Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[3396] trunk/scripts/src/main/java/org/dllearner/ examples/pdb
|
From: <km...@us...> - 2011-11-11 15:05:35
|
Revision: 3396
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3396&view=rev
Author: kmpf
Date: 2011-11-11 15:05:29 +0000 (Fri, 11 Nov 2011)
Log Message:
-----------
Untested workaround to get the positive/negative class mapping correct.
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/examples/pdb/AminoAcids.java
trunk/scripts/src/main/java/org/dllearner/examples/pdb/HelixRDFCreator.java
trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java
Modified: trunk/scripts/src/main/java/org/dllearner/examples/pdb/AminoAcids.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/examples/pdb/AminoAcids.java 2011-11-11 09:22:08 UTC (rev 3395)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/AminoAcids.java 2011-11-11 15:05:29 UTC (rev 3396)
@@ -117,29 +117,29 @@
return resourceString;
}
- public static HashMap<Resource, String> getAminoAcidArffAttributeMap(){
- HashMap<Resource, String> resdata = new HashMap<Resource, String>(30);
- resdata.put(ALA, new String("2,0,0.5,?,?"));
- resdata.put(CYS, new String("1,0,1,?,0"));
- resdata.put(ASP, new String("0,-1,1,?,-1"));
- resdata.put(GLU, new String("0,-1,2,?,-1"));
- resdata.put(PHE, new String("2,0,2,1,?"));
- resdata.put(GLY, new String("2,0,0.5,?,?"));
- resdata.put(HIS, new String("1,1,2,1,1"));
- resdata.put(ILE, new String("2,0,2,0,?"));
- resdata.put(LYS, new String("1,1,2,?,1"));
- resdata.put(LEU, new String("2,0,2,0,?"));
- resdata.put(MET, new String("2,0,2,?,?"));
- resdata.put(ASN, new String("0,0,1,?,0"));
- resdata.put(PRO, new String("?,0,1,?,?"));
- resdata.put(GLN, new String("0,0,2,?,0"));
- resdata.put(ARG, new String("0,1,2,?,1"));
- resdata.put(SER, new String("0,0,0.5,?,0"));
- resdata.put(THR, new String("1,0,1,?,0,"));
- resdata.put(VAL, new String("2,0,1,0,?"));
- resdata.put(TRP, new String("1,0,2,1,1"));
- resdata.put(TYR, new String("1,0,2,1,0"));
- resdata.put(SEL, new String("?,?,?,?,?"));
+ public static HashMap<String, String> getAminoAcidArffAttributeMap(){
+ HashMap<String, String> resdata = new HashMap<String, String>(30);
+ resdata.put(new String("A"), new String("2,0,0.5,?,?"));
+ resdata.put(new String("C"), new String("1,0,1,?,0"));
+ resdata.put(new String("D"), new String("0,-1,1,?,-1"));
+ resdata.put(new String("E"), new String("0,-1,2,?,-1"));
+ resdata.put(new String("F"), new String("2,0,2,1,?"));
+ resdata.put(new String("G"), new String("2,0,0.5,?,?"));
+ resdata.put(new String("H"), new String("1,1,2,1,1"));
+ resdata.put(new String("I"), new String("2,0,2,0,?"));
+ resdata.put(new String("K"), new String("1,1,2,?,1"));
+ resdata.put(new String("L"), new String("2,0,2,0,?"));
+ resdata.put(new String("M"), new String("2,0,2,?,?"));
+ resdata.put(new String("N"), new String("0,0,1,?,0"));
+ resdata.put(new String("P"), new String("?,0,1,?,?"));
+ resdata.put(new String("Q"), new String("0,0,2,?,0"));
+ resdata.put(new String("R"), new String("0,1,2,?,1"));
+ resdata.put(new String("S"), new String("0,0,0.5,?,0"));
+ resdata.put(new String("T"), new String("1,0,1,?,0,"));
+ resdata.put(new String("V"), new String("2,0,1,0,?"));
+ resdata.put(new String("W"), new String("1,0,2,1,1"));
+ resdata.put(new String("Y"), new String("1,0,2,1,0"));
+ resdata.put(new String("U"), new String("?,?,?,?,?"));
return resdata;
}
Modified: trunk/scripts/src/main/java/org/dllearner/examples/pdb/HelixRDFCreator.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/examples/pdb/HelixRDFCreator.java 2011-11-11 09:22:08 UTC (rev 3395)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/HelixRDFCreator.java 2011-11-11 15:05:29 UTC (rev 3396)
@@ -156,6 +156,7 @@
_logger.info("PDB ID: " + protein.getPdbID());
_logger.info("chain ID: " + protein.getChainID());
+
trainmodel = new PDBIdRdfModel(protein);
if (fasta){
@@ -491,13 +492,55 @@
out.println(data);
// HashMap containing information about the properties of every amino acid
- HashMap<Resource, String> resdata = AminoAcids.getAminoAcidArffAttributeMap();
+ HashMap<String, String> resdata = AminoAcids.getAminoAcidArffAttributeMap();
ArrayList<Resource> positives = model.getPositives();
ArrayList<Resource> negatives = model.getNegatives();
Property type = ResourceFactory.createProperty("http://www.w3.org/1999/02/22-rdf-syntax-ns#", "type");
Property iib = ResourceFactory.createProperty("http://bio2rdf.org/pdb:", "isImmediatelyBefore");
-
+ String sequence = protein.getSequence();
+ HashMap<Integer, Resource> posRes = model.getPositionResource();
+ for ( int i = 0; i < sequence.length(); i++) {
+ StringBuffer dataLine = new StringBuffer("");
+ String key = Character.toString( sequence.charAt(i) );
+
+ // add amino acid description to dataLine
+ if ( resdata.containsKey(key) ){
+ dataLine.append( resdata.get(key) + "," );
+ } else {
+ //
+ dataLine.append( resdata.get("U") + "," );
+ }
+
+ // add information about neighbouring amino acids to dataLine
+ for (int j = (i - 8); j <= (i + 8) ; j++){
+ try {
+ dataLine.append( protein.getSequence().charAt(j) + "," );
+ } catch (IndexOutOfBoundsException e) {
+ dataLine.append( "?," );
+ }
+ }
+
+ // add information about positive or negative to dataLine
+ if (positives.contains( posRes.get( new Integer(i) ))){
+ dataLine.append( "1" );
+ } else if (negatives.contains( posRes.get( new Integer(i) ))){
+ dataLine.append( "0" );
+ } else {
+ dataLine.append( "?" );
+ }
+
+ _logger.info(dataLine);
+ out.println(dataLine);
+
+ }
+
+ } catch (FileNotFoundException e){
+ e.printStackTrace();
+ }
+
+
+/* // to be exchanged
while (firstAAs.hasNext()){
Resource firstAA = firstAAs.next();
Resource currentAA = firstAA;
@@ -529,6 +572,8 @@
dataLine.append( "1" );
} else if (negatives.contains(currentAA)){
dataLine.append( "0" );
+ } else {
+ dataLine.append( "?" );
}
@@ -537,12 +582,6 @@
if (model.getModel().contains(currentAA, iib)){
nextAA = model.getModel().getProperty(currentAA, iib).getResource();
}
- _logger.info(dataLine);
- out.println(dataLine);
- }
- }
- } catch (FileNotFoundException e){
- e.printStackTrace();
- }
+*/
}
}
Modified: trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java 2011-11-11 09:22:08 UTC (rev 3395)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java 2011-11-11 15:05:29 UTC (rev 3396)
@@ -5,7 +5,13 @@
import java.io.IOException;
import java.io.PrintStream;
import java.util.ArrayList;
+import java.util.HashMap;
+import org.apache.log4j.ConsoleAppender;
+import org.apache.log4j.FileAppender;
+import org.apache.log4j.Level;
+import org.apache.log4j.Logger;
+import org.apache.log4j.SimpleLayout;
import org.xml.sax.InputSource;
import com.dumontierlab.pdb2rdf.model.PdbRdfModel;
@@ -27,13 +33,17 @@
public class PDBIdRdfModel {
+ private static Logger _logger = Logger.getRootLogger();
+
private PdbRdfModel _pdbIdModel = new PdbRdfModel();
private PdbRdfModel _removedFromModel = new PdbRdfModel();
private PDBProtein _protein = null ;
private ArrayList<Resource> _positives = null;
private ArrayList<Resource> _negatives = null;
+ private HashMap<Integer, Resource> _positionResource = null;
- public PDBIdRdfModel (PDBProtein protein){
+ public PDBIdRdfModel (PDBProtein protein) {
+
this._protein = protein;
this._pdbIdModel = this.getPdbRdfModel();
this.getProtein().setSequence(extractSequence(_pdbIdModel));
@@ -41,6 +51,7 @@
this.getProtein().setSpecies(extractSpecies(_pdbIdModel));
System.out.println("Species: " + this.getProtein().getSpecies());
createPositivesAndNegatives();
+ _positionResource = createPositionResidueMap();
}
public PdbRdfModel getModel(){
@@ -58,6 +69,10 @@
public ArrayList<Resource> getNegatives(){
return _negatives;
}
+
+ public HashMap<Integer, Resource> getPositionResource(){
+ return _positionResource;
+ }
private PdbRdfModel getPdbRdfModel() {
String[] pdbIDs = {_protein.getPdbID()};
@@ -192,13 +207,15 @@
"PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> " +
"PREFIX fn: <http://www.w3.org/2005/xpath-functions#> " +
"CONSTRUCT { ?x1 pdb:beginsAt ?x2 ." +
- " ?x1 pdb:endsAt ?x3 . " +
- " ?x5 dcterms:isPartOf ?x4 . " +
+ " ?x1 pdb:endsAt ?x3 ." +
+ " ?x5 dcterms:isPartOf ?x4 ." +
" ?x5 rdf:type ?x6 ." +
" ?x5 pdb:isImmediatelyBefore ?x7 ." +
+ " ?x5 pdb:hasChainPosition ?x8 ." +
+ " ?x8 rdfs:label ?residuePosition ." +
" ?organism rdfs:label ?organismName ." +
" ?seq rdf:type pdb:PolymerSequence ." +
- " ?seq pdb:hasValue ?sequence. } " +
+ " ?seq pdb:hasValue ?sequence . } " +
"WHERE { ?x1 rdf:type pdb:Helix ." +
" ?x1 pdb:beginsAt ?x2 ." +
" ?x1 pdb:endsAt ?x3 ." +
@@ -206,7 +223,9 @@
" ?x4 rdf:type <http://bio2rdf.org/pdb:Polypeptide(L)> ." +
" ?x5 dcterms:isPartOf ?x4 ." +
" ?x5 rdf:type ?x6 ." +
- " ?x5 pdb:hasChainPosition ?x8 . " +
+ " ?x5 pdb:hasChainPosition ?x8 ." +
+ " ?x8 dcterms:isPartOf ?x4 ." +
+ " ?x8 rdfs:label ?residuePosition ." +
" OPTIONAL { ?x5 pdb:isImmediatelyBefore ?x7 . } . ";
if (chainID.length() == 1 && pdbID.length() == 4)
@@ -256,6 +275,59 @@
return niter;
}
+ private HashMap<Integer, Resource> createPositionResidueMap(){
+
+ HashMap<Integer, Resource> posres = new HashMap<Integer, Resource>(150);
+ Property iib = ResourceFactory.createProperty("http://bio2rdf.org/pdb:", "isImmediatelyBefore");
+
+ ResIterator firstAAs = this.getFirstAA();
+ while ( firstAAs.hasNext()){
+ Resource firstAA = firstAAs.next();
+ Resource nextAA = firstAA;
+ Resource currentAA = firstAA;
+ do {
+ currentAA = nextAA;
+ posres.put(new Integer(this.getResiduePosition(currentAA)), currentAA);
+ nextAA = _pdbIdModel.getProperty(currentAA, iib).getResource();
+ } while (currentAA.hasProperty(iib));
+ }
+
+ return posres;
+ }
+
+ private int getResiduePosition(Resource res) {
+ Property hasChainPosition = ResourceFactory.createProperty("http://bio2rdf.org/pdb:", "hasChainPosition");
+ Property label = ResourceFactory.createProperty("http://www.w3.org/2000/01/rdf-schema#", "label");
+ ResourceFactory.createResource();
+
+ NodeIterator residuePosition = _pdbIdModel.listObjectsOfProperty(res, hasChainPosition );
+ ArrayList<RDFNode> positionNodes = new ArrayList<RDFNode>();
+ ArrayList<String> positionLabels = new ArrayList<String>();
+ while ( residuePosition.hasNext() ) {
+ RDFNode positionNode = residuePosition.next();
+ positionNodes.add(positionNode);
+ NodeIterator positionLabelNodes = _pdbIdModel.listObjectsOfProperty( positionNode.asResource(), label );
+ while ( positionLabelNodes.hasNext() ) {
+ positionLabels.add(positionLabelNodes.next().toString());
+ }
+
+ }
+
+
+ Integer position = null;
+ if ( positionNodes.size() == 1 && positionLabels.size() == 1 ) {
+ String positionLabel = positionLabels.get(0);
+ String a = new String( "Position " );
+ String b = new String( " on chain" );
+ position = Integer.parseInt(
+ positionLabel.substring(positionLabel.indexOf(a) + a.length(), positionLabel.indexOf(b)));
+ } else {
+ position = new Integer(0);
+ _logger.error("");
+ }
+ return position.intValue();
+ }
+
public void addDistanceInfo(){
String queryString =
"PREFIX pdb: <http://bio2rdf.org/pdb:> " +
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
