Fixed B37 VCF output renderer to keep B37 contig names in the VCF out…

…put file

src/main/java/org/broadinstitute/hellbender/tools/funcotator/Funcotator.java

@@ -874,7 +874,6 @@ public void apply(final VariantContext variant, final ReadsContext readsContext,
         // Get the correct reference for B37/HG19 compliance:
         // This is necessary because of the variant transformation that gets applied in VariantWalkerBase::apply.
         final ReferenceContext correctReferenceContext = funcotatorEngine.getCorrectReferenceContext(variant, referenceContext);
-
         // Place the variant on our queue to be funcotated:
         enqueueAndHandleVariant(variant, correctReferenceContext, featureContext);
     }
@@ -924,7 +923,10 @@ protected void enqueueAndHandleVariant(final VariantContext variant, final Refer
 
         final FuncotationMap funcotationMap = funcotatorEngine.createFuncotationMapForVariant(variant, referenceContext, featureContext);
 
+        // This is necessary because we want to revert the variant contig namechange if it was applied in the VariantWalkerBase::apply method before output.
+        final VariantContext variantContextForOutput = funcotatorEngine.getCorrectVariantContextForOutput(variant);
+
         // At this point there is only one transcript ID in the funcotation map if canonical or best effect are selected
-        outputRenderer.write(variant, funcotationMap);
+        outputRenderer.write(variantContextForOutput, funcotationMap);
     }
 }

src/main/java/org/broadinstitute/hellbender/tools/funcotator/FuncotatorEngine.java

@@ -79,6 +79,12 @@ public final class FuncotatorEngine implements AutoCloseable {
      */
     private final boolean mustConvertInputContigsToHg19;
 
+    /**
+     * Whether the output variant contigs must be converted back to B37 from hg19 before being returned.
+     * (NOTE: this means that the output contigs will continue to use B37 contig names even if internally we converted them to hg19)
+     */
+    private boolean mustRevertVariantContigsFromHg19ToB37 = false;
+
     /**
      * Whether this {@link FuncotatorEngine} has only produced annotations on variants that have been labeled by the
      * {@link org.broadinstitute.hellbender.tools.funcotator.dataSources.gencode.GencodeFuncotationFactory} as {@link org.broadinstitute.hellbender.tools.funcotator.dataSources.gencode.GencodeFuncotation.VariantClassification#IGR}.
@@ -327,6 +333,22 @@ private VariantContext getCorrectVariantContextForReference(final VariantContext
         }
     }
 
+    /**
+     * Create a new {@link VariantContext} which will match the given Reference if there is a mismatch for input between the B37 reference and the HG19 reference.
+     * @param variant A {@link VariantContext} object containing the variant to convert.
+     * @return A {@link VariantContext} whose contig has been transformed to HG19 if requested by the user.  Otherwise, an identical variant.
+     */
+    VariantContext getCorrectVariantContextForOutput(final VariantContext variant) {
+        if ( mustRevertVariantContigsFromHg19ToB37 ) {
+            final VariantContextBuilder vcb = new VariantContextBuilder(variant);
+            vcb.chr(FuncotatorUtils.convertHG19ContigToB37Contig(variant.getContig()));
+            return vcb.make();
+        }
+        else {
+            return variant;
+        }
+    }
+
     /**
      * @return The default {@link VariantTransformer} which will automatically convert from the B37 reference standard to the HG19 reference standard for contig names.
      */
@@ -483,7 +505,7 @@ private boolean determineReferenceAndDatasourceCompatibility() {
         }
         else if ( funcotatorArgs.referenceVersion.equals(BaseFuncotatorArgumentCollection.FuncotatorReferenceVersionHg19) &&
                         FuncotatorUtils.isSequenceDictionaryUsingB37Reference(sequenceDictionaryForDrivingVariants) ) {
-            logger.info("VCF sequence dictionary detected as B37 in HG19 annotation mode.  Performing conversion.");
+            logger.info("VCF sequence dictionary detected as B37 in HG19 annotation mode.  Performing conversion. (NOTE: the output VCF will still be B37)");
             mustConvertInputContigsToHg19 = true;
         }
         else {
@@ -505,6 +527,11 @@ else if ( funcotatorArgs.referenceVersion.equals(BaseFuncotatorArgumentCollectio
                     "There MAY be some errors (e.g. in the Y chromosome, but possibly in other places as well) due to changes between the two references.");
         }
 
+        // Record whether we need to revert the contigs back to B37 after annotation:
+        if (FuncotatorUtils.isSequenceDictionaryUsingB37Reference(sequenceDictionaryForDrivingVariants) && mustConvertInputContigsToHg19) {
+            this.mustRevertVariantContigsFromHg19ToB37 = true;
+        }
+
         return mustConvertInputContigsToHg19;
     }
 

src/main/java/org/broadinstitute/hellbender/tools/funcotator/vcfOutput/VcfOutputRenderer.java

@@ -197,7 +197,8 @@ public void write(final VariantContext variant, final FuncotationMap txToFuncota
         variantContextOutputBuilder.genotypes( variant.getGenotypes() );
 
         // Render and add our VCF line:
-        vcfWriter.add( variantContextOutputBuilder.make() );
+        VariantContext out = variantContextOutputBuilder.make();
+        vcfWriter.add( out );
     }
 
     private Funcotation createManualAnnotationFuncotation(final Allele altAllele) {

src/test/resources/org/broadinstitute/hellbender/tools/funcotator/validationTestData/hashSetOrderingIssue_expected.vcf

@@ -91,4 +91,4 @@
 ##reference=/cromwell_root/broad-references/hg19/v0/Homo_sapiens_assembly19.fasta
 ##source=Funcotator
 #CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO
-chr2	70120909	rs3214822	GA	G	722.12	PASS	FUNCOTATION=[SNRNP27|hg19|chr2|70120910|70120910|FIVE_PRIME_UTR||DEL|A|A|-|g.chr2:70120910delA|ENST00000244227.3|+|1|||||0.43640897755610975|GGGAAAAATGAAAGCTGTGTT|SNRNP27_ENST00000409116.1_FIVE_PRIME_FLANK/SNRNP27_ENST00000488986.1_FIVE_PRIME_FLANK|||||||||||||||||||||||||91|biliary_tract(2)_%7C_breast(12)_%7C_central_nervous_system(44)_%7C_large_intestine(11)_%7C_pancreas(22)|||||||X76302|NM_006857.2|NP_006848.1|HGNC:30240|small_%20_nuclear_%20_ribonucleoprotein_%20_U4/U6.U5_%20_subunit_%20_27|Approved|gene_%20_with_%20_protein_%20_product|protein-coding_%20_gene||"small_%20_nuclear_%20_ribonucleoprotein_%20_27kDa_%20_(U4/U6.U5)"_%2C__%20_"small_%20_nuclear_%20_ribonucleoprotein_%2C__%20_U4/U6.U5_%20_27kDa_%20_subunit"|RY1_%2C__%20_U4/U6.U5-27K|"nucleic_%20_acid_%20_binding_%20_protein_%20_RY_%20_1"_%2C__%20_"U4/U6.U5_%20_small_%20_nuclear_%20_ribonucleoprotein_%20_27_%20_kDa_%20_protein"|2p13.3|2016-10-05||2016-03-11|X76302||11017|ENSG00000124380|7931148_%2C__%20_9085842|NM_006857|||CCDS33219|OTTHUMG00000152689|11017||NM_006857|Q8WVK2|ENSG00000124380|uc002sfw.4|SNR27_HUMAN||Q15410|Q8WVK2|mRNA_%20_processing_%20_(GO:0006397)_%7C_RNA_%20_splicing_%20_(GO:0008380)|nucleus_%20_(GO:0005634)|nucleic_%20_acid_%20_binding_%20_(GO:0003676)|_%7C_|_%7C_|_%7C_|true_%7C_true|false_%7C_false|0.6222_%2C_0.3778_%7C_0.6222_%2C_0.3778|false_%7C_false|false_%7C_false|1_%7C_1|_%7C_|false_%7C_false|true_%7C_false|true_%7C_false|SNRNP27:11017_%7C_SNRNP27:11017|true_%7C_false|false_%7C_false|false_%7C_false|true_%7C_false|true_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|true_%7C_true|false_%7C_false|3214822_%7C_397747233|70120910_%7C_70120912|false_%7C_false|false_%7C_false|0_%7C_0|true_%7C_true|0_%7C_0|false_%7C_false|0.661217_%2C_0.338783_%7C_|false_%7C_false|false_%7C_false|false_%7C_false|DIV_%7C_DIV|true_%7C_false|0x05010002000517013e000200_%7C_0x050100020005000002000200|1_%7C_1|false_%7C_false|134_%7C_138|rs3214822_%7C_rs397747233|_%7C_]
+2	70120909	rs3214822	GA	G	722.12	PASS	FUNCOTATION=[SNRNP27|hg19|chr2|70120910|70120910|FIVE_PRIME_UTR||DEL|A|A|-|g.chr2:70120910delA|ENST00000244227.3|+|1|||||0.43640897755610975|GGGAAAAATGAAAGCTGTGTT|SNRNP27_ENST00000409116.1_FIVE_PRIME_FLANK/SNRNP27_ENST00000488986.1_FIVE_PRIME_FLANK|||||||||||||||||||||||||91|biliary_tract(2)_%7C_breast(12)_%7C_central_nervous_system(44)_%7C_large_intestine(11)_%7C_pancreas(22)|||||||X76302|NM_006857.2|NP_006848.1|HGNC:30240|small_%20_nuclear_%20_ribonucleoprotein_%20_U4/U6.U5_%20_subunit_%20_27|Approved|gene_%20_with_%20_protein_%20_product|protein-coding_%20_gene||"small_%20_nuclear_%20_ribonucleoprotein_%20_27kDa_%20_(U4/U6.U5)"_%2C__%20_"small_%20_nuclear_%20_ribonucleoprotein_%2C__%20_U4/U6.U5_%20_27kDa_%20_subunit"|RY1_%2C__%20_U4/U6.U5-27K|"nucleic_%20_acid_%20_binding_%20_protein_%20_RY_%20_1"_%2C__%20_"U4/U6.U5_%20_small_%20_nuclear_%20_ribonucleoprotein_%20_27_%20_kDa_%20_protein"|2p13.3|2016-10-05||2016-03-11|X76302||11017|ENSG00000124380|7931148_%2C__%20_9085842|NM_006857|||CCDS33219|OTTHUMG00000152689|11017||NM_006857|Q8WVK2|ENSG00000124380|uc002sfw.4|SNR27_HUMAN||Q15410|Q8WVK2|mRNA_%20_processing_%20_(GO:0006397)_%7C_RNA_%20_splicing_%20_(GO:0008380)|nucleus_%20_(GO:0005634)|nucleic_%20_acid_%20_binding_%20_(GO:0003676)|_%7C_|_%7C_|_%7C_|true_%7C_true|false_%7C_false|0.6222_%2C_0.3778_%7C_0.6222_%2C_0.3778|false_%7C_false|false_%7C_false|1_%7C_1|_%7C_|false_%7C_false|true_%7C_false|true_%7C_false|SNRNP27:11017_%7C_SNRNP27:11017|true_%7C_false|false_%7C_false|false_%7C_false|true_%7C_false|true_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|false_%7C_false|true_%7C_true|false_%7C_false|3214822_%7C_397747233|70120910_%7C_70120912|false_%7C_false|false_%7C_false|0_%7C_0|true_%7C_true|0_%7C_0|false_%7C_false|0.661217_%2C_0.338783_%7C_|false_%7C_false|false_%7C_false|false_%7C_false|DIV_%7C_DIV|true_%7C_false|0x05010002000517013e000200_%7C_0x050100020005000002000200|1_%7C_1|false_%7C_false|134_%7C_138|rs3214822_%7C_rs397747233|_%7C_]