Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[3433] trunk/scripts/src/main/java/org/dllearner/ examples/pdb
|
From: <km...@us...> - 2011-11-23 18:49:18
|
Revision: 3433
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3433&view=rev
Author: kmpf
Date: 2011-11-23 18:49:11 +0000 (Wed, 23 Nov 2011)
Log Message:
-----------
Amino Acid Properties changed
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 2011-11-23 14:24:16 UTC (rev 3432)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/AminoAcids.java 2011-11-23 18:49:11 UTC (rev 3433)
@@ -117,32 +117,122 @@
return resourceString;
}
- public static HashMap<String, String> getAminoAcidArffAttributeMap(){
+/*
+ ++++ Amino acid names and numbers ++++
+Every line starts with the one-letter-code, followed by their numeric representation for .arff files,
+followed by their three-letter-code and finally their name.
+ A = 1 Ala Alanin
+ C = 3 Cys Cystein
+ D = 4 Asp Aspartat
+ E = 5 Glu Glutamat
+ F = 6 Phe Phenylalanin
+ G = 7 Gly Glycin
+ H = 8 His Histidin
+ I = 9 Ile Isoleucin
+ K = 11 Lys Lysin
+ L = 12 Leu Leucin
+ M = 13 Met Methionin
+ N = 14 Asn Asparagin
+ O = 15 Pyl Pyrrolysin
+ P = 16 Pro Prolin
+ Q = 17 Gln Glutamin
+ R = 18 Arg Arginin
+ S = 19 Ser Serin
+ T = 20 Thr Threonin
+ U = 21 Sec Selenocystein
+ V = 22 Val Valin
+ W = 23 Trp Tryptophan
+ Y = 25 Tyr Tyrosin
+*/
+ public static HashMap<String, String> getAminoAcidNumber(){
+ HashMap<String,String> resnum = new HashMap<String, String>(30);
+ resnum.put(new String("A"), new String("1"));
+ resnum.put(new String("C"), new String("3"));
+ resnum.put(new String("D"), new String("4"));
+ resnum.put(new String("E"), new String("5"));
+ resnum.put(new String("F"), new String("6"));
+ resnum.put(new String("G"), new String("7"));
+ resnum.put(new String("H"), new String("8"));
+ resnum.put(new String("I"), new String("9"));
+ resnum.put(new String("K"), new String("11"));
+ resnum.put(new String("L"), new String("12"));
+ resnum.put(new String("M"), new String("13"));
+ resnum.put(new String("N"), new String("14"));
+ resnum.put(new String("O"), new String("15"));
+ resnum.put(new String("P"), new String("16"));
+ resnum.put(new String("Q"), new String("17"));
+ resnum.put(new String("R"), new String("18"));
+ resnum.put(new String("S"), new String("19"));
+ resnum.put(new String("T"), new String("20"));
+ resnum.put(new String("U"), new String("21"));
+ resnum.put(new String("V"), new String("22"));
+ resnum.put(new String("W"), new String("23"));
+ resnum.put(new String("Y"), new String("25"));
+
+ return resnum;
+ }
+
+ public static HashMap<String, String> getAminoAcidNumericArffAttributeMap(){
+ // Hydrophobicity hydrophilic = 0; Hydrophobic = 1; aromatic = 2; aliphatic = 3
+ // Polarity unpolar = 0; polar = 1; positive = 2; negative = 3;
+ // Size Tiny = 0; Small = 1; Large = 2;
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("?,?,?,?,?"));
+ resdata.put(new String("A"), new String("1,0,0"));
+ resdata.put(new String("C"), new String("1,1,0"));
+ resdata.put(new String("D"), new String("0,3,1"));
+ resdata.put(new String("E"), new String("0,3,2"));
+ resdata.put(new String("F"), new String("2,0,2"));
+ resdata.put(new String("G"), new String("1,0,0"));
+ resdata.put(new String("H"), new String("2,2,2"));
+ resdata.put(new String("I"), new String("3,0,2"));
+ resdata.put(new String("K"), new String("1,2,2"));
+ resdata.put(new String("L"), new String("3,0,2"));
+ resdata.put(new String("M"), new String("1,0,2"));
+ resdata.put(new String("N"), new String("0,1,1"));
+ resdata.put(new String("O"), new String("?,?,?"));
+ resdata.put(new String("P"), new String("0,0,1"));
+ resdata.put(new String("Q"), new String("0,1,2"));
+ resdata.put(new String("R"), new String("0,2,2"));
+ resdata.put(new String("S"), new String("0,1,0"));
+ resdata.put(new String("T"), new String("1,1,1"));
+ resdata.put(new String("V"), new String("3,0,1"));
+ resdata.put(new String("W"), new String("2,1,2"));
+ resdata.put(new String("X"), new String("?,?,?")); // unknown residue (e.g. modified amino acids)
+ resdata.put(new String("Y"), new String("2,1,2"));
+ resdata.put(new String("U"), new String("?,?,?"));
+ return resdata;
+ }
+
+ public static HashMap<String, String> getAminoAcidNominalArffAttributeMap(){
+ // Hydrophobicity hydrophilic = 0; Hydrophobic = 1; aromatic = 2; aliphatic = 3
+ // Polarity unpolar = 0 polar = 1; positive = 2; negative = 3;
+ // Size Tiny = 0; Small = 1; Large = 2;
+ HashMap<String, String> resdata = new HashMap<String, String>(30);
+
+ resdata.put(new String("A"), new String("Hydrophobic,Unpolar,Tiny"));
+ resdata.put(new String("C"), new String("Hydrophobic,Polar,Tiny"));
+ resdata.put(new String("D"), new String("Hydrophilic,Negative,Small"));
+ resdata.put(new String("E"), new String("Hydrophilic,Negative,Large"));
+ resdata.put(new String("F"), new String("Aromatic,Unpolar,Large"));
+ resdata.put(new String("G"), new String("Hydrophobic,Unpolar,Tiny"));
+ resdata.put(new String("H"), new String("Aromatic,Positive,Large"));
+ resdata.put(new String("I"), new String("Aliphatic,Unpolar,Large"));
+ resdata.put(new String("K"), new String("Hydrophobic,Positive,Large"));
+ resdata.put(new String("L"), new String("Aliphatic,Unpolar,Large"));
+ resdata.put(new String("M"), new String("Hydrophobic,Unpolar,Large"));
+ resdata.put(new String("N"), new String("Hydrophilic,Polar,Small"));
+ resdata.put(new String("O"), new String("?,?,?"));
+ resdata.put(new String("P"), new String("Hydrophilic,Unpolar,Small"));
+ resdata.put(new String("Q"), new String("Hydrophilic,Polar,Large"));
+ resdata.put(new String("R"), new String("Hydrophilic,Positive,Large"));
+ resdata.put(new String("S"), new String("Hydrophilic,Polar,Tiny"));
+ resdata.put(new String("T"), new String("Hydrophobic,Polar,Small"));
+ resdata.put(new String("V"), new String("Aliphatic,Unpolar,Small"));
+ resdata.put(new String("W"), new String("Aromatic,Polar,Large"));
+ resdata.put(new String("X"), new String("?,?,?")); // unknown residue (e.g. modified amino acids)
+ resdata.put(new String("Y"), new String("Aromatic,Polar,Large"));
+ 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 2011-11-23 14:24:16 UTC (rev 3432)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/HelixRDFCreator.java 2011-11-23 18:49:11 UTC (rev 3433)
@@ -20,8 +20,6 @@
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.LearningProblemUnsupportedException;
-
-import com.hp.hpl.jena.rdf.model.NodeIterator;
import com.hp.hpl.jena.rdf.model.Property;
import com.hp.hpl.jena.rdf.model.ResIterator;
import com.hp.hpl.jena.rdf.model.Resource;
@@ -176,7 +174,8 @@
if (arff)
{
ResIterator niter = trainmodel.getFirstAA();
- createArffFile(pdbDir, trainmodel, niter);
+ createNumericArffFile(pdbDir, trainmodel, niter);
+ createNominalArffFile(pdbDir, trainmodel, niter);
}
/*
@@ -361,7 +360,7 @@
// add knowledge source definition to <PDB ID>.conf files
confFile.println(ks);
-
+
// add knowledge source definition to <PDB ID>_<Amino Acid>.conf files
Iterator<Resource> resources = resprint.keySet().iterator();
while (resources.hasNext()){
@@ -451,13 +450,14 @@
}
}
- private static void createArffFile(String pdbDir, PDBIdRdfModel model, ResIterator firstAAs){
+ private static void createNumericArffFile(String pdbDir, PDBIdRdfModel model, ResIterator firstAAs){
try {
PDBProtein protein = model.getProtein();
String arffFilePath = pdbDir + protein.getArffFileName();
+ arffFilePath = arffFilePath.replace(".arff", ".numeric.arff");
PrintStream out = new PrintStream (arffFilePath);
- _logger.debug("Creating ARFF file: " + arffFilePath);
+ _logger.debug("Creating numeric ARFF file: " + arffFilePath);
/*
* RELATION
@@ -470,14 +470,16 @@
* ATTRIBUTES
*/
// Integer declaring Position in chain
- StringBuffer attributes = new StringBuffer("@ATTRIBUTE hydrophob NUMERIC\n" + // Hydrophilic = 0; Hydrophobic = 1; Very_hydrophobic = 2
- "@ATTRIBUTE charge NUMERIC\n" + // Negative = -1; Neutral = 0; Positive = 1
- "@ATTRIBUTE size NUMERIC\n" + // Large = 2; Small = 1; Tiny = 0.5
- "@ATTRIBUTE aromaticity NUMERIC\n" + // Aliphatic = 0; Aromatic = 1
- "@ATTRIBUTE hydrogen_bonding NUMERIC\n"); // Donor = 1; Donor/Acceptor = 0; Acceptor = -1
+ StringBuffer attributes = new StringBuffer(
+ // Hydrophobicity hydrophilic = 0; Hydrophobic = 1; aromatic = 2; aliphatic = 3
+ "@ATTRIBUTE hydrophobicity NUMERIC\n" +
+ // Polarity unpolar = 0 polar = 1; positive = 2; negative = 3;
+ "@ATTRIBUTE polarity NUMERIC\n" +
+ // Size Tiny = 0; Small = 1; Large = 2;
+ "@ATTRIBUTE size NUMERIC\n");
for (int i = -8; i <= 8; i++) {
- attributes.append("@ATTRIBUTE aa_position_" + i + " {A,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,Y}\n"); // amino acid at position $i from current amino acid
+ attributes.append("@ATTRIBUTE aa_position_" + i + " NUMERIC\n"); // amino acid at position $i from current amino acid
}
attributes.append("@ATTRIBUTE in_helix NUMERIC\n"); // Helix = 1 Other = 0
@@ -492,14 +494,14 @@
out.println(data);
// HashMap containing information about the properties of every amino acid
- HashMap<String, String> resdata = AminoAcids.getAminoAcidArffAttributeMap();
+ HashMap<String, String> resdata = AminoAcids.getAminoAcidNumericArffAttributeMap();
+ HashMap<String, String> resnum = AminoAcids.getAminoAcidNumber();
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) );
@@ -509,13 +511,14 @@
dataLine.append( resdata.get(key) + "," );
} else {
//
- dataLine.append( resdata.get("U") + "," );
+ dataLine.append( resdata.get("X") + "," );
}
// add information about neighbouring amino acids to dataLine
for (int j = (i - 8); j <= (i + 8) ; j++){
try {
- dataLine.append( protein.getSequence().charAt(j) + "," );
+ dataLine.append( resnum.get(
+ Character.toString(protein.getSequence().charAt(j))) + "," );
} catch (IndexOutOfBoundsException e) {
dataLine.append( "?," );
}
@@ -538,50 +541,96 @@
} catch (FileNotFoundException e){
e.printStackTrace();
}
+ }
+
+ private static void createNominalArffFile(String pdbDir, PDBIdRdfModel model, ResIterator firstAAs){
+
+ try {
+ PDBProtein protein = model.getProtein();
+ String arffFilePath = pdbDir + protein.getArffFileName();
+ arffFilePath = arffFilePath.replace(".arff", ".nominal.arff");
+ PrintStream out = new PrintStream (arffFilePath);
+ _logger.debug("Creating nominal ARFF file: " + arffFilePath);
+
+ /*
+ * RELATION
+ */
+ String relation = "@RELATION " + protein.getPdbID();
+ out.println(relation);
+ _logger.debug(relation);
+ /*
+ * ATTRIBUTES
+ */
+ // Integer declaring Position in chain
+ StringBuffer attributes = new StringBuffer(
+ // Hydrophobicity hydrophilic = 0; hydrophobic = 1; aromatic = 2; aliphatic = 3
+ "@ATTRIBUTE hydrophob {hydrophilic, hydrophobic, aromatic, aliphatic}\n" + // Hydrophilic = 0; Hydrophobic = 1; Very_hydrophobic = 2
+ // Polarity unpolar = 0; polar = 1; positive = 2; negative = 3;
+ "@ATTRIBUTE charge {unpolar, polar, positive, negative}\n" + // Negative = -1; Neutral = 0; Positive = 1
+ // Size tiny = 0; small = 1; large = 2;
+ "@ATTRIBUTE size {tiny, small, large}\n");
+
+ for (int i = -8; i <= 8; i++) {
+ attributes.append("@ATTRIBUTE aa_position_" + i + " {A,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,Y}\n"); // amino acid at position $i from current amino acid
+ }
+ attributes.append("@ATTRIBUTE in_helix {Helix, Non_helix}\n"); // Helix = 1 Other = 0
+
+ _logger.debug(attributes);
+ out.println(attributes);
-/* // to be exchanged
- while (firstAAs.hasNext()){
- Resource firstAA = firstAAs.next();
- Resource currentAA = firstAA;
- Resource nextAA = firstAA;
+ /*
+ * @DATA
+ */
+ String data = "@DATA\n";
+ _logger.debug(data);
+ out.println(data);
+
+ // HashMap containing information about the properties of every amino acid
+ HashMap<String, String> resdata = AminoAcids.getAminoAcidNominalArffAttributeMap();
+ ArrayList<Resource> positives = model.getPositives();
+ ArrayList<Resource> negatives = model.getNegatives();
+ 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("X") + "," );
+ }
+ // 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( "Helix" );
+ } else if (negatives.contains( posRes.get( new Integer(i) ))){
+ dataLine.append( "Non_helix" );
+ } else {
+ dataLine.append( "?" );
+ }
- for ( int i = 0; currentAA.hasProperty(iib); i++ ) {
- StringBuffer dataLine = new StringBuffer("");
- currentAA = nextAA;
-
- NodeIterator niter = model.getModel().listObjectsOfProperty(currentAA, type);
- while (niter.hasNext()){
- Resource key = niter.next().asResource();
- if (resdata.containsKey(key)){
- dataLine.append( resdata.get(key) + "," );
- }
- }
-
- for (int j = (i - 8); j <= (i + 8) ; j++){
- try {
- dataLine.append( protein.getSequence().charAt(j) + "," );
- } catch (IndexOutOfBoundsException e) {
- dataLine.append( "?," );
- }
- }
-
- if (positives.contains(currentAA)){
- dataLine.append( "1" );
- } else if (negatives.contains(currentAA)){
- dataLine.append( "0" );
- } else {
- dataLine.append( "?" );
- }
+ _logger.info(dataLine);
+ out.println(dataLine);
+
+ }
+
+ } catch (FileNotFoundException e){
+ e.printStackTrace();
+ }
+ }
-
-
- // get next AA if there is one
- if (model.getModel().contains(currentAA, iib)){
- nextAA = model.getModel().getProperty(currentAA, iib).getResource();
- }
-*/
- }
}
Modified: trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java 2011-11-23 14:24:16 UTC (rev 3432)
+++ trunk/scripts/src/main/java/org/dllearner/examples/pdb/PDBIdRdfModel.java 2011-11-23 18:49:11 UTC (rev 3433)
@@ -52,7 +52,7 @@
this.getProtein().setSpecies(extractSpecies(_pdbIdModel));
System.out.println("Species: " + this.getProtein().getSpecies());
createPositivesAndNegatives();
- _positionResource = createPositionResidueMap();
+ this._positionResource = createPositionResidueMap();
}
public PdbRdfModel getModel(){
@@ -160,15 +160,15 @@
Resource polymerSequence = ResourceFactory.createResource("http://bio2rdf.org/pdb:PolymerSequence");
ResIterator riter = model.listResourcesWithProperty(type, polymerSequence);
- while (riter.hasNext()){
- Resource nextRes = riter.next();
+ while (riter.hasNext()) {
+ Resource nextRes = riter.nextResource();
if (model.contains(nextRes, hasValue)){
NodeIterator niter = model.listObjectsOfProperty(nextRes, hasValue);
sequence = niter.next().toString();
System.out.println("Sequence: " + sequence);
}
- }
+ } ;
return sequence;
}
@@ -230,7 +230,6 @@
// a amino acid is followed
" OPTIONAL { ?x5 pdb:isImmediatelyBefore ?x7 . } . " +
" ?x5 pdb:hasChainPosition ?x8 ." +
- " ?x8 rdfs:label ?residuePosition ." +
" ?x8 pdb:hasValue ?x9 Filter (xsd:int(?x9)) .";
if (chainID.length() == 1 && pdbID.length() == 4)
{
@@ -283,7 +282,7 @@
Property iib = ResourceFactory.createProperty("http://bio2rdf.org/pdb:", "isImmediatelyBefore");
ResIterator firstAAs = this.getFirstAA();
- while ( firstAAs.hasNext()){
+ while ( firstAAs.hasNext()) {
Resource firstAA = firstAAs.next();
Resource currentAA = firstAA;
posres.put(new Integer(this.getResiduePosition(currentAA)), currentAA);
@@ -311,8 +310,7 @@
while ( positionLabelNodes.hasNext() ) {
positionLabels.add(positionLabelNodes.next().asLiteral().getInt());
}
-
- }
+ }
Integer position = null;
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
