Menu
▾
▴
[Treebase-guts] SF.net SVN: treebase:[1029] trunk/treebase-core/src
|
From: <rv...@us...> - 2012-02-01 21:00:38
|
Revision: 1029
http://treebase.svn.sourceforge.net/treebase/?rev=1029&view=rev
Author: rvos
Date: 2012-02-01 21:00:31 +0000 (Wed, 01 Feb 2012)
Log Message:
-----------
Adding state set mappers, which help translate character state sequences from ones that contain mini-syntax for indicating ambiguity as {ACGT} to IUPAC single character ambiguity codes such as N.
Added Paths:
-----------
trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapper.java
trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperDna.java
trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperProtein.java
trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperRna.java
trunk/treebase-core/src/test/java/org/cipres/treebase/domain/matrix/StateSetMapperTest.java
Added: trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapper.java
===================================================================
--- trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapper.java (rev 0)
+++ trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapper.java 2012-02-01 21:00:31 UTC (rev 1029)
@@ -0,0 +1,93 @@
+package org.cipres.treebase.domain.matrix;
+
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Map;
+
+/**
+ * This class (and its subclasses) help with mapping between sets
+ * of fundamental states (such as {ACGT}) and IUPAC single character
+ * ambiguity symbols. This is especially important for valid NeXML
+ * generation, which requires that molecular <seq> elements only
+ * contain single character symbols, not sets.
+ * @author rvosa
+ *
+ */
+public abstract class StateSetMapper {
+ private Map<Character,HashSet<Character>> symbolToAmbig;
+ private Map<HashSet<Character>,Character> ambigToSymbol;
+
+ /**
+ * Returns a map keyed on IUPAC single character ambiguity symbols
+ * http://droog.gs.washington.edu/parc/images/iupac.html with values
+ * containing a set (possibly of 1) with fundamental states.
+ * @return
+ */
+ protected abstract Map<Character,HashSet<Character>> makeSymbolToAmbigMap();
+
+
+ public StateSetMapper () {
+ symbolToAmbig = makeSymbolToAmbigMap();
+ ambigToSymbol = new HashMap<HashSet<Character>,Character>();
+ for ( Character key : symbolToAmbig.keySet() ) {
+ HashSet<Character> value = symbolToAmbig.get(key);
+ ambigToSymbol.put(value, key);
+ }
+ }
+
+ /**
+ * Instantiates appropriate subclass given datatype
+ * @param datatype - one of the types in MatrixDataType
+ * @return
+ */
+ public static StateSetMapper createMapperForDataType(String datatype) {
+ if ( MatrixDataType.MATRIX_DATATYPE_DNA.equals(datatype) ) {
+ return new StateSetMapperDna();
+ }
+ else if ( MatrixDataType.MATRIX_DATATYPE_RNA.equals(datatype) ) {
+ return new StateSetMapperRna();
+ }
+ else if ( MatrixDataType.MATRIX_DATATYPE_PROTEIN.equals(datatype) ) {
+ return new StateSetMapperProtein();
+ }
+ else {
+ return null;
+ }
+ }
+
+ /**
+ * Looks up the IUPAC single character symbol for a set of fundamental states
+ * @return
+ */
+ public Character getSymbolForAmbiguousSet(HashSet<Character> ambiguousSet) {
+ return ambigToSymbol.get(ambiguousSet);
+ }
+
+ /**
+ * Looks up the set of fundamental states for an IUPAC single character symbol
+ * @return
+ */
+ public HashSet<Character> getAmbiguousSetForSymbol(Character symbol) {
+ return symbolToAmbig.get(symbol);
+ }
+
+ /**
+ * Returns a data type from MatrixDataType.MATRIX_DATATYPE_.*
+ * @return
+ */
+ public abstract String getDataType();
+
+ /**
+ * Helper method for populating the underlying Map object
+ * @param symbol
+ * @param symbolArray
+ * @param result
+ */
+ protected void addMapping(Character symbol,Character[] symbolArray,Map<Character, HashSet<Character>> result) {
+ HashSet<Character> symbolSet = new HashSet<Character>(symbolArray.length);
+ for ( int i = 0; i < symbolArray.length; i++ ) {
+ symbolSet.add(symbolArray[i]);
+ }
+ result.put(symbol, symbolSet);
+ }
+}
Added: trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperDna.java
===================================================================
--- trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperDna.java (rev 0)
+++ trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperDna.java 2012-02-01 21:00:31 UTC (rev 1029)
@@ -0,0 +1,65 @@
+package org.cipres.treebase.domain.matrix;
+
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Map;
+import org.cipres.treebase.domain.matrix.MatrixDataType;
+
+public class StateSetMapperDna extends StateSetMapper {
+
+ /*
+ * (non-Javadoc)
+ * @see org.cipres.treebase.domain.matrix.StateSetMapper#makeSymbolToAmbigMap()
+ */
+ @Override
+ protected Map<Character, HashSet<Character>> makeSymbolToAmbigMap() {
+ return this.makeSymbolToAmbigMapWithSymbols('A', 'C', 'G', 'T');
+ }
+
+ /**
+ *
+ * @param A
+ * @param C
+ * @param G
+ * @param T
+ * @return
+ */
+ protected Map<Character, HashSet<Character>> makeSymbolToAmbigMapWithSymbols(char A, char C, char G, char T) {
+ Map<Character, HashSet<Character>> result = new HashMap<Character, HashSet<Character>>();
+
+ addMapping(A,new Character[]{A}, result); // A
+ addMapping(C,new Character[]{C}, result); // C
+ addMapping(G,new Character[]{G}, result); // G
+ addMapping(T,new Character[]{T}, result); // T
+
+ addMapping('M',new Character[]{A,C}, result); // M => (A,C)
+ addMapping('R',new Character[]{A,G}, result); // R => (A,G)
+ addMapping('W',new Character[]{A,T}, result); // W => (A,T)
+ addMapping('S',new Character[]{C,G}, result); // S => (C,G)
+ addMapping('Y',new Character[]{C,T}, result); // Y => (C,T) XXX
+ addMapping('K',new Character[]{G,T}, result); // K => (G,T)
+
+ addMapping('V',new Character[]{A,C,G}, result); // V => (A,C,G)
+ addMapping('H',new Character[]{A,C,T}, result); // H => (A,C,T)
+ addMapping('D',new Character[]{A,G,T}, result); // D => (A,G,T)
+ addMapping('B',new Character[]{C,G,T}, result); // B => (C,G,T)
+
+ addMapping('N',new Character[]{A,C,G,T}, result); // N => (A,C,G,T)
+ addMapping('?',new Character[]{A,C,G,T,'-'}, result); // ? => (A,C,G,T,-)
+ addMapping('-',new Character[]{'-'}, result); // - => ()
+
+ return result;
+ }
+
+ /**
+ * Returns MatrixDataType.MATRIX_DATATYPE_DNA
+ */
+ @Override
+ public String getDataType() {
+ return MatrixDataType.MATRIX_DATATYPE_DNA;
+ }
+
+
+
+
+}
Added: trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperProtein.java
===================================================================
--- trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperProtein.java (rev 0)
+++ trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperProtein.java 2012-02-01 21:00:31 UTC (rev 1029)
@@ -0,0 +1,56 @@
+package org.cipres.treebase.domain.matrix;
+
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Map;
+import org.cipres.treebase.domain.matrix.MatrixDataType;
+
+public class StateSetMapperProtein extends StateSetMapper {
+
+ /*
+ * (non-Javadoc)
+ * @see org.cipres.treebase.domain.matrix.StateSetMapper#makeSymbolToAmbigMap()
+ */
+ @Override
+ protected Map<Character, HashSet<Character>> makeSymbolToAmbigMap() {
+ Map<Character, HashSet<Character>> result = new HashMap<Character, HashSet<Character>>();
+
+ addMapping('A',new Character[]{'A'}, result); // Alanine
+ addMapping('B',new Character[]{'B'}, result); // Aspartic acid or asparagine
+ addMapping('C',new Character[]{'C'}, result); // Cysteine
+ addMapping('D',new Character[]{'D'}, result); // Aspartic acid
+ addMapping('E',new Character[]{'E'}, result); // Glutamic acid
+ addMapping('F',new Character[]{'F'}, result); // Phenylanine
+ addMapping('G',new Character[]{'G'}, result); // Glycine
+ addMapping('H',new Character[]{'H'}, result); // Histidine
+ addMapping('I',new Character[]{'I'}, result); // Isoleucine
+ addMapping('K',new Character[]{'K'}, result); // Lysine
+ addMapping('L',new Character[]{'L'}, result); // Leucine
+ addMapping('M',new Character[]{'M'}, result); // Methionine
+ addMapping('N',new Character[]{'N'}, result); // Asparagine
+ addMapping('P',new Character[]{'P'}, result); // Proline
+ addMapping('Q',new Character[]{'Q'}, result); // Glutamine
+ addMapping('R',new Character[]{'R'}, result); // Arginine
+ addMapping('S',new Character[]{'S'}, result); // Serine
+ addMapping('T',new Character[]{'T'}, result); // Threonine
+ addMapping('U',new Character[]{'U'}, result); // Selenocysteine
+ addMapping('V',new Character[]{'V'}, result); // Valine
+ addMapping('W',new Character[]{'W'}, result); // Tryptophan
+ addMapping('Y',new Character[]{'Y'}, result); // Tyrosine
+ addMapping('Z',new Character[]{'Z'}, result); // Glutamic acid or glutamine
+ addMapping('X',new Character[]{'A','B','C','D','E','F','G','H','I','K','L','M','N','P','Q','R','S','T','U','V','W','Y','Z'}, result);
+ addMapping('?',new Character[]{'A','B','C','D','E','F','G','H','I','K','L','M','N','P','Q','R','S','T','U','V','W','Y','Z','-'}, result);
+ addMapping('-',new Character[]{'-'}, result);
+
+ return result;
+ }
+
+ /**
+ * Returns MatrixDataType.MATRIX_DATATYPE_PROTEIN
+ */
+ @Override
+ public String getDataType() {
+ return MatrixDataType.MATRIX_DATATYPE_PROTEIN;
+ }
+
+}
Added: trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperRna.java
===================================================================
--- trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperRna.java (rev 0)
+++ trunk/treebase-core/src/main/java/org/cipres/treebase/domain/matrix/StateSetMapperRna.java 2012-02-01 21:00:31 UTC (rev 1029)
@@ -0,0 +1,25 @@
+package org.cipres.treebase.domain.matrix;
+
+import java.util.HashSet;
+import java.util.Map;
+import org.cipres.treebase.domain.matrix.MatrixDataType;
+
+public class StateSetMapperRna extends StateSetMapperDna {
+
+ /*
+ * (non-Javadoc)
+ * @see org.cipres.treebase.domain.matrix.StateSetMapperDna#makeSymbolToAmbigMap()
+ */
+ @Override
+ protected Map<Character, HashSet<Character>> makeSymbolToAmbigMap() {
+ return makeSymbolToAmbigMapWithSymbols('A', 'C', 'G', 'U');
+ }
+
+ /**
+ * Returns MatrixDataType.MATRIX_DATATYPE_RNA
+ */
+ @Override
+ public String getDataType() {
+ return MatrixDataType.MATRIX_DATATYPE_RNA;
+ }
+}
Added: trunk/treebase-core/src/test/java/org/cipres/treebase/domain/matrix/StateSetMapperTest.java
===================================================================
--- trunk/treebase-core/src/test/java/org/cipres/treebase/domain/matrix/StateSetMapperTest.java (rev 0)
+++ trunk/treebase-core/src/test/java/org/cipres/treebase/domain/matrix/StateSetMapperTest.java 2012-02-01 21:00:31 UTC (rev 1029)
@@ -0,0 +1,54 @@
+package org.cipres.treebase.domain.matrix;
+
+import junit.framework.Assert;
+import junit.framework.TestCase;
+
+import org.cipres.treebase.domain.matrix.MatrixDataType;
+import org.cipres.treebase.domain.matrix.MatrixRow;
+import org.cipres.treebase.domain.matrix.StateSetMapper;
+
+public class StateSetMapperTest extends TestCase {
+ /**
+ * The inputSeq contains all possible permutations of the fundamental
+ * states, which are to be mapped onto the IUPAC single character symbols
+ */
+ public void testDnaMapper() {
+ String inputSeq = "ACGTK{GT}M{AC}R{AG}S{CG}W{AT}B{CGT}D{AGT}H{ACT}V{ACG}N{ACGT}-?{ACGT-}";
+ String outputSeq = "ACGTKKMMRRSSWWBBDDHHVVNN-??";
+ String result = MatrixRow.buildNormalizedSymbolString(StateSetMapper.createMapperForDataType(MatrixDataType.MATRIX_DATATYPE_DNA), inputSeq);
+ Assert.assertEquals(outputSeq, result);
+ }
+
+ /**
+ * The inputSeq contains all possible permutations of the fundamental
+ * states, which are to be mapped onto the IUPAC single character symbols
+ */
+ public void testRnaMapper() {
+ String inputSeq = "ACGUK{GU}M{AC}R{AG}S{CG}W{AU}B{CGU}D{AGU}H{ACU}V{ACG}N{ACGU}-?{ACGU-}";
+ String outputSeq = "ACGUKKMMRRSSWWBBDDHHVVNN-??";
+ String result = MatrixRow.buildNormalizedSymbolString(StateSetMapper.createMapperForDataType(MatrixDataType.MATRIX_DATATYPE_RNA), inputSeq);
+ Assert.assertEquals(outputSeq, result);
+ }
+
+ /**
+ * The inputSeq contains all possible permutations of the fundamental
+ * states, which are to be mapped onto the IUPAC single character symbols
+ */
+ public void testProteinMapper() {
+ String inputSeq = "ABCDEFGHIKLMNPQRSTUVWYZ{ABCDEFGHIKLMNPQRSTUVWYZ}{ABCDEFGHIKLMNPQRSTUVWYZ-}-";
+ String outputSeq = "ABCDEFGHIKLMNPQRSTUVWYZX?-";
+ String result = MatrixRow.buildNormalizedSymbolString(StateSetMapper.createMapperForDataType(MatrixDataType.MATRIX_DATATYPE_PROTEIN), inputSeq);
+ Assert.assertEquals(outputSeq, result);
+ }
+
+ /**
+ * For STANDARD data there is no implied mapping. For now we will simply
+ * pass these through, though that's technically not valid.
+ */
+ public void testStandardMapper() {
+ String inputSeq = "012{1}";
+ String outputSeq = "012{1}";
+ String result = MatrixRow.buildNormalizedSymbolString(StateSetMapper.createMapperForDataType(MatrixDataType.MATRIX_DATATYPE_STANDARD), inputSeq);
+ Assert.assertEquals(outputSeq, result);
+ }
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
