fix for the eval tests, and a change to the backedbygenotypes interface, more changes to come

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@1661 348d0f76-0448-11de-a6fe-93d51630548a

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

@@ -266,7 +266,6 @@ public class RodGeliText extends BasicReferenceOrderedDatum implements Variation
 
     public List<String> getGenotype() throws IllegalStateException {
         return Arrays.asList(getBestGenotype());
-        //throw new IllegalStateException("huh?");
     }
 
     public int getPloidy() throws IllegalStateException {
@@ -348,14 +347,13 @@ public class RodGeliText extends BasicReferenceOrderedDatum implements Variation
 
 
     /**
-     * get the likelihoods
+     * get the genotype
      *
-     * @return an array in lexigraphical order of the likelihoods
+     * @return a map in lexigraphical order of the genotypes
      */
     @Override
-    public Genotype getGenotype(DiploidGenotype x) {
+    public Genotype getCallexGenotype() {
-        if (x.toString() != this.getAltBasesFWD()) throw new IllegalStateException("We don't contain genotype " + x); 
+        return new BasicGenotype(getLocation(), this.getAltBasesFWD(), refBase, lodBtnb);
-        return new BasicGenotype(getLocation(), x.toString(), refBase, lodBtnb);
     }
 
     /**

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

@@ -259,6 +259,16 @@ public class RodGenotypeChipAsGFF extends BasicReferenceOrderedDatum implements
     public boolean isBiallelic() { return true; }
     public int length() { return 1; }
 
+    /**
+     * get the genotype
+     *
+     * @return a map in lexigraphical order of the genotypes
+     */
+    @Override
+    public Genotype getCallexGenotype() {
+        return new BasicGenotype(this.getLocation(),this.feature,this.getRefSnpFWD(),this.getConsensusConfidence());
+    }
+
     /**
      * get the likelihoods
      *
@@ -271,17 +281,6 @@ public class RodGenotypeChipAsGFF extends BasicReferenceOrderedDatum implements
         return ret;
     }
     
-    /**
-     * get the likelihoods
-     *
-     * @return an array in lexigraphical order of the likelihoods
-     */
-    @Override
-    public Genotype getGenotype(DiploidGenotype x) {
-        if (!x.toString().equals(this.getAltBasesFWD())) throw new IllegalStateException("Unable to retrieve genotype");
-        return new BasicGenotype(this.getLocation(),this.feature,this.getRefSnpFWD(),this.getConsensusConfidence());
-    }
-    
     /**
      * do we have the specified genotype?  not all backedByGenotypes
      * have all the genotype data.

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

@@ -249,6 +249,16 @@ public class RodVCF extends BasicReferenceOrderedDatum implements Variation, Var
         return mCurrentRecord.getReferenceBase();
     }
 
+    /**
+     * get the genotype
+     *
+     * @return a map in lexigraphical order of the genotypes
+     */
+    @Override
+    public Genotype getCallexGenotype() {
+        throw new UnsupportedOperationException("We don't support this right now"); 
+    }
+
     /**
      * get the genotypes
      *
@@ -273,29 +283,6 @@ public class RodVCF extends BasicReferenceOrderedDatum implements Variation, Var
         return genotypes;
     }
 
-
-    /**
-     * get the likelihoods
-     *
-     * @return an array in lexigraphical order of the likelihoods
-     */
-    @Override
-    public Genotype getGenotype(DiploidGenotype x) {
-        if (x.toString().equals(getReference()))
-            return new BasicGenotype(this.getLocation(), getReference(), this.getReference().charAt(0), 0);
-        for (VCFGenotypeRecord record : mCurrentRecord.getVCFGenotypeRecords()) {
-            if (Utils.join("", record.getAlleles()).equals(x.toString())) {
-                double qual = 0.0;
-                if (record.getAlleles().equals(this.getReference()))
-                    qual = this.getNegLog10PError();
-                else if (record.getFields().containsKey("GQ"))
-                    qual = Double.valueOf(record.getFields().get("GQ")) / 10.0;
-                return new BasicGenotype(this.getLocation(), Utils.join("", record.getAlleles()), this.getReference().charAt(0), qual);
-            }
-        }
-        return null;
-    }
-
     /**
      * do we have the specified genotype?  not all backedByGenotypes
      * have all the genotype data.
@@ -306,9 +293,7 @@ public class RodVCF extends BasicReferenceOrderedDatum implements Variation, Var
      */
     @Override
     public boolean hasGenotype(DiploidGenotype x) {
-        if (getGenotype(x) != null)
+        return (this.getAlternateBases().contains(x.toString()));
-            return true;
-        return false;
     }
 
     @Override

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

@@ -2,10 +2,7 @@ package org.broadinstitute.sting.gatk.refdata;
 
 import net.sf.picard.util.SequenceUtil;
 import org.broadinstitute.sting.utils.*;
-import org.broadinstitute.sting.utils.genotype.BasicGenotype;
+import org.broadinstitute.sting.utils.genotype.*;
-import org.broadinstitute.sting.utils.genotype.DiploidGenotype;
-import org.broadinstitute.sting.utils.genotype.VariantBackedByGenotype;
-import org.broadinstitute.sting.utils.genotype.Variation;
 
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -355,6 +352,16 @@ public class rodDbSNP extends BasicReferenceOrderedDatum implements Variation, V
         return (int) (loc.getStop() - loc.getStart() + 1);
     }
 
+    /**
+     * get the genotype
+     *
+     * @return a map in lexigraphical order of the genotypes
+     */
+    @Override
+    public org.broadinstitute.sting.utils.genotype.Genotype getCallexGenotype() {
+        return new BasicGenotype(this.getLocation(), this.getAltBasesFWD(), this.getRefSnpFWD(), this.getConsensusConfidence());
+    }
+
     /**
      * get the likelihoods
      *
@@ -367,17 +374,6 @@ public class rodDbSNP extends BasicReferenceOrderedDatum implements Variation, V
         return list;
     }
 
-    /**
-     * get the likelihoods
-     *
-     * @return an array in lexigraphical order of the likelihoods
-     */
-    @Override
-    public org.broadinstitute.sting.utils.genotype.Genotype getGenotype(DiploidGenotype x) {
-        if (!x.toString().equals(this.getAltBasesFWD())) throw new IllegalStateException("Unable to retrieve genotype");
-        return new BasicGenotype(this.getLocation(), this.getAltBasesFWD(), this.getRefSnpFWD(), this.getConsensusConfidence());
-    }
-
     /**
      * do we have the specified genotype?  not all backedByGenotypes
      * have all the genotype data.

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

@@ -11,18 +11,18 @@ import java.util.List;
  */
 public interface VariantBackedByGenotype {
     /**
-     * get the likelihoods
+     * get the genotype
      *
-     * @return a map in lexigraphical order of the likelihoods
+     * @return a specific genotype that represents the called genotype
      */
-    public List<Genotype> getGenotypes();
+    public Genotype getCallexGenotype();
 
     /**
-     * get the likelihoods
+     * get the genotype
      *
-     * @return a map in lexigraphical order of the likelihoods
+     * @return a map in lexigraphical order of the genotypes
      */
-    public Genotype getGenotype(DiploidGenotype x);
+    public List<Genotype> getGenotypes();
     /**
      * do we have the specified genotype?  not all backedByGenotypes
      * have all the genotype data.

java/test/org/broadinstitute/sting/playground/gatk/walkers/varianteval/VariantEvalWalkerIntegrationTest.java

@@ -114,7 +114,7 @@ public class VariantEvalWalkerIntegrationTest extends WalkerTest {
     @Test
     public void testEvalGenotypeROD() {
         List<String> md5 = new ArrayList<String>();
-        md5.add("439c2983a6e1250cd845c686a0e5a085");
+        md5.add("d337f22057d2e9cfda03af3be70ce35a");
         /**
          * the above MD5 was calculated after running the following command:
          *
@@ -144,5 +144,32 @@ public class VariantEvalWalkerIntegrationTest extends WalkerTest {
                 md5);
         List<File> result = executeTest("testEvalGenotypeROD", spec).getFirst();
     }
+
+    @Test
+    public void testEvalMarksGenotypingExample() {
+        List<String> md5 = new ArrayList<String>();
+        md5.add("39ea64299cad44f816f033db54c8e622");
+        /**
+         * Run with the following commands:
+         * 
+         * java -Xmx2048m -jar /humgen/gsa-hphome1/depristo/dev/GenomeAnalysisTK/trunk/dist/GenomeAnalysisTK.jar
+         * -T VariantEval -R /broad/1KG/reference/human_b36_both.fasta -l INFO
+         * -B eval,Variants,/humgen/gsa-scr1/ebanks/concordanceForMark/UMichVsBroad.venn.set1Only.calls
+         * -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -hc /humgen/gsa-scr1/GATK_Data/1KG_gffs/NA12878.1kg.gff
+         * -G -L 1 -o /humgen/gsa-scr1/ebanks/concordanceForMark/UMichVsBroad.venn.set1Only.calls.eval
+         */
+
+        WalkerTest.WalkerTestSpec spec = new WalkerTest.WalkerTestSpec(
+                "-T VariantEval -R /broad/1KG/reference/human_b36_both.fasta " +
+                "-B eval,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/UMichVsBroad.venn.set1Only.calls " +
+                "-D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -hc /humgen/gsa-scr1/GATK_Data/1KG_gffs/NA12878.1kg.gff " +
+                "-G " +
+                "--supressDateInformation " +
+                "-L 1:1-10,000,000 " +
+                "--outerr %s",
+                1, // just one output file
+                md5);
+        List<File> result = executeTest("testEvalMarksGenotypingExample", spec).getFirst();
+    }
 }