Merge branch 'hotfix/v1.0.3'v1.0.3

4 files changed, 176 insertions, 159 deletions

diff --git a/README.md b/README.md
index cf1f519..0eb641c 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,7 @@
 OpenTox Algorithm
 =================
 
-- An [OpenTox](http://www.opentox.org) REST Webservice 
+- An [OpenTox](http://www.opentox.org) REST Webservice
 - Implements the OpenTox algorithm API for
     - fminer
     - lazar
@@ -9,18 +9,26 @@ OpenTox Algorithm
 REST operations
 ---------------
 
-    Get a list of all algorithms  GET   /       -                       URIs of algorithms        200
-    Get a representation of the   GET   /fminer -                       fminer representation     200,404
-    fminer algorithm  
-    Get a representation of the   GET   /lazar  -                       lazar representation      200,404
-    lazar algorithm   
-    Create fminer features        POST  /fminer dataset_uri,            URI for feature dataset   200,400,404,500
-                                                feature_uri   
-    Create lazar model            POST  /lazar  dataset_uri,            URI for lazar model       200,400,404,500
-                                                prediction_feature, 
-                                                feature_generation_uri  
-
-Supported MIME formats 
+    Get a list of all algorithms  GET   /             -                       URIs of algorithms        200
+    Get a representation of the   GET   /fminer/      -                       fminer representation     200,404
+     fminer algorithms
+    Get a representation of the   GET   /fminer/bbrc  -                       bbrc representation       200,404
+    bbrc algorithm
+    Get a representation of the   GET   /fminer/last  -                       last representation       200,404
+     last algorithm
+    Get a representation of the   GET   /lazar        -                       lazar representation      200,404
+     lazar algorithm
+     Create bbrc features          POST  /fminer/bbrc dataset_uri,            URI for feature dataset   200,400,404,500
+                                                      feature_uri,
+                                                      min_frequency
+     Create last features          POST  /fminer/last dataset_uri,            URI for feature dataset   200,400,404,500
+                                                      feature_uri,
+                                                      min_frequency
+     Create lazar model            POST  /lazar       dataset_uri,            URI for lazar model       200,400,404,500
+                                                      prediction_feature,
+                                                      feature_generation_uri
+
+Supported MIME formats
 ----------------------
 
 - application/rdf+xml (default): read/write OWL-DL
@@ -29,19 +37,25 @@ Supported MIME formats
 Examples
 --------
 
-### Get the OWL-DL representation of fminer 
+### Get the OWL-DL representation of fminer
 
     curl http://webservices.in-silico.ch/algorithm/fminer
 
-### Get the OWL-DL representation of lazar 
+### Get the OWL-DL representation of lazar
 
     curl http://webservices.in-silico.ch/algorithm/lazar
 
-### Create fminer features
+### Create [BBRC](http://bbrc.maunz.de) features
 
-    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} http://webservices.in-silico.ch/algorithm/fminer
+    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d min_frequency={min_frequency} http://webservices.in-silico.ch/algorithm/fminer/bbrc
 
-(feature_uri specifies the dependent variable, e.g. http://www.epa.gov/NCCT/dsstox/CentralFieldDef.html#ActivityOutcome_CPDBAS_Hamster)
+feature_uri specifies the dependent variable from the dataset.
+
+### Create [LAST-PM](http://last-pm.maunz.de) features
+
+    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d min_frequency={min_frequency} http://webservices.in-silico.ch/algorithm/fminer/last
+
+feature_uri specifies the dependent variable from the dataset.
 
 Creates a dataset with fminer features (backbone refinement class representatives from supervised graph mining, see http://www.maunz.de/libfminer-doc/). These features can be used e.g. as structural alerts, as descriptors (fingerprints) for prediction models or for similarity calculations.
 
@@ -49,10 +63,9 @@ Creates a dataset with fminer features (backbone refinement class representative
 
     curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d feature_generation_uri=http://webservices.in-silico.ch/algorithm/fminer http://webservices.in-silico.ch/test/algorithm/lazar
 
-(feaure_uri specifies the dependent variable, e.g. http://www.epa.gov/NCCT/dsstox/CentralFieldDef.html#ActivityOutcome_CPDBAS_Hamster)
+feature_uri specifies the dependent variable from the dataset
 
 [API documentation](http://rdoc.info/github/opentox/algorithm)
 --------------------------------------------------------------
 
 Copyright (c) 2009-2011 Christoph Helma, Martin Guetlein, Micha Rautenberg, Andreas Maunz, David Vorgrimmler, Denis Gebele. See LICENSE for details.
-
diff --git a/fminer.rb b/fminer.rb
index 5cc83ed..91e73cd 100644
--- a/fminer.rb
+++ b/fminer.rb
@@ -67,148 +67,150 @@ end
 # @return [text/uri-list] Task URI
 post '/fminer/bbrc/?' do 
 
-    # TODO: is this thread safe??
-    #@@bbrc = Bbrc::Bbrc.new 
-    minfreq = 5 unless minfreq = params[:min_frequency]
-    @@bbrc.SetMinfreq(minfreq)
-    @@bbrc.SetType(1) if params[:feature_type] == "paths"
-    @@bbrc.SetBackbone(params[:backbone]) if params[:backbone]
-    @@bbrc.SetChisqSig(params[:min_chisq_significance]) if params[:min_chisq_significance]
-    @@bbrc.SetConsoleOut(false)
-
-    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]
-
-    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])
-
-    task = OpenTox::Task.create("Mining BBRC features", url_for('/fminer',:full)) do 
-
-      feature_dataset = OpenTox::Dataset.new(nil, @subjectid)
-      feature_dataset.add_metadata({
-        DC.title => "BBRC representatives for " + training_dataset.metadata[DC.title].to_s,
-        DC.creator => url_for('/fminer/bbrc',:full),
-        OT.hasSource => url_for('/fminer/bbrc', :full),
-        OT.parameters => [
-          { DC.title => "dataset_uri", OT.paramValue => params[:dataset_uri] },
-          { DC.title => "prediction_feature", OT.paramValue => params[:prediction_feature] }
-        ]
-      })
-      feature_dataset.save(@subjectid)
-
-      id = 1 # fminer start id is not 0
-      compounds = []
-      nr_active=0
-      nr_inactive=0
-      all_activities = Hash.new# DV: for effect calculation in regression part
-
-      @@bbrc.Reset
-      training_dataset.data_entries.each do |compound,entry|
-        begin
-          smiles = OpenTox::Compound.new(compound.to_s).to_smiles
-        rescue
-          LOGGER.warn "No resource for #{compound.to_s}"
-          next
-        end
-        if smiles == '' or smiles.nil?
-          LOGGER.warn "Cannot find smiles for #{compound.to_s}."
-          next
-        end
-        entry.each do |feature,values|
-          values.each do |value|
-            if value.nil? 
-              LOGGER.warn "No #{feature} activiity for #{compound.to_s}."
+  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]
+
+  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])
+
+  unless minfreq = params[:min_frequency]
+    minfreq = 5*training_dataset.compounds.size/1000 # 8 promille according to Andreas suggestions
+    minfreq = 2 unless minfreq > 2
+  end
+
+  @@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.SetConsoleOut(false)
+
+  task = OpenTox::Task.create("Mining BBRC features", url_for('/fminer',:full)) do 
+
+    feature_dataset = OpenTox::Dataset.new(nil, @subjectid)
+    feature_dataset.add_metadata({
+      DC.title => "BBRC representatives for " + training_dataset.metadata[DC.title].to_s,
+      DC.creator => url_for('/fminer/bbrc',:full),
+      OT.hasSource => url_for('/fminer/bbrc', :full),
+      OT.parameters => [
+        { DC.title => "dataset_uri", OT.paramValue => params[:dataset_uri] },
+        { DC.title => "prediction_feature", OT.paramValue => params[:prediction_feature] }
+      ]
+    })
+    feature_dataset.save(@subjectid)
+
+    id = 1 # fminer start id is not 0
+    compounds = []
+    nr_active=0
+    nr_inactive=0
+    all_activities = Hash.new# DV: for effect calculation in regression part
+
+    @@bbrc.Reset
+    training_dataset.data_entries.each do |compound,entry|
+      begin
+        smiles = OpenTox::Compound.new(compound.to_s).to_smiles
+      rescue
+        LOGGER.warn "No resource for #{compound.to_s}"
+        next
+      end
+      if smiles == '' or smiles.nil?
+        LOGGER.warn "Cannot find smiles for #{compound.to_s}."
+        next
+      end
+      entry.each do |feature,values|
+        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
-              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)
-              end
-              begin
-                @@bbrc.AddCompound(smiles,id)
-                @@bbrc.AddActivity(activity, id)
-                all_activities[id]=activity # DV: insert global information
-                compounds[id] = compound
-                id += 1
-              rescue
-                LOGGER.warn "Could not add " + smiles + "\t" + value.to_s + " to fminer"
-              end
+              activity = value.to_f
+              @@bbrc.SetRegression(true)
+            end
+            begin
+              @@bbrc.AddCompound(smiles,id)
+              @@bbrc.AddActivity(activity, id)
+              all_activities[id]=activity # DV: insert global information
+              compounds[id] = compound
+              id += 1
+            rescue
+              LOGGER.warn "Could not add " + smiles + "\t" + value.to_s + " to fminer"
             end
           end
         end
       end
+    end
 
-      g_array=all_activities.values # DV: calculation of global median for effect calculation
-      g_median=OpenTox::Algorithm.median(g_array)
-      
-      raise "No compounds in dataset #{training_dataset.uri}" if compounds.size==0
-
-      features = Set.new
-      # run @@bbrc
-      (0 .. @@bbrc.GetNoRootNodes()-1).each do |j|
-
-        results = @@bbrc.MineRoot(j)
-        results.each do |result|
-          f = YAML.load(result)[0]
-          smarts = f[0]
-          p_value = f[1]
-
-          if (!@@bbrc.GetRegression) 
-            ids = f[2] + f[3]
-            if f[2].size.to_f/ids.size > nr_active.to_f/(nr_active+nr_inactive)
-              effect = 'activating'
-            else
-              effect = 'deactivating'
-            end
-          else #regression part
-            ids = f[2]
-            # DV: effect calculation
-            f_arr=Array.new
-            f[2].each do |id|
-              f_arr.push(all_activities[id]) 
-            end 
-            f_median=OpenTox::Algorithm.median(f_arr)
-            if g_median >= f_median 
-              effect = 'activating'
-            else
-              effect = 'deactivating'
-            end
-          end
+    g_array=all_activities.values # DV: calculation of global median for effect calculation
+    g_median=OpenTox::Algorithm.median(g_array)
+    
+    raise "No compounds in dataset #{training_dataset.uri}" if compounds.size==0
+
+    features = Set.new
+    # run @@bbrc
+    (0 .. @@bbrc.GetNoRootNodes()-1).each do |j|
+
+      results = @@bbrc.MineRoot(j)
+      results.each do |result|
+        f = YAML.load(result)[0]
+        smarts = f[0]
+        p_value = f[1]
 
-          feature_uri = File.join feature_dataset.uri,"feature","bbrc", features.size.to_s
-          unless features.include? smarts
-            features << smarts
-            metadata = {
-              OT.hasSource => url_for('/fminer/bbrc', :full),
-              OT.isA => OT.Substructure,
-              OT.smarts => smarts,
-              OT.pValue => p_value.to_f,
-              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
-            #feature_dataset.add_feature_parameters feature_uri, feature_dataset.parameters
+        if (!@@bbrc.GetRegression) 
+          ids = f[2] + f[3]
+          if f[2].size.to_f/ids.size > nr_active.to_f/(nr_active+nr_inactive)
+            effect = 'activating'
+          else
+            effect = 'deactivating'
+          end
+        else #regression part
+          ids = f[2]
+          # DV: effect calculation
+          f_arr=Array.new
+          f[2].each do |id|
+            f_arr.push(all_activities[id]) 
+          end 
+          f_median=OpenTox::Algorithm.median(f_arr)
+          if g_median >= f_median 
+            effect = 'activating'
+          else
+            effect = 'deactivating'
           end
-          ids.each { |id| feature_dataset.add(compounds[id], feature_uri, true)}
         end
+
+        feature_uri = File.join feature_dataset.uri,"feature","bbrc", features.size.to_s
+        unless features.include? smarts
+          features << smarts
+          metadata = {
+            OT.hasSource => url_for('/fminer/bbrc', :full),
+            OT.isA => OT.Substructure,
+            OT.smarts => smarts,
+            OT.pValue => p_value.to_f,
+            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
+          #feature_dataset.add_feature_parameters feature_uri, feature_dataset.parameters
+        end
+        ids.each { |id| feature_dataset.add(compounds[id], feature_uri, true)}
       end
-      feature_dataset.save(@subjectid) 
-      feature_dataset.uri
     end
-    response['Content-Type'] = 'text/uri-list'
-    halt 503,task.uri+"\n" if task.status == "Cancelled"
-    halt 202,task.uri.to_s+"\n"
+    feature_dataset.save(@subjectid) 
+    feature_dataset.uri
   end
+  response['Content-Type'] = 'text/uri-list'
+  halt 503,task.uri+"\n" if task.status == "Cancelled"
+  halt 202,task.uri.to_s+"\n"
+end
 #end
 
 # Run last algorithm on a dataset
@@ -221,22 +223,24 @@ post '/fminer/bbrc/?' do
 #   - hops Maximum number of hops
 # @return [text/uri-list] Task URI
 post '/fminer/last/?' do
-  #@@last = Last::Last.new 
-  minfreq = 5 unless minfreq = params[:min_frequency]
-  @@last.SetMinfreq(minfreq)
-  @@last.SetType(1) if params[:feature_type] == "paths"
-  @@last.SetMaxHops(params[:hops]) if params[:hops]
-  @@last.SetConsoleOut(false)
 
   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]
-
   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])
 
+  unless minfreq = params[:min_frequency]
+    minfreq = 8*training_dataset.compounds.size/100 # 8% according to Andreas suggestions
+    minfreq = 2 unless minfreq > 2
+  end
+
+  @@last.SetMinfreq(minfreq)
+  @@last.SetType(1) if params[:feature_type] == "paths"
+  @@last.SetMaxHops(params[:hops]) if params[:hops]
+  @@last.SetConsoleOut(false)
+
   task = OpenTox::Task.create("Mining LAST features", url_for('/fminer',:full)) do 
 
     feature_dataset = OpenTox::Dataset.new
@@ -319,7 +323,7 @@ post '/fminer/last/?' do
 
     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!)
-    smarts=lu.smarts_rb(dom,'msa')          # AM LAST: converts patterns to LAST-SMARTS using msa variant (see last-pm.maunz.de)
+    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|
       feat_hash = Hash[*(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
diff --git a/last-utils b/last-utils
-Subproject f8a4631495309d86cb4bec5493af4110d25e4ad
+Subproject daafa32e330b27111df6dc7193a6ed72fae2be4
diff --git a/libfminer b/libfminer
-Subproject 03a3588cbea1c175d8573c2a8aad867f7a27e3e
+Subproject 01b8e50e8e6fb3ce29fc8bf0a65a8c6f6af94b3