GenotypeLocusData now extends Variation.

Also, Variations should be INSERTIONs or DELETIONs (and not just INDELs).
Technically, VCF records can be indels now.
More changes coming


git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@2150 348d0f76-0448-11de-a6fe-93d51630548a

java/src/org/broadinstitute/sting/gatk/refdata/RodGLF.java

@@ -263,7 +263,8 @@ public class RodGLF implements VariationRod, Iterator<RodGLF> {
     @Override
     public VARIANT_TYPE getType() {
         if (this.isSNP()) return VARIANT_TYPE.SNP;
-        else if (this.isInsertion() || this.isDeletion()) return VARIANT_TYPE.INDEL;
+        else if (this.isInsertion()) return VARIANT_TYPE.INSERTION;
+        else if (this.isDeletion()) return VARIANT_TYPE.DELETION;
         else return VARIANT_TYPE.REFERENCE;
     }
 

java/src/org/broadinstitute/sting/gatk/refdata/RodVCF.java

@@ -126,7 +126,8 @@ public class RodVCF extends BasicReferenceOrderedDatum implements VariationRod,
     @Override
     public VARIANT_TYPE getType() {
         if (this.isSNP()) return VARIANT_TYPE.SNP;
-        else if (this.isIndel()) return VARIANT_TYPE.INDEL;
+        else if (this.isInsertion()) return VARIANT_TYPE.INSERTION;
+        else if (this.isDeletion()) return VARIANT_TYPE.DELETION;
         return VARIANT_TYPE.REFERENCE;
     }
 

java/src/org/broadinstitute/sting/gatk/refdata/SimpleIndelROD.java

@@ -61,7 +61,7 @@ public class SimpleIndelROD extends TabularROD implements Genotype, VariationRod
     /** @return the VARIANT_TYPE of the current variant */
     @Override
     public VARIANT_TYPE getType() {
-        return VARIANT_TYPE.INDEL;
+        return isInsertion() ? VARIANT_TYPE.INSERTION : VARIANT_TYPE.DELETION;
     }
 
     public boolean isSNP() { return false; }

java/src/org/broadinstitute/sting/gatk/refdata/rodPicardDbSNP.java

@@ -82,8 +82,9 @@ public class rodPicardDbSNP implements VariationRod {
             case SNP:
                 return VARIANT_TYPE.SNP;
             case insertion:
+                return VARIANT_TYPE.INSERTION;
             case deletion:
-                return VARIANT_TYPE.INDEL;
+                return VARIANT_TYPE.DELETION;
         }
         return null;
     }
@@ -131,7 +132,7 @@ public class rodPicardDbSNP implements VariationRod {
      * @return true if we're an insertion or deletion
      */
     public boolean isIndel() {
-        return getType() == VARIANT_TYPE.INDEL;
+        return getType() == VARIANT_TYPE.INSERTION || getType() == VARIANT_TYPE.DELETION;
     }
 
     public String getName() {

java/src/org/broadinstitute/sting/gatk/walkers/genotyper/DiploidGenotypeCalculationModel.java

@@ -2,7 +2,6 @@ package org.broadinstitute.sting.gatk.walkers.genotyper;
 
 import org.broadinstitute.sting.utils.*;
 import org.broadinstitute.sting.utils.genotype.*;
-import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
 
 import java.util.*;
 

java/src/org/broadinstitute/sting/gatk/walkers/genotyper/EMGenotypeCalculationModel.java

@@ -48,7 +48,7 @@ public abstract class EMGenotypeCalculationModel extends GenotypeCalculationMode
         }
 
         // generate the calls
-        GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, context.getLocation());
+        GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, context.getLocation(), Variation.VARIANT_TYPE.SNP);
         if ( locusdata != null ) {
             if ( locusdata instanceof ConfidenceBacked ) {
                 ((ConfidenceBacked)locusdata).setConfidence(phredScaledConfidence);
@@ -77,9 +77,7 @@ public abstract class EMGenotypeCalculationModel extends GenotypeCalculationMode
 
                 ((SLODBacked)locusdata).setSLOD(strandScore);
             }
-            if ( locusdata instanceof AlleleFrequencyBacked ) {
+            locusdata.setAlleleFrequency(overall.getMAF());
-                ((AlleleFrequencyBacked)locusdata).setAlleleFrequency(overall.getMAF());
-            }
         }
         return new Pair<List<Genotype>, GenotypeLocusData>(genotypeCallsFromGenotypeLikelihoods(overall, ref, contexts), locusdata);
     }

java/src/org/broadinstitute/sting/gatk/walkers/genotyper/JointEstimateGenotypeCalculationModel.java

@@ -2,6 +2,7 @@ package org.broadinstitute.sting.gatk.walkers.genotyper;
 
 import org.broadinstitute.sting.utils.*;
 import org.broadinstitute.sting.utils.genotype.*;
+import org.broadinstitute.sting.utils.genotype.Variation.VARIANT_TYPE;
 import org.broadinstitute.sting.gatk.refdata.RefMetaDataTracker;
 import org.broadinstitute.sting.gatk.refdata.rodDbSNP;
 import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
@@ -314,17 +315,13 @@ public abstract class JointEstimateGenotypeCalculationModel extends GenotypeCalc
 
         // next, the general locus data
         // note that calculating strand bias involves overwriting data structures, so we do that last
-        GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, loc);
+        GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, loc, VARIANT_TYPE.SNP);
         if ( locusdata != null ) {
+            locusdata.addAlternateAllele(bestAlternateAllele.toString());
+            locusdata.setAlleleFrequency((double)bestAFguess / (double)(frequencyEstimationPoints-1));
             if ( locusdata instanceof ConfidenceBacked ) {
                 ((ConfidenceBacked)locusdata).setConfidence(phredScaledConfidence);
             }
-            if ( locusdata instanceof AlternateAlleleBacked ) {
-                ((AlternateAlleleBacked)locusdata).setAlternateAllele(bestAlternateAllele);
-            }
-            if ( locusdata instanceof AlleleFrequencyBacked ) {
-                ((AlleleFrequencyBacked)locusdata).setAlleleFrequency((double)bestAFguess / (double)(frequencyEstimationPoints-1));
-            }
             if ( locusdata instanceof IDBacked ) {
                 rodDbSNP dbsnp = getDbSNP(tracker);
                 if ( dbsnp != null )

java/src/org/broadinstitute/sting/gatk/walkers/genotyper/PointEstimateGenotypeCalculationModel.java

@@ -86,7 +86,7 @@ public class PointEstimateGenotypeCalculationModel extends EMGenotypeCalculation
                 ((PosteriorsBacked)call).setPosteriors(discoveryGL.second.getPosteriors());
             }
 
-            GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, context.getLocation());
+            GenotypeLocusData locusdata = GenotypeWriterFactory.createSupportedGenotypeLocusData(OUTPUT_FORMAT, ref, context.getLocation(), Variation.VARIANT_TYPE.SNP);
             if ( locusdata != null ) {
                 if ( locusdata instanceof ConfidenceBacked ) {
                     ((ConfidenceBacked)locusdata).setConfidence(phredScaledConfidence);

java/src/org/broadinstitute/sting/utils/genotype/AlleleFrequencyBacked.java

@@ -1,23 +0,0 @@
-package org.broadinstitute.sting.utils.genotype;
-
-/**
- * @author ebanks
- * Interface AlleleFrequencyBacked
- *
- * this interface indicates that the genotype is
- * backed up by allele frequency information.
- */
-public interface AlleleFrequencyBacked {
-
-    /**
-     *
-     * @return returns the best allele frequency for this genotype
-     */
-    public double getAlleleFrequency();
-
-    /**
-     *
-     * @param   frequency    the allele frequency for this genotype
-     */
-    public void setAlleleFrequency(double frequency);
-}

java/src/org/broadinstitute/sting/utils/genotype/AlternateAlleleBacked.java

@@ -1,24 +0,0 @@
-package org.broadinstitute.sting.utils.genotype;
-
-/**
- * @author ebanks
- * Interface AlternateAlleleBacked
- *
- * this interface indicates that the genotype is
- * backed up by alternate allele information.
- */
-public interface AlternateAlleleBacked {
-
-    /**
-     *
-     * @return returns the alternate allele for this genotype
-     */
-    public char getAlternateAllele();
-
-    /**
-     *
-     * @param   alt    the alternate allele base for this genotype
-     */
-    public void setAlternateAllele(char alt);
-
-}

java/src/org/broadinstitute/sting/utils/genotype/BasicVariation.java

@@ -65,7 +65,7 @@ public class BasicVariation implements Variation {
      */
     @Override
     public VARIANT_TYPE getType() {
-        if (mLength != 0) return VARIANT_TYPE.INDEL;
+        if (mLength != 0) return VARIANT_TYPE.INSERTION;
         return (isSNP()) ? VARIANT_TYPE.SNP : VARIANT_TYPE.REFERENCE;
     }
 

java/src/org/broadinstitute/sting/utils/genotype/GenotypeLocusData.java

@@ -1,7 +1,5 @@
 package org.broadinstitute.sting.utils.genotype;
 
-import org.broadinstitute.sting.utils.GenomeLoc;
-
 
 /**
  * @author ebanks
@@ -10,19 +8,18 @@ import org.broadinstitute.sting.utils.GenomeLoc;
  *         <p/>
  *         represents the locus specific data associated with a genotype object.
  */
-public interface GenotypeLocusData {
+public interface GenotypeLocusData extends Variation {
 
     /**
-      * get the reference base.
-      * @return a character, representing the reference base
-      */
-    public char getReference();
-
-    /**
-     * get the genotype's location
      *
-     * @return a GenomeLoc representing the location
+     * @param   alt    the alternate allele base for this genotype
      */
-    public GenomeLoc getLocation();    
+    public void addAlternateAllele(String alt);
+
+    /**
+     *
+     * @param   frequency    the allele frequency for this genotype
+     */
+    public void setAlleleFrequency(double frequency);
 
 }

java/src/org/broadinstitute/sting/utils/genotype/GenotypeWriterFactory.java

@@ -99,12 +99,13 @@ public class GenotypeWriterFactory {
      * @param format the format
      * @param ref the reference base
      * @param loc the location
+     * @param type the variant type
      * @return an unpopulated genotype locus data object
      */
-    public static GenotypeLocusData createSupportedGenotypeLocusData(GENOTYPE_FORMAT format, char ref, GenomeLoc loc) {
+    public static GenotypeLocusData createSupportedGenotypeLocusData(GENOTYPE_FORMAT format, char ref, GenomeLoc loc, Variation.VARIANT_TYPE type) {
         switch (format) {
             case VCF:
-                return new VCFGenotypeLocusData(ref, loc);
+                return new VCFGenotypeLocusData(ref, loc, type);
             case GELI:
             case GELI_BINARY:
                 return null;

java/src/org/broadinstitute/sting/utils/genotype/Variation.java

@@ -14,10 +14,12 @@ import java.util.List;
 public interface Variation {
     // the types of variants we currently allow
     public enum VARIANT_TYPE {
-        SNP, INDEL, REFERENCE // though reference is not really a variant, we need to represent it
+        SNP, INSERTION, DELETION, REFERENCE // though reference is not really a variant, we need to represent it
     }
 
-    /** are we bi-allelic? */
+    /**
+     * @return true if we are bi-allelic?
+     */
     public boolean isBiallelic();
 
     /**
@@ -39,7 +41,7 @@ public interface Variation {
     public VARIANT_TYPE getType();
 
     /**
-     * are we a SNP? If not we're a Indel/deletion or the reference.  This method must be call before you use
+     * are we a SNP? If not we're a Indel/deletion or the reference.  This method must be called  before you use
      * the convenience methods getAlternativeBaseForSNP or getReferenceForSNP, to ensure that you're working with a SNP
      *
      * @return true if we're a SNP
@@ -127,7 +129,7 @@ public interface Variation {
     /**
      * gets the reference base is the case of a SNP.  Throws an IllegalStateException if we're not a SNP
      *
-     * @return a char, representing the alternate base
+     * @return a char, representing the reference base
      */
     public char getReferenceForSNP();
 

java/src/org/broadinstitute/sting/utils/genotype/vcf/VCFGenotypeLocusData.java

@@ -3,8 +3,7 @@ package org.broadinstitute.sting.utils.genotype.vcf;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.utils.genotype.*;
 
-import java.util.HashMap;
+import java.util.*;
-import java.util.Map;
 
 /**
  * @author ebanks
@@ -13,7 +12,7 @@ import java.util.Map;
  *         <p/>
  *         represents the meta data for a genotype object.
  */
-public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked, SLODBacked, IDBacked, AlternateAlleleBacked, AlleleFrequencyBacked, ArbitraryFieldsBacked {
+public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked, SLODBacked, IDBacked, ArbitraryFieldsBacked {
 
     // the discovery lod score
     private double mConfidence = 0.0;
@@ -29,7 +28,10 @@ public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked
 
     // the ref base and alt bases
     private char mRefBase;
-    private char mAltBase = 'N';
+    private List<String> mAltBases = new ArrayList<String>();
+
+    // the variant type
+    private VARIANT_TYPE mType = VARIANT_TYPE.SNP;
 
     // the id
     private String mID;
@@ -42,18 +44,20 @@ public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked
      *
      * @param ref       the reference base
      * @param loc       the locus
+     * @param type      the variant type
      */
-    public VCFGenotypeLocusData(char ref, GenomeLoc loc) {
+    public VCFGenotypeLocusData(char ref, GenomeLoc loc, VARIANT_TYPE type) {
         mRefBase = ref;
         mLoc = loc;
+        mType = type;
     }
 
     /**
       * get the reference base.
       * @return a character, representing the reference base
       */
-    public char getReference() {
+    public String getReference() {
-        return mRefBase;
+        return String.valueOf(mRefBase);
     }
 
     /**
@@ -65,20 +69,68 @@ public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked
         return mLoc;
     }
 
-    /**
+    public boolean isBiallelic() {
-     *
+        return mAltBases.size() == 1;
-     * @return returns the alternate allele for this genotype
-     */
-    public char getAlternateAllele() {
-        return mAltBase;
     }
 
-    /**
+    public boolean isSNP() {
-     *
+        return mType == VARIANT_TYPE.SNP;
-     * @param   alt    the alternate allele base for this genotype
+    }
-     */
+
-    public void setAlternateAllele(char alt) {
+    public boolean isInsertion() {
-        mAltBase = alt;
+        return mType == VARIANT_TYPE.INSERTION;
+    }
+
+    public boolean isIndel() {
+        return mType == VARIANT_TYPE.INSERTION || mType == VARIANT_TYPE.DELETION;
+    }
+
+    public boolean isDeletion() {
+        return mType == VARIANT_TYPE.DELETION;
+    }
+
+    public boolean isReference() {
+        return mType == VARIANT_TYPE.REFERENCE;
+    }
+
+    public VARIANT_TYPE getType() {
+        return mType;
+    }
+
+    public double getNonRefAlleleFrequency() {
+        return mAlleleFrequency;
+    }
+
+    public double getNegLog10PError() {
+        return mConfidence / 10.0;
+    }
+
+    public List<String> getAlternateAlleleList() {
+        return mAltBases;
+    }
+
+    public void addAlternateAllele(String alt) {
+        mAltBases.add(alt);
+    }
+
+    public List<String> getAlleleList() {
+        LinkedList<String> alleles = new LinkedList<String>(mAltBases);
+        alleles.addFirst(getReference());
+        return alleles;
+    }
+
+    public char getAlternativeBaseForSNP() {
+        if ( !isSNP() )
+            throw new IllegalStateException("This variant is not a SNP");
+        if ( mAltBases.size() == 0 )
+            throw new IllegalStateException("No alternate alleles have been set");
+        return mAltBases.get(0).charAt(0);
+    }
+
+    public char getReferenceForSNP() {
+        if ( !isSNP() )
+            throw new IllegalStateException("This variant is not a SNP");
+        return mRefBase;
     }
 
     /**
@@ -115,15 +167,6 @@ public class VCFGenotypeLocusData implements GenotypeLocusData, ConfidenceBacked
         mSLOD = slod;
     }
 
-    /**
-     * get the allele frequency
-     *
-     * @return the allele frequency
-     */
-    public double getAlleleFrequency() {
-        return mAlleleFrequency;
-    }
-
     /**
      *
      * @param   frequency    the allele frequency for this genotype

java/src/org/broadinstitute/sting/utils/genotype/vcf/VCFGenotypeWriterAdapter.java

@@ -121,10 +121,11 @@ public class VCFGenotypeWriterAdapter implements GenotypeWriter {
             if ( locusdata == null )
                 throw new IllegalArgumentException("Unable to parse out the current location: genotype array must contain at least one entry or have locusdata");
 
-            params.setLocations(locusdata.getLocation(), locusdata.getReference());
+            params.setLocations(locusdata.getLocation(), locusdata.getReference().charAt(0));
 
             // if there is no genotype data, we'll also need to set an alternate allele
-            params.addAlternateBase(new VCFGenotypeEncoding(String.valueOf(((VCFGenotypeLocusData)locusdata).getAlternateAllele())));
+            if ( locusdata.isSNP() && locusdata.isBiallelic() )
+                params.addAlternateBase(new VCFGenotypeEncoding(locusdata.getAlternateAlleleList().get(0)));
         } else {
             params.setLocations(genotypes.get(0).getLocation(), genotypes.get(0).getReference());
         }
@@ -189,7 +190,7 @@ public class VCFGenotypeWriterAdapter implements GenotypeWriter {
         if ( locusdata != null ) {
             if ( locusdata.getSLOD() != null )
                 infoFields.put("SB", String.format("%.2f", locusdata.getSLOD()));
-            infoFields.put("AF", String.format("%.2f", locusdata.getAlleleFrequency()));
+            infoFields.put("AF", String.format("%.2f", locusdata.getNonRefAlleleFrequency()));
             Map<String, String> otherFields = locusdata.getFields();
             if ( otherFields != null ) {
                 infoFields.putAll(otherFields);