support for output file; support for resnik-ic, small fixes to output…

… and error reporting.

klarigi/build.gradle

@@ -78,4 +78,4 @@ jacocoTestReport {
   }
 }
 
-version = '0.0.4'
+version = '0.0.5'

klarigi/src/main/groovy/klarigi/App.groovy

@@ -20,6 +20,7 @@ class App {
       d longOpt: 'data', 'The data describing entities and associations. See documentation for format.', args: 1
       o longOpt: 'ontology', 'The ontology to use for explanations (should be the same as the ontology used to describe patients).', args: 1
 
+      ri longOpt: 'resnik-ic', 'Use Resnik annotation frequency for information content calculation', type: Boolean
       ic longOpt: 'ic', 'List of classes and associated information content values.', args: 1
       _ longOpt: 'save-ic', 'Save the IC values to the given file', args:1
 
@@ -36,6 +37,9 @@ class App {
 
       _ longOpt: 'output-scores', 'Output the results of the scorer. This can be useful for debugging, or identifying coefficient settings.', type: Boolean
 
+      _ longOpt: 'latex', 'Output the results in LaTeX table format', type: Boolean
+      _ longOpt: 'output', 'File to output results to. If not given, will print to stdout', args: 1
+
       _ longOpt: 'verbose', 'Verbose output, mostly progress', type: Boolean, args: 0
     }
 
@@ -53,9 +57,12 @@ class App {
 
     def k = new Klarigi(o)
     if(!o['group'] || (o['group'] && o['group'] == '*')) {
-      k.explainAllClusters(o['output-scores'])
+      k.explainAllClusters(o['output-scores']).each {
+        k.output(it.cluster, it.results, o['latex'], o['output'])
+      }
     } else {
-      k.explainCluster(o['group'], o['output-scores'])
+      def r = k.explainCluster(o['group'], o['output-scores'])
+      k.output(o['group'], r, o['latex'], o['output'])
     }
   }
 }

klarigi/src/main/groovy/klarigi/InformationContent.groovy

@@ -51,6 +51,10 @@ public class InformationContent {
   private factory
 
   InformationContent(ontologyPath) {
+    this(ontologyPath, false, false)
+  }
+
+  InformationContent(ontologyPath, dataPath, annotIC) {
     factory = URIFactoryMemory.getSingleton()
 
     /*def graphURI = factory.getURI('http://purl.obolibrary.org/obo/HP_')
@@ -65,12 +69,21 @@ public class InformationContent {
 
     def gConf = new GraphConf()
     gConf.addGDataConf(dataConf)
+
+    def icMeasure = DEFAULT_IC
+    if(annotIC) {
+      gConf.addGDataConf(new GDataConf(GFormat.TSV_ANNOT, dataPath));
+    }
     //gConf.addGAction(actionRerootConf)
 
     GraphLoaderGeneric.load(gConf, graph)
     def roots = new ValidatorDAG().getTaxonomicRoots(graph)
 
     icConf = new IC_Conf_Topo(DEFAULT_IC)
+    if(annotIC) {
+      icConf = new IC_Conf_Corpus(SMConstants.FLAG_IC_ANNOT_RESNIK_1995_NORMALIZED)
+    }
+
     engine = new SM_Engine(graph)
   }
 
@@ -80,7 +93,7 @@ public class InformationContent {
       try {
         def cTerm = factory.getURI(c)
         res[c] = engine.getIC(icConf, cTerm)
-      } catch(e) { e.printStackTrace() }
+      } catch(e) { println 'Warning: ' + e.getMessage() }
     }
     res
   }

klarigi/src/main/groovy/klarigi/Klarigi.groovy

@@ -35,9 +35,13 @@ public class Klarigi {
   Klarigi(o) {
     loadData(o['data'])
     loadOntology(o['ontology'])
-    loadIc(o['ic'], o['ontology'], o['save-ic'])
+    loadIc(o['ic'], o['ontology'], o['data'], o['resnik-ic'], o['save-ic'])
     coefficients = Coefficients.Generate(o)
     verbose = o['verbose']
+
+    if(o['output']) { // blank the output file, since we will subsequently append to it. all the output stuff could probs be better abstracted.
+      new File(o['output']).text = ''
+    }
   }
 
   def loadData(dataFile) {
@@ -63,7 +67,7 @@ public class Klarigi {
     }
   }
 
-  def loadIc(icFile, ontologyFile, saveIc) {
+  def loadIc(icFile, ontologyFile, annotFile, resnikIc, saveIc) {
     if(icFile) {
       try {
       new File(icFile).splitEachLine('\t') {
@@ -74,7 +78,7 @@ public class Klarigi {
       }
     } else {
       try {
-        def icFactory = new InformationContent(ontologyFile)
+        def icFactory = new InformationContent(ontologyFile, annotFile, resnikIc)
         def allClasses = ontoHelper.reasoner.getSubClasses(ontoHelper.dataFactory.getOWLThing(), false).collect { it.getRepresentativeElement().getIRI().toString() }.unique(false)
         allClasses = allClasses.findAll { it != 'http://www.w3.org/2002/07/owl#Nothing' } // heh
         data.ic = icFactory.getInformationContent(allClasses)
@@ -145,14 +149,21 @@ public class Klarigi {
       }
     }
 
-
-    def res = StepDown.Run(coefficients, cid, candidates, data)
-    StepDown.Print(cid, res, ontoHelper.labels)
+    StepDown.Run(coefficients, cid, candidates, data)
   }
 
   def explainAllClusters(outputScores) {
-    data.groupings.each { g, v ->
-      explainCluster(g, outputScores)
+    data.groupings.collect { g, v ->
+      [ cluster: g, results: explainCluster(g, outputScores) ]
+    }
+  }
+
+  def output(cid, results, latex, toFile) {
+    def cSize = data.groupings[cid].size()
+    if(latex) {
+      StepDown.PrintLaTeX(cid, results, ontoHelper.labels, cSize, toFile)
+    } else {
+      StepDown.Print(cid, results, ontoHelper.labels, cSize, toFile)
     }
   }
 

klarigi/src/main/groovy/klarigi/StepDown.groovy

@@ -43,15 +43,38 @@ public class StepDown {
     return stepDown(candidates, c.MAX_IC, c.MAX_EXCLUSION, c.MAX_INCLUSION, c.MAX_TOTAL_INCLUSION)
   }
 
-  static def Print(cid, res, labels) {
-    println "----------------"
-    println "Group: $cid"
-    println "Overall inclusion: ${res[1]}%"
-    println "Explanatory classes:"
+  static def Print(cid, res, labels, s, toFile) {
+    def out = []
+    out << "----------------"
+    out << "Group: $cid ($s members)"
+    out << "Overall inclusion: ${res[1]}%"
+    out << "Explanatory classes:"
     res[0].each { z ->
-      println "  IRI: ${labels[z.iri]} (${z.iri}), Inclusivity: ${z.nInclusion}, Exclusivity: ${z.nExclusion}, IC: ${z.nIc}"
+      out << "  IRI: ${labels[z.iri]} (${z.iri}), Inclusivity: ${z.nInclusion}, Exclusivity: ${z.nExclusion}, IC: ${z.nIc}"
+    }
+    out << "----------------"
+    out << ""
+    out = out.join('\n')
+
+    if(toFile) {
+      new File(toFile).text += '\n' + out 
+    } else {
+      println out
+    }
+  }
+
+  static def PrintLaTeX(cid, res, labels, s, toFile) {
+    def out = []
+    out << "{\\bf Group: $cid ($s members), overall inclusivity: ${res[1].toDouble().round(2)})} & {\\bf Exclusion} & {\\bf Inclusion} & {\\bf IC} \\\\"
+    res[0].sort { -it.nIc }.each {
+      out << "${labels[it.iri]} (HP:${it.iri.tokenize('_')[1]}) & ${it.nExclusion.toDouble().round(2)} & ${it.nInclusion.toDouble().round(2)} & ${it.ic.toDouble().round(2)} \\\\"
+    }
+    out = out.join('\n')
+
+    if(toFile) {
+      new File(toFile).text += '\n' + out 
+    } else {
+      println out
     }
-    println "----------------"
-    println ""
   }
 }