Initial commit

2 files changed, 187 insertions, 25 deletions

diff --git a/application.rb b/application.rb
index ef123da..b00ba9f 100644
--- a/application.rb
+++ b/application.rb
@@ -38,7 +38,7 @@ end
 #
 # @return [text/uri-list] algorithm URIs
 get '/?' do
-	list = [ url_for('/lazar', :full), url_for('/fminer/bbrc', :full), url_for('/fminer/last', :full), url_for('/feature_selection/rfe', :full), url_for('/pc', :full) ].join("\n") + "\n"
+	list = [ url_for('/lazar', :full), url_for('/fminer/bbrc', :full), url_for('/fminer/bbrc/sample', :full), url_for('/fminer/last', :full), url_for('/feature_selection/rfe', :full), url_for('/pc', :full) ].join("\n") + "\n"
   case request.env['HTTP_ACCEPT']
   when /text\/html/
     content_type "text/html"
diff --git a/fminer.rb b/fminer.rb
index da06fc8..fd32043 100644
--- a/fminer.rb
+++ b/fminer.rb
@@ -4,14 +4,14 @@ ENV['FMINER_PVALUES'] = 'true'
 ENV['FMINER_SILENT'] = 'true'
 ENV['FMINER_NR_HITS'] = 'true'
 
-@@bbrc = Bbrc::Bbrc.new 
-@@last = Last::Last.new 
+@@bbrc = Bbrc::Bbrc.new
+@@last = Last::Last.new
 
 # Get list of fminer algorithms
 #
 # @return [text/uri-list] URIs of fminer algorithms
 get '/fminer/?' do
-  list = [ url_for('/fminer/bbrc', :full), url_for('/fminer/last', :full) ].join("\n") + "\n"
+  list = [ url_for('/fminer/bbrc', :full), url_for('/fminer/bbrc/sample', :full), url_for('/fminer/last', :full) ].join("\n") + "\n"
   case request.env['HTTP_ACCEPT']
   when /text\/html/
     content_type "text/html"
@@ -50,7 +50,38 @@ get "/fminer/bbrc/?" do
     content_type "application/x-yaml"
     algorithm.to_yaml
   else
-    response['Content-Type'] = 'application/rdf+xml'  
+    response['Content-Type'] = 'application/rdf+xml'
+    algorithm.to_rdfxml
+  end
+end
+
+# Get RDF/XML representation of fminer bbrc algorithm
+# @return [application/rdf+xml] OWL-DL representation of fminer bbrc algorithm
+get "/fminer/bbrc/sample?" do
+  algorithm = OpenTox::Algorithm::Generic.new(url_for('/fminer/bbrc/sample',:full))
+  algorithm.metadata = {
+    DC.title => 'fminer backbone refinement class representatives, obtained from samples of a dataset',
+    DC.creator => "andreas@maunz.de",
+#    BO.instanceOf => "http://opentox.org/ontology/ist-algorithms.owl#fminer_bbrc",
+    RDF.type => [OT.Algorithm,OTA.PatternMiningSupervised],
+    OT.parameters => [
+      { DC.description => "Dataset URI", OT.paramScope => "mandatory", DC.title => "dataset_uri" },
+      { DC.description => "Feature URI for dependent variable", OT.paramScope => "mandatory", DC.title => "prediction_feature" },
+      { DC.description => "Number of bootstrap samples", OT.paramScope => "optional", DC.title => "num_boots" },
+      { DC.description => "Minimum sampling support", OT.paramScope => "optional", DC.title => "min_sampling_support" },
+      { DC.description => "Minimum frequency", OT.paramScope => "optional", DC.title => "min_frequency" },
+      { DC.description => "Whether subgraphs should be weighted with their occurrence counts in the instances (frequency)", OT.paramScope => "optional", DC.title => "nr_hits" },
+  ]
+  }
+  case request.env['HTTP_ACCEPT']
+  when /text\/html/
+    content_type "text/html"
+    OpenTox.text_to_html algorithm.to_yaml
+  when /yaml/
+    content_type "application/x-yaml"
+    algorithm.to_yaml
+  else
+    response['Content-Type'] = 'application/rdf+xml'
     algorithm.to_rdfxml
   end
 end
@@ -81,17 +112,17 @@ get "/fminer/last/?" do
     content_type "application/x-yaml"
     algorithm.to_yaml
   else
-    response['Content-Type'] = 'application/rdf+xml'  
+    response['Content-Type'] = 'application/rdf+xml'
     algorithm.to_rdfxml
   end
 end
 
 # Creates same features for dataset <dataset_uri> that have been created
 # with fminer in dataset <feature_dataset_uri>
-# accept params[:nr_hits] as used in other fminer methods 
-post '/fminer/:method/match?' do 
+# accept params[:nr_hits] as used in other fminer methods
+post '/fminer/:method/match?' do
   raise OpenTox::BadRequestError.new "feature_dataset_uri not given" unless params[:feature_dataset_uri]
-  raise OpenTox::BadRequestError.new "dataset_uri not given" unless params[:dataset_uri] 
+  raise OpenTox::BadRequestError.new "dataset_uri not given" unless params[:dataset_uri]
   task = OpenTox::Task.create("Matching features", url_for('/fminer/match',:full)) do |task|
     f_dataset = OpenTox::Dataset.find params[:feature_dataset_uri],@subjectid
     c_dataset = OpenTox::Dataset.find params[:dataset_uri],@subjectid
@@ -105,7 +136,7 @@ post '/fminer/:method/match?' do
       f_dataset.features.each do |f,m|
         if params[:nr_hits] == "true"
           hits = comp.match_hits([m[OT.smarts]])
-          res_dataset.add(c,f,hits[m[OT.smarts]]) if hits[m[OT.smarts]]          
+          res_dataset.add(c,f,hits[m[OT.smarts]]) if hits[m[OT.smarts]]
         else
           res_dataset.add(c,f,1) if comp.match?(m[OT.smarts])
         end
@@ -115,7 +146,7 @@ post '/fminer/:method/match?' do
     res_dataset.uri
   end
   return_task(task)
-end 
+end
 
 # Run bbrc algorithm on dataset
 #
@@ -128,7 +159,7 @@ end
 #   - min_chisq_significance Significance threshold (between 0 and 1)
 #   - nr_hits Set to "true" to get hit count instead of presence
 # @return [text/uri-list] Task URI
-post '/fminer/bbrc/?' do 
+post '/fminer/bbrc/?' do
 
   fminer=OpenTox::Algorithm::Fminer.new
   fminer.check_params(params,5,@subjectid)
@@ -185,7 +216,7 @@ post '/fminer/bbrc/?' do
         smarts = f[0]
         p_value = f[1]
 
-        if (!@@bbrc.GetRegression) 
+        if (!@@bbrc.GetRegression)
           id_arrs = f[2..-1].flatten
           max = OpenTox::Algorithm.effect(f[2..-1], fminer.db_class_sizes)
           effect = f[2..-1].size-max
@@ -195,10 +226,10 @@ post '/fminer/bbrc/?' do
           f_arr=Array.new
           f[2].each do |id|
             id=id.keys[0] # extract id from hit count hash
-            f_arr.push(fminer.all_activities[id]) 
-          end 
+            f_arr.push(fminer.all_activities[id])
+          end
           f_median=f_arr.to_scale.median
-          if g_median >= f_median 
+          if g_median >= f_median
             effect = 'activating'
           else
             effect = 'deactivating'
@@ -232,13 +263,13 @@ post '/fminer/bbrc/?' do
           end
         }
 
-      end # end of 
+      end # end of
     end   # feature parsing
 
     # AM: add feature values for non-present features
-    # feature_dataset.complete_data_entries 
+    # feature_dataset.complete_data_entries
 
-    feature_dataset.save(@subjectid) 
+    feature_dataset.save(@subjectid)
     feature_dataset.uri
   end
   response['Content-Type'] = 'text/uri-list'
@@ -247,6 +278,137 @@ post '/fminer/bbrc/?' do
 end
 #end
 
+
+# Run bbrc/sample algorithm on a dataset
+#
+# @param [String] dataset_uri URI of the training dataset
+# @param [String] prediction_feature URI of the prediction feature (i.e. dependent variable)
+# @param [optional] BBRC sample parameters, accepted are
+#   - num_boots Number of bootstrap samples (default 150)
+#   - min_sampling_support Minimum sampling support (default 30% of num_boots)
+#   - min_frequency  Minimum frequency (default 10% of dataset size)
+#   - nr_hits Whether subgraphs should be weighted with their occurrence counts in the instances (frequency)
+#
+# @return [text/uri-list] Task URI
+post '/fminer/bbrc/sample/?' do
+
+  fminer=OpenTox::Algorithm::Fminer.new
+  fminer.check_params(params,80,@subjectid)
+
+  # num_boots
+  unless params[:num_boots]
+    num_boots = 150
+    LOGGER.debug "Set num_boots to default value #{num_boots}"
+  else
+    raise OpenTox::BadRequestError.new "num_boots is not numeric" unless OpenTox::Algorithm.numeric? params[:num_boots]
+	  num_boots = params[:num_boots].to_i.ceil
+  end
+
+  # min_sampling_support
+  unless params[:min_sampling_support]
+    min_sampling_support = (num_boots * 0.3).ceil
+    LOGGER.debug "Set num_boots to default value #{min_sampling_support}"
+  else
+    raise OpenTox::BadRequestError.new "min_sampling_support is not numeric" unless OpenTox::Algorithm.numeric? params[:min_sampling_support]
+	  min_sampling_support= params[:min_sampling_support].to_i.ceil
+  end
+
+  # re-set min_frequency
+  unless params[:min_frequency]
+    min_frequency = (fminer.training_dataset.compounds.size * 0.1).ceil
+    LOGGER.debug "Set min_frequency to default value #{fminer.minfreq}"
+  else
+    raise OpenTox::BadRequestError.new "min_frequency is not numeric" unless OpenTox::Algorithm.numeric? params[:min_frequency]
+	  min_frequency= params[:min_frequency].to_i.ceil
+  end
+
+    fminer.training_dataset.compounds.size
+
+  task = OpenTox::Task.create("Mining BBRC sample features", url_for('/fminer',:full)) do |task|
+    if fminer.prediction_feature.feature_type == "regression"
+      raise OpenTox::BadRequestError.new "BBRC sampling is only for classification"
+    else
+      raise "no accept values for dataset '"+fminer.training_dataset.uri.to_s+"' and feature '"+fminer.prediction_feature.uri.to_s+
+        "'" unless fminer.training_dataset.accept_values(fminer.prediction_feature.uri)
+      @value_map=fminer.training_dataset.value_map(fminer.prediction_feature.uri)
+    end
+
+    feature_dataset = OpenTox::Dataset.new(nil, @subjectid)
+    feature_dataset.add_metadata({
+      DC.title => "BBRC sampled representatives for " + fminer.training_dataset.metadata[DC.title].to_s,
+      DC.creator => url_for('/fminer/bbrc/sample',:full),
+      OT.hasSource => url_for('/fminer/bbrc/sample', :full),
+      OT.parameters => [
+        { DC.title => "dataset_uri", OT.paramValue => params[:dataset_uri] },
+        { DC.title => "prediction_feature", OT.paramValue => params[:prediction_feature] }
+        # TODO: add more params
+    ]
+    })
+    feature_dataset.save(@subjectid)
+
+    fminer.compounds = []
+    fminer.db_class_sizes = Array.new # AM: effect
+    fminer.all_activities = Hash.new # DV: for effect calculation (class and regr)
+    fminer.smi = [] # AM LAST: needed for matching the patterns back
+
+    # Add data to fminer
+    fminer.add_fminer_data(@@last, @value_map)
+
+    raise "No compounds in dataset #{fminer.training_dataset.uri}" if fminer.compounds.size==0
+
+    # run bbrc-sample
+    features = Set.new
+    task.progress 10
+
+
+    # matching
+    task.progress 90
+    lu = LU.new                             # AM LAST: uses last-utils here
+    params[:nr_hits] == "true" ? hit_count=true: hit_count=false
+    matches, counts = lu.match_rb(fminer.smi,smarts,hit_count)       # AM LAST: creates instantiations
+
+    matches.each do |smarts, ids|
+      feat_hash = Hash[*(fminer.all_activities.select { |k,v| ids.include?(k) }.flatten)] # AM LAST: get activities of feature occurrences; see http://www.softiesonrails.com/2007/9/18/ruby-201-weird-hash-syntax
+      p_value = @@last.ChisqTest(fminer.all_activities.values, feat_hash.values).to_f
+      g=Array.new
+      @value_map.each { |y,act| g[y-1]=Array.new }
+      feat_hash.each  { |x,y|   g[y-1].push(x)   }
+      max = OpenTox::Algorithm.effect(g, fminer.db_class_sizes)
+      effect = g.size-max
+      feature_uri = File.join feature_dataset.uri,"feature","last", features.size.to_s
+      unless features.include? smarts
+        features << smarts
+        metadata = {
+          RDF.type => [OT.Feature, OT.Substructure],
+          OT.hasSource => feature_dataset.uri,
+          OT.smarts => smarts,
+          OT.pValue => p_value.abs,
+          OT.effect => effect,
+          OT.parameters => [
+            { DC.title => "dataset_uri", OT.paramValue => params[:dataset_uri] },
+            { DC.title => "prediction_feature", OT.paramValue => params[:prediction_feature] }
+        ]
+        }
+        feature_dataset.add_feature feature_uri, metadata
+      end
+      if !hit_count
+        ids.each { |id| feature_dataset.add(fminer.compounds[id], feature_uri, 1)}
+      else
+        ids.each_with_index { |id,i| feature_dataset.add(fminer.compounds[id], feature_uri, counts[smarts][i])}
+      end
+    end
+
+    # AM: add feature values for non-present features
+    # feature_dataset.complete_data_entries
+
+    feature_dataset.save(@subjectid)
+    feature_dataset.uri
+  end
+  response['Content-Type'] = 'text/uri-list'
+  raise OpenTox::ServiceUnavailableError.newtask.uri+"\n" if task.status == "Cancelled"
+  halt 202,task.uri.to_s+"\n"
+end
+
 # Run last algorithm on a dataset
 #
 # @param [String] dataset_uri URI of the training dataset
@@ -312,14 +474,14 @@ post '/fminer/last/?' do
     end
 
     lu = LU.new                             # AM LAST: uses last-utils here
-    dom=lu.read(xml)                        # AM LAST: parse GraphML 
+    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)
     params[:nr_hits] == "true" ? hit_count=true: hit_count=false
     matches, counts = lu.match_rb(fminer.smi,smarts,hit_count)       # AM LAST: creates instantiations
 
     matches.each do |smarts, ids|
       feat_hash = Hash[*(fminer.all_activities.select { |k,v| ids.include?(k) }.flatten)] # AM LAST: get activities of feature occurrences; see http://www.softiesonrails.com/2007/9/18/ruby-201-weird-hash-syntax
-      if @@last.GetRegression() 
+      if @@last.GetRegression()
         p_value = @@last.KSTest(fminer.all_activities.values, feat_hash.values).to_f # AM LAST: use internal function for test
         effect = (p_value > 0) ? "activating" : "deactivating"
       else
@@ -343,20 +505,20 @@ post '/fminer/last/?' do
             { DC.title => "dataset_uri", OT.paramValue => params[:dataset_uri] },
             { DC.title => "prediction_feature", OT.paramValue => params[:prediction_feature] }
         ]
-        } 
+        }
         feature_dataset.add_feature feature_uri, metadata
       end
       if !hit_count
         ids.each { |id| feature_dataset.add(fminer.compounds[id], feature_uri, 1)}
       else
-        ids.each_with_index { |id,i| feature_dataset.add(fminer.compounds[id], feature_uri, counts[smarts][i])} 
+        ids.each_with_index { |id,i| feature_dataset.add(fminer.compounds[id], feature_uri, counts[smarts][i])}
       end
     end
 
     # AM: add feature values for non-present features
-    # feature_dataset.complete_data_entries 
+    # feature_dataset.complete_data_entries
 
-    feature_dataset.save(@subjectid) 
+    feature_dataset.save(@subjectid)
     feature_dataset.uri
   end
   response['Content-Type'] = 'text/uri-list'