Merge branch 'release/v3.2.0'

.gitmodules

@@ -3,4 +3,7 @@
 	url = git://github.com/amaunz/fminer2.git
 [submodule "last-utils"]
 	path = last-utils
-	url = git://github.com/amaunz/last-utils.git
+  url = git://github.com/amaunz/last-utils.git
+[submodule "bbrc-sample"]
+	path = bbrc-sample
+	url = git://github.com/amaunz/bbrc-sample

ChangeLog

@@ -1,3 +1,16 @@
+v4.0.0 2012-07-12
+	* bbrc-sample as submodule
+	* fminer feature datasets carry metadata (minfreq, nr_hits, ...)
+	* matching service uses now last-utils, calculates p-values
+	* fminer support for percentage and per-mil frequencies
+	* switch to opentox-ruby version 4.0.0
+
+2012-04-20
+	* Dichotomy between nominal and numeric features removed, which allows for uniform handling of all descriptors
+	* Uniform interface /pc for feature generation (for PC descriptors)
+	* Uniform interface /fs for feature selection (using recursive feature elimination)
+	* min_sim for cosine similarity corrected
+
 v3.1.0 2012-02-24
 	* lazar.rb: pc type parameter in model, cleaned all parameters,	propositionalized learning only for SVM, switch for minimal training performance, removed conf_stdev
 	* fminer.rb: feature match service for datasets, also with number of hits

README.md

@@ -3,66 +3,104 @@ OpenTox Algorithm
 
 - An [OpenTox](http://www.opentox.org) REST Webservice
 - Implements the OpenTox algorithm API for
-    - fminer
     - lazar
+    - subgraph descriptor calculation (fminer)
+    - physico-chemical descriptor calculation (pc) for more than 300 descriptors
+    - feature selection (fs) using recursive feature elimination (rfe)
+- See [opentox-ruby on maunz.de](http://opentox-ruby.maunz.de) for high-level workflow documentation
 
 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 algorithms
-    Get a representation of the   GET   /fminer/bbrc            -  bbrc representation                   200,404
+    DESCRIPTION                  TYPE  ADDRESS           ARGUMENTS                      RETURN TYPE               RETURN CODE
+    Get a representation of the  GET   /lazar            -                              lazar representation      200,404
+    lazar algorithm
+    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
-    Get a representation of the   GET   /feature_selection      -  feature selection representation      200,404
-     feature selection algorithms
-    Get a representation of the   GET   /feature_selection/rfe  -  rfe representation                    200,404
-     rfe algorithm
-
-
-    Create bbrc features          POST  /fminer/bbrc            dataset_uri,                URI for feature dataset   200,400,404,500
-                                                                feature_uri,
-                                                                [min_frequency=5 per-mil],
-                                                                [feature_type=trees],
-                                                                [backbone=true],
-                                                                [min_chisq_significance=0.95],
-                                                                [nr_hits=false]
-    Create last features          POST  /fminer/last            dataset_uri,                URI for feature dataset   200,400,404,500
-                                                                feature_uri,
-                                                                [min_frequency=8 %],
-                                                                [feature_type=trees],
-                                                                [nr_hits=false]
-    Create lazar model            POST  /lazar                  dataset_uri,                URI for lazar model       200,400,404,500
-                                                                [prediction_feature],
-                                                                [feature_generation_uri],
-                                                                [prediction_algorithm],
-                                                                [feature_dataset_uri],
-                                                                [pc_type=null],
-                                                                [nr_hits=false (class. using wt. maj. vote), true (else)],
-                                                                [min_sim=0.3 (nominal), 0.4 (numeric features)]
-                                                                [min_train_performance=0.1]
-
-    Create selected features      POST /feature_selection/rfe   dataset_uri,                URI for dataset 200,400,404,500
-                                                                prediction_feature,
-                                                                feature_dataset_uri,
-                                                                [del_missing=false]
-
+    Get a representation of the  GET   /fminer/last      -                              last representation       200,404
+    last algorithm
+    Get a representation of the  GET   /pc               -                              URIs of algorithms        200,404
+    pc algorithms
+    Get a representation of the  GET   /pc/<name>        -                              descriptor representation 200,404
+    pc algorithm <name>
+    Get a representation of the  GET   /fs               -                              URIs of algorithms        200,404
+    fs algorithms
+    Get a representation of the  GET   /fs/rfe           -                              rfe representation        200,404
+    rfe algorithm
+    Create lazar model           POST  /lazar            dataset_uri,                   URI for lazar model       200,400,404,500
+                                                         [prediction_feature],
+                                                         [feature_generation_uri],
+                                                         [feature_dataset_uri],
+                                                         [prediction_algorithm],
+                                                         [pc_type=null],
+                                                         [lib=null],
+                                                         [nr_hits=false (cl+wmv), 
+                                                           true (else)],
+                                                         [min_sim=0.3 (nominal), 0.4 
+                                                           (numeric features)],
+                                                         [min_train_performance=0.1]
+    Create bbrc features         POST  /fminer/bbrc      dataset_uri,                   URI for feature dataset   200,400,404,500
+                                                         prediction_feature,
+                                                         [min_frequency=5 per-mil],
+                                                         [feature_type=trees],
+                                                         [backbone=true],
+                                                         [min_chisq_significance=0.95],
+                                                         [nr_hits=false]
+    Create last features         POST  /fminer/last      dataset_uri,                   URI for feature dataset   200,400,404,500
+                                                         prediction_feature,
+                                                         [min_frequency=8 %],
+                                                         [feature_type=trees],
+                                                         [nr_hits=false]
+    Create features              POST /pc/AllDescriptors dataset_uri,                   URI for dataset           200,400,404,500
+                                                         [pc_type=constitutional,
+                                                         topological,geometrical,
+                                                         electronic,cpsa,hybrid],
+                                                         [lib=cdk,joelib,openbabel]
+    Create feature               POST /pc/<name>         dataset_uri                    URI for dataset           200,400,404,500
+    Select features              POST /fs/rfe            dataset_uri,                   URI for dataset           200,400,404,500
+                                                         prediction_feature,
+                                                         feature_dataset_uri,
+                                                         [del_missing=false]
 
 Synopsis
 --------
 
-- prediction\_algorithm: One of "weighted\_majority\_vote" (default for classification),  "local\_svm\_classification", "local\_svm\_regression" (default for regression). "weighted\_majority\_vote"  is not applicable for regression.
-- pc_type: Mandatory for feature dataset, one of [geometrical, topological, electronic, constitutional, hybrid, cpsa].
-- nr_hits: Whether nominal features should be instantiated with their occurrence counts in the instances. One of "true", "false". 
-- min_sim: The minimum similarity threshold for neighbors. Numeric value in [0,1].
-- min_train_performance. The minimum training performance for "local\_svm\_classification" (Accuracy) and "local\_svm\_regression" (R-squared). Numeric value in [0,1].
-- del_missing: one of true, false
+- *del_missing*: one of 
+    - *true*
+    - *false*
+
+- *feature\_type*: Type of subgraphs when no feature dataset is supplied, one of
+    - *trees*
+    - *paths*
+
+- *lib*: Mandatory for feature datasets that do not contain appropriate feature metadata, one of 
+    - *cdk*
+    - *openbabel*
+    - *joelib*
+
+- *min_sim*: The minimum similarity threshold for neighbors. Numeric value in [0,1].
+
+- *min_train_performance*. The minimum training performance for *local\_svm\_classification* (Accuracy) and *local\_svm\_regression* (R-squared). Numeric value in [0,1].
 
-See http://www.maunz.de/wordpress/opentox/2011/lazar-models-and-how-to-trigger-them for a graphical overview.
+- *nr_hits*: Whether nominal features should be instantiated with their occurrence counts in the instances. One of 
+    - *true*
+    - *false*
+
+- *pc_type*: Mandatory for feature datasets that do not contain appropriate feature metadata, one of 
+    - *geometrical*
+    - *topological* 
+    - *electronic*
+    - *constitutional*
+    - *hybrid*
+    - *cpsa*
+
+- *prediction\_algorithm*: One of 
+    - *weighted\_majority\_vote* (default for classification, n.a. for regression)
+    - *local\_svm\_classification*
+    - *local\_svm\_regression* (default for regression). 
 
 
 Supported MIME formats
@@ -76,17 +114,39 @@ Examples
 
 NOTE: http://webservices.in-silico.ch hosts the stable version that might not have complete functionality yet. **Please try http://ot-test.in-silico.ch** for latest versions.
 
+### Get the OWL-DL representation of lazar
+
+    curl http://webservices.in-silico.ch/algorithm/lazar
+
 ### 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 pc
 
-    curl http://webservices.in-silico.ch/algorithm/lazar
+    curl http://webservices.in-silico.ch/algorithm/pc
+
+### Get the OWL-DL representation of fs
+
+    curl http://webservices.in-silico.ch/algorithm/fs
 
 * * * 
 
-The following creates datasets with backbone refinement class representatives or latent structure patterns, using supervised graph mining, see http://cs.maunz.de. These features can be used e.g. as structural alerts, as descriptors (fingerprints) for prediction models or for similarity calculations.
+### Create lazar model
+
+Creates a standard Lazar model with subgraph descriptors.
+
+    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d feature_generation_uri=http://webservices.in-silico.ch/algorithm/fminer/bbrc http://webservices.in-silico.ch/test/algorithm/lazar 
+
+Creates a Lazar model with physico-chemical descriptors.
+
+    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d feature_dataset_uri={feature_dataset_uri} http://webservices.in-silico.ch/test/algorithm/lazar 
+
+feature_uri specifies the dependent variable from the dataset.
+
+* * *
+
+Creates subgraph descriptors with backbone refinement class representatives or latent structure patterns, using supervised graph mining, see http://cs.maunz.de. These features can be used e.g. as structural alerts, as descriptors (fingerprints) for prediction models or for similarity calculations.
 
 ### Create the full set of frequent and significant subtrees
 
@@ -101,30 +161,34 @@ backbone=false reduces BBRC mining to frequent and correlated subtree mining (mu
 
 feature_uri specifies the dependent variable from the dataset.   
 Adding -d nr_hits=true produces frequency counts per pattern and molecule.
-Please click [here](http://bbrc.maunz.de#usage) for more guidance on usage.
+Click [here](http://bbrc.maunz.de#usage) for more guidance on usage.
 
 ### Create [LAST-PM](http://last-pm.maunz.de) descriptors, recommended for small to medium-sized datasets.
 
     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.   
 Adding -d nr_hits=true produces frequency counts per pattern and molecule.
-Please click [here](http://last-pm.maunz.de#usage) for guidance for more guidance on usage.
+Click [here](http://last-pm.maunz.de#usage) for guidance for more guidance on usage.
 
-* * * 
 
-### Create lazar model
+* * * 
 
-Creates a standard Lazar model.
+### Create a feature dataset of physico-chemical descriptors with CDK
 
-    curl -X POST -d dataset_uri={datset_uri} -d prediction_feature={feature_uri} -d feature_generation_uri=http://webservices.in-silico.ch/algorithm/fminer/bbrc http://webservices.in-silico.ch/test/algorithm/lazar 
+    curl -X POST -d dataset_uri={dataset_uri} -d lib=cdk http://webservices.in-silico.ch/test/algorithm/pc/AllDescriptors
 
-[API documentation](http://rdoc.info/github/opentox/algorithm)
---------------------------------------------------------------
+lib specifies the library to use.
 
 * * *
 
-### Create a feature dataset of selected features
-    curl -X POST -d dataset_uri={dataset_uri} -d prediction_feature_uri={prediction_feature_uri} -d feature_dataset_uri={feature_dataset_uri} -d del_missing=true http://webservices.in-silico.ch/test/algorithm/feature_selection/rfe
+### Select features from a feature dataset
+
+    curl -X POST -d dataset_uri={dataset_uri} -d prediction_feature={feature_uri} -d feature_dataset_uri={feature_dataset_uri} http://webservices.in-silico.ch/test/algorithm/fs/rfe
+
+feature_uri specifies the dependent variable from the dataset.   
+
+* * *
 
 Copyright (c) 2009-2011 Christoph Helma, Martin Guetlein, Micha Rautenberg, Andreas Maunz, David Vorgrimmler, Denis Gebele. See LICENSE for details.
+

application.rb

@@ -1,17 +1,33 @@
+# Java Klimbim
+ENV["JAVA_HOME"] = "/usr/lib/jvm/java-6-sun" unless ENV["JAVA_HOME"]
+ENV["JOELIB2"] = File.join File.expand_path(File.dirname(__FILE__)),"java"
+deps = []
+deps << "#{ENV["JAVA_HOME"]}/lib/tools.jar"
+deps << "#{ENV["JAVA_HOME"]}/lib/classes.jar"
+deps << "#{ENV["JOELIB2"]}"
+jars = Dir[ENV["JOELIB2"]+"/*.jar"].collect {|f| File.expand_path(f) }
+deps = deps + jars
+ENV["CLASSPATH"] = deps.join(":")
+
 require 'rubygems'
-# AM LAST: can include both libs, no problems
-require File.join(File.expand_path(File.dirname(__FILE__)), 'libfminer/libbbrc/bbrc') # has to be included before openbabel, otherwise we have strange SWIG overloading problems
-require File.join(File.expand_path(File.dirname(__FILE__)), 'libfminer/liblast/last') # has to be included before openbabel, otherwise we have strange SWIG overloading problems
-require File.join(File.expand_path(File.dirname(__FILE__)), 'last-utils/lu.rb') # AM LAST
-gem "opentox-ruby", "~> 3"
+
+
+# fminer libs to be included before openbabel, otherwise strange SWIG overloading problems
+require File.join(File.expand_path(File.dirname(__FILE__)), 'libfminer/libbbrc/bbrc') 
+require File.join(File.expand_path(File.dirname(__FILE__)), 'libfminer/liblast/last')
+require File.join(File.expand_path(File.dirname(__FILE__)), 'last-utils/lu.rb')
+
+gem "opentox-ruby", "~> 4"
 require 'opentox-ruby'
+require 'rjb'
+require 'rinruby'
+
 
-#require 'smarts.rb'
-#require 'similarity.rb'
-require 'openbabel.rb'
+# main
 require 'fminer.rb'
 require 'lazar.rb'
-require 'feature_selection.rb'
+require 'fs.rb'
+require 'pc.rb'
 
 set :lock, true
 
@@ -23,7 +39,7 @@
 #
 # @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) ].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('/fminer/bbrc/match', :full), url_for('/fminer/last/match', :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"