3c08bb5a81bc02af04f6f
+Subproject 75bea7645601fd296aa68c6678ee9b0a49a7b91
diff --git a/lazar.rb b/lazar.rb
index a4e006b..6bedf95 100644
--- a/lazar.rb
+++ b/lazar.rb
@@ -35,17 +35,17 @@ post '/lazar/?' do
   halt 404, "Dataset #{dataset_uri} not found." unless training_activities = OpenTox::Dataset.new(dataset_uri)
   training_activities.load_all(@subjectid)
 
-  prediction_feature = params[:prediction_feature]
-  unless prediction_feature # try to read prediction_feature from dataset
+  prediction_feature = OpenTox::Feature.find(params[:prediction_feature],@subjectid)
+  unless params[:prediction_feature] # try to read prediction_feature from dataset
     halt 404, "#{training_activities.features.size} features in dataset #{dataset_uri}. Please provide a  prediction_feature parameter." unless training_activities.features.size == 1
-    prediction_feature = training_activities.features.keys.first
-    params[:prediction_feature] = prediction_feature
+    prediction_feature = OpenTox::Feature.find(training_activities.features.keys.first,@subjectid)
+    params[:prediction_feature] = prediction_feature.uri # pass to feature mining service
   end
 
   feature_generation_uri = @@feature_generation_default unless feature_generation_uri = params[:feature_generation_uri]
 
-	halt 404, "No feature #{prediction_feature} in dataset #{params[:dataset_uri]}. (features: "+
-    training_activities.features.inspect+")" unless training_activities.features and training_activities.features.include?(prediction_feature)
+	halt 404, "No feature #{prediction_feature.uri} in dataset #{params[:dataset_uri]}. (features: "+
+    training_activities.features.inspect+")" unless training_activities.features and training_activities.features.include?(prediction_feature.uri)
 
   task = OpenTox::Task.create("Create lazar model",url_for('/lazar',:full)) do |task|
 
@@ -55,7 +55,7 @@ post '/lazar/?' do
 		if params[:feature_dataset_uri]
       feature_dataset_uri = params[:feature_dataset_uri]
       training_features = OpenTox::Dataset.new(feature_dataset_uri)
-      case training_features.feature_type
+      case training_features.feature_type(@subjectid)
       when "classification"
         lazar.similarity_algorithm = "Similarity.tanimoto"
       when "regression"
@@ -73,7 +73,6 @@ post '/lazar/?' do
       if prediction_feature.feature_type == "regression" && feature_generation_uri.match(/fminer/) 
         params[:feature_type] = "paths"
       end
-      prediction_feature = prediction_feature.uri  #hotfix this will change in future version see development branch 2011/04/06 mr
       feature_dataset_uri = OpenTox::Algorithm::Generic.new(feature_generation_uri).run(params).to_s
       training_features = OpenTox::Dataset.new(feature_dataset_uri)
     end
@@ -82,21 +81,24 @@ post '/lazar/?' do
 		halt 404, "Dataset #{feature_dataset_uri} not found." if training_features.nil?
 
     # sorted features for index lookups
-    lazar.features = training_features.features.sort if training_features.feature_type == "regression"
+
+    lazar.features = training_features.features.sort if prediction_feature.feature_type == "regression" and lazar.feature_calculation_algorithm != "Substructure.match"
 
     training_features.data_entries.each do |compound,entry|
       lazar.fingerprints[compound] = [] unless lazar.fingerprints[compound]
       entry.keys.each do |feature|
-        if feature_generation_uri.match(/fminer/)
-          smarts = training_features.features[feature][OT.smarts]
-          lazar.fingerprints[compound] << smarts
-          unless lazar.features.include? smarts
-            lazar.features << smarts
-            lazar.p_values[smarts] = training_features.features[feature][OT.pValue]
-            lazar.effects[smarts] = training_features.features[feature][OT.effect]
+        if lazar.feature_calculation_algorithm == "Substructure.match"
+          if training_features.features[feature]
+            smarts = training_features.features[feature][OT.smarts]
+            lazar.fingerprints[compound] << smarts
+            unless lazar.features.include? smarts
+              lazar.features << smarts
+              lazar.p_values[smarts] = training_features.features[feature][OT.pValue]
+              lazar.effects[smarts] = training_features.features[feature][OT.effect]
+            end
           end
         else
-          case training_features.feature_type
+          case training_features.feature_type(@subjectid)
           when "classification"
             # fingerprints are sets
             if entry[feature].flatten.size == 1
@@ -117,27 +119,37 @@ post '/lazar/?' do
       end
     end
       
+    @training_classes = training_activities.feature_classes(prediction_feature.uri, @subjectid) if prediction_feature.feature_type == "classification"
+    lazar.prediction_algorithm = "Neighbors.local_svm_regression" if  prediction_feature.feature_type == "regression"
+
     training_activities.data_entries.each do |compound,entry| 
 			lazar.activities[compound] = [] unless lazar.activities[compound]
-      unless entry[params[:prediction_feature]].empty?
-        entry[params[:prediction_feature]].each do |value|
-          case value.to_s
-          when "true"
-            lazar.activities[compound] << true
-          when "false"
-            lazar.activities[compound] << false
-          else 
+      unless entry[prediction_feature.uri].empty?
+        entry[prediction_feature.uri].each do |value|
+          if prediction_feature.feature_type == "classification"
+            case value.to_s
+            when "true"
+              lazar.activities[compound] << true
+            when "false"
+              lazar.activities[compound] << false
+            when /#{@training_classes.last}/
+              lazar.activities[compound] << true
+            when /#{@training_classes.first}/
+              lazar.activities[compound] << false
+            else
+              LOGGER.warn "Unknown class \"#{value.to_s}\"."
+            end
+          elsif prediction_feature.feature_type == "regression"
             halt 404, "0 values not allowed in training dataset. log10 is calculated internally." if value.to_f == 0
             lazar.activities[compound] << value.to_f
-            lazar.prediction_algorithm = "Neighbors.local_svm_regression"
           end
         end
       end
     end
 
-    lazar.metadata[DC.title] = "lazar model for #{URI.decode(File.basename(prediction_feature))}"
+    lazar.metadata[DC.title] = "lazar model for #{URI.decode(File.basename(prediction_feature.uri))}"
     # TODO: fix dependentVariable
-    lazar.metadata[OT.dependentVariables] = params[:prediction_feature]
+    lazar.metadata[OT.dependentVariables] = prediction_feature.uri
     lazar.metadata[OT.trainingDataset] = dataset_uri
 		lazar.metadata[OT.featureDataset] = feature_dataset_uri
     if training_activities.feature_type.to_s == "classification"
@@ -148,7 +160,7 @@ post '/lazar/?' do
 
     lazar.metadata[OT.parameters] = [
       {DC.title => "dataset_uri", OT.paramValue => dataset_uri},
-      {DC.title => "prediction_feature", OT.paramValue => prediction_feature},
+      {DC.title => "prediction_feature", OT.paramValue => prediction_feature.uri},
       {DC.title => "feature_generation_uri", OT.paramValue => feature_generation_uri}
     ]
 		
diff --git a/libfminer b/libfminer
-Subproject 6514520638b8309ec9bf379bce54f45d3a53419
+Subproject 2af90d80f66649901002f223fac371fc86ca03b