added initial plugin n.b.ds.liblinear and n.b.liblinear

bioclipse · Jan 14, 2015 · 0f1bb57 · 0f1bb57
1 parent 2d08519
commit 0f1bb57
Show file tree

Hide file tree

Showing 11 changed files with 900 additions and 0 deletions.
diff --git a/plugins/net.bioclipse.ds.liblinear/.classpath b/plugins/net.bioclipse.ds.liblinear/.classpath
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+	<classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.7"/>
+	<classpathentry kind="con" path="org.eclipse.pde.core.requiredPlugins"/>
+	<classpathentry kind="src" path="src"/>
+	<classpathentry kind="output" path="bin"/>
+</classpath>
diff --git a/plugins/net.bioclipse.ds.liblinear/.project b/plugins/net.bioclipse.ds.liblinear/.project
@@ -0,0 +1,28 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+	<name>net.bioclipse.ds.liblinear</name>
+	<comment></comment>
+	<projects>
+	</projects>
+	<buildSpec>
+		<buildCommand>
+			<name>org.eclipse.jdt.core.javabuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
+		<buildCommand>
+			<name>org.eclipse.pde.ManifestBuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
+		<buildCommand>
+			<name>org.eclipse.pde.SchemaBuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
+	</buildSpec>
+	<natures>
+		<nature>org.eclipse.pde.PluginNature</nature>
+		<nature>org.eclipse.jdt.core.javanature</nature>
+	</natures>
+</projectDescription>
diff --git a/plugins/net.bioclipse.ds.liblinear/META-INF/MANIFEST.MF b/plugins/net.bioclipse.ds.liblinear/META-INF/MANIFEST.MF
@@ -0,0 +1,26 @@
+Manifest-Version: 1.0
+Bundle-ManifestVersion: 2
+Bundle-Name: DS Liblinear
+Bundle-SymbolicName: net.bioclipse.ds.liblinear
+Bundle-Version: 0.1.1
+Bundle-RequiredExecutionEnvironment: JavaSE-1.7
+Require-Bundle: org.eclipse.ui,
+ org.eclipse.core.runtime,
+ org.eclipse.help,
+ org.eclipse.core.resources,
+ net.bioclipse.ds,
+ net.bioclipse.liblinear,
+ net.bioclipse.cdk.business,
+ net.bioclipse.core,
+ net.bioclipse.ds.signatures,
+ org.openscience.cdk.interfaces,
+ org.openscience.cdk.standard,
+ org.openscience.cdk.io,
+ net.bioclipse.ds.common,
+ net.bioclipse.cdk.ui.sdfeditor,
+ signatures,
+ org.openscience.cdk.valencycheck,
+ net.bioclipse.ds.libsvm;bundle-version="2.6.2"
+Import-Package: org.apache.log4j,
+ org.slf4j
+Export-Package: net.bioclipse.ds.liblinear
diff --git a/plugins/net.bioclipse.ds.liblinear/build.properties b/plugins/net.bioclipse.ds.liblinear/build.properties
@@ -0,0 +1,4 @@
+source.. = src/
+output.. = bin/
+bin.includes = META-INF/,\
+               .
diff --git a/plugins/net.bioclipse.ds.liblinear/src/net/bioclipse/ds/liblinear/SignLibLinearModel.java b/plugins/net.bioclipse.ds.liblinear/src/net/bioclipse/ds/liblinear/SignLibLinearModel.java
@@ -0,0 +1,120 @@
+package net.bioclipse.ds.liblinear;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+
+import org.apache.log4j.Logger;
+
+import de.bwaldvogel.liblinear.Feature;
+import de.bwaldvogel.liblinear.FeatureNode;
+import de.bwaldvogel.liblinear.Linear;
+import de.bwaldvogel.liblinear.Model;
+import libsvm.svm;
+import libsvm.svm_model;
+import libsvm.svm_node;
+
+
+/**
+ * A SignaturesLibLinearModel. Consists of a Model and a list of signatures.
+ * 
+ * @author ola
+ *
+ */
+public class SignLibLinearModel {
+
+	private static final Logger logger = Logger.getLogger(SignLibLinearModel.class);
+
+	List<String> modelSignatures;
+	Model model;
+
+	public SignLibLinearModel(Model model, List<String> signatures) {
+		super();
+		this.model = model;
+		this.modelSignatures = signatures;
+	}
+
+
+	public List<String> getModelSignatures() {
+		return modelSignatures;
+	}
+	public void setModelSignatures(List<String> modelSignatures) {
+		this.modelSignatures = modelSignatures;
+	}
+	public Model getModel() {
+		return model;
+	}
+	public void setModel(Model model) {
+		this.model = model;
+	}
+
+
+	/**
+	 * Wrap a standard internal LibLinear prediction.
+	 * 
+	 * @return
+	 */
+	public double predict(Feature[] instance){
+		return Linear.predict(model, instance);
+	}
+
+
+
+	/**
+	 * For a given set of signatures, predict the results for this model.
+	 * 
+	 * @param querySignatures
+	 * @return
+	 */
+	public double predict(List<String> querySignatures){
+
+
+		// Contains the signatures for a molecule and the count. 
+		//We store the count as a double although it is an integer. libsvm wants a double.
+		Map<String, Double> moleculeSignatures = new HashMap<String, Double>(); 
+
+		Iterator<String> signsIter = querySignatures.iterator();
+		int signsIndex = 0;
+		while (signsIter.hasNext()){
+			String currentSignature = signsIter.next();
+			if (modelSignatures.contains(currentSignature)){
+				if (moleculeSignatures.containsKey(currentSignature)){
+					moleculeSignatures.put(currentSignature, (Double)moleculeSignatures.get(currentSignature)+1.00);
+				}
+				else{
+					moleculeSignatures.put(currentSignature, 1.0);
+				}
+			}
+			signsIndex++;
+		}
+
+//		logger.debug("Signatures with count: " + moleculeSignatures.toString());
+
+		//Set up input for LibLinear
+		List<FeatureNode> nodes = new ArrayList<FeatureNode>();
+		Iterator<String> signaturesIter = modelSignatures.iterator();
+		while (signaturesIter.hasNext()){
+			String currentSignature = signaturesIter.next();
+			if (moleculeSignatures.containsKey(currentSignature)){
+				int index = modelSignatures.indexOf(currentSignature)+1; // libsvm assumes that the index starts at one.
+				Double value = (Double) moleculeSignatures.get(currentSignature);
+				FeatureNode node = new FeatureNode(index, value);
+				nodes.add(node);
+
+//				System.out.println(currentSignature + " ==> " 
+//						+ node.index + "=" + node.value);
+
+			}
+		}
+
+		//This is the array with signature descriptor values
+		Feature[] instance = nodes.toArray(new Feature[0]);
+
+		return predict(instance);
+	}
+
+
+
+}