Now non-locatable data sources can create funcotations again.

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

@@ -135,6 +135,12 @@ public String getVersion() {
         return version;
     }
 
+    /**
+     * @return {@code True} if this {@link DataSourceFuncotationFactory} requires features to create {@link Funcotation}s.  {@code False} otherwise.
+     */
+    @VisibleForTesting
+    public boolean requiresFeatures() { return true; }
+
     /**
      * @return An ordered {@link LinkedHashSet} of the names of annotations that this Data Source supports.
      */
@@ -169,11 +175,15 @@ public List<Funcotation> createFuncotations(final VariantContext variant, final
         Utils.nonNull(referenceContext);
         Utils.nonNull(featureContext);
 
-        // Query this funcotation factory to get the list of overlapping features.
-        final List<Feature> featureList = queryFeaturesFromFeatureContext(featureContext);
-
         final List<Funcotation> outputFuncotations;
 
+        // Query this funcotation factory to get the list of overlapping features.
+        // NOTE: This will only get features that are LOCATABLE!
+        //       This corresponds to requiresFeatures() returning `True`.
+        final List<Feature> featureList = requiresFeatures() ?
+                queryFeaturesFromFeatureContext(featureContext) :
+                Collections.emptyList();
+
         // If our featureList is compatible with this DataSourceFuncotationFactory, then we make our funcotations:
         if ( isFeatureListCompatible(featureList) ) {
 
@@ -203,12 +213,22 @@ public List<Funcotation> createFuncotations(final VariantContext variant, final
      * Checks to see if the given featureList is compatible with this {@link DataSourceFuncotationFactory}.
      * Cues off of the feature type in the feature list and whether the given list contains any non-null features.
      * This method acts as a sanity-check before attempting to do any annotations on features.
+     * If this {@link DataSourceFuncotationFactory} does not require features as per {@link #requiresFeatures()}, then
+     * this method will always return {@code True}.
      * @param featureList {@link List} of {@link Feature} that might be applicable to this {@link DataSourceFuncotationFactory} for annotation.
      * @return {@code true} if the given {@code featureList} contains at least one non-null feature of type {@link #getAnnotationFeatureClass()}; {@code false} otherwise.
      */
     private boolean isFeatureListCompatible(final List<Feature> featureList) {
-        // Make sure these features can be annotated by this DataSourceFuncotationFactory:
+        // Make sure these features can be annotated by this DataSourceFuncotationFactory.
         // NOTE: We only check the first non-null element of the list for feature type:
+
+        // The feature list is compatible if we found a compatible feature
+        // OR
+        // if this DataSourceFuncotationFactory does not require features.
+        if ( !requiresFeatures() ) {
+            return true;
+        }
+
         boolean foundCompatibleFeature = false;
         for ( final Feature f : featureList ) {
             if (f != null) {

src/main/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/cosmic/CosmicFuncotationFactory.java

@@ -202,6 +202,15 @@ public String getName() {
         return name;
     }
 
+    @Override
+    /**
+     * {@inheritDoc}
+     * Since {@link CosmicFuncotationFactory} primarily keys off a gene name, we don't actually need
+     * any features to create annotations.
+     */
+    @VisibleForTesting
+    public boolean requiresFeatures() { return false; }
+
     @Override
     public LinkedHashSet<String> getSupportedFuncotationFields() {
         return supportedFields;
@@ -250,7 +259,7 @@ protected List<Funcotation> createFuncotationsOnVariant(final VariantContext var
         // Then query our DB for matches on the gene name.
         // Then grab Genome position / Protein position and see if we overlap.
         // If any do, we create our CosmicFuncotation
-        for (  final GencodeFuncotation gencodeFuncotation : gencodeFuncotations ) {
+        for ( final GencodeFuncotation gencodeFuncotation : gencodeFuncotations ) {
             final String geneName = gencodeFuncotation.getHugoSymbol();
 
             final SimpleInterval genomePosition = new SimpleInterval(variant.getContig(), variant.getStart(), variant.getEnd());

src/main/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/xsv/SimpleKeyXsvFuncotationFactory.java

@@ -1,5 +1,6 @@
 package org.broadinstitute.hellbender.tools.funcotator.dataSources.xsv;
 
+import com.google.common.annotations.VisibleForTesting;
 import htsjdk.tribble.Feature;
 import htsjdk.variant.variantcontext.Allele;
 import htsjdk.variant.variantcontext.VariantContext;
@@ -177,6 +178,15 @@ public String getName() {
         return name;
     }
 
+    @Override
+    /**
+     * {@inheritDoc}
+     * Since {@link SimpleKeyXsvFuncotationFactory} keys off a gene name or transcript ID, we don't actually need
+     * any features to create annotations.
+     */
+    @VisibleForTesting
+    public boolean requiresFeatures() { return false; }
+
     @Override
     public LinkedHashSet<String> getSupportedFuncotationFields() {
         return new LinkedHashSet<>(annotationColumnNames);

src/test/java/org/broadinstitute/hellbender/tools/funcotator/FuncotatorIntegrationTest.java

@@ -86,6 +86,9 @@ public class FuncotatorIntegrationTest extends CommandLineProgramTest {
     private static final String PIK3CA_DBSNP_DS          = toolsTestDir + "funcotator" + File.separator + "small_pik3ca_dbsnp_ds";
     private static final String MAF_DBSNP_TEST           = toolsTestDir + "funcotator" + File.separator + "maf_dbsnp_test_input.vcf";
 
+    // Non-locatable funcotation file:
+    private static final String NON_LOCATABLE_FUNCOTATED_INPUT_VCF = toolsTestDir + "funcotator" + File.separator + "non_locatable_proof_input.vcf";
+
     private static final List<String> VCF_FIELDS_GENCODE_19_DS = Arrays.asList("Gencode_19_hugoSymbol","Gencode_19_ncbiBuild","Gencode_19_chromosome","Gencode_19_start","Gencode_19_end","Gencode_19_variantClassification","Gencode_19_variantType","Gencode_19_refAllele","Gencode_19_tumorSeqAllele1","Gencode_19_tumorSeqAllele2","Gencode_19_genomeChange","Gencode_19_annotationTranscript","Gencode_19_transcriptStrand","Gencode_19_transcriptExon","Gencode_19_transcriptPos","Gencode_19_cDnaChange","Gencode_19_codonChange","Gencode_19_proteinChange","Gencode_19_gcContent","Gencode_19_referenceContext","Gencode_19_otherTranscripts");//,"Achilles_Top_Genes","CGC_Name","CGC_GeneID","CGC_Chr","CGC_Chr_Band","CGC_Cancer_Somatic_Mut","CGC_Cancer_Germline_Mut","CGC_Tumour_Types__(Somatic_Mutations)","CGC_Tumour_Types_(Germline_Mutations)","CGC_Cancer_Syndrome","CGC_Tissue_Type","CGC_Cancer_Molecular_Genetics","CGC_Mutation_Type","CGC_Translocation_Partner","CGC_Other_Germline_Mut","CGC_Other_Syndrome/Disease","ClinVar_HGMD_ID","ClinVar_SYM","ClinVar_TYPE","ClinVar_ASSEMBLY","ClinVar_rs","Cosmic_overlapping_mutations","CosmicFusion_fusion_genes","CosmicFusion_fusion_id","CosmicTissue_total_alterations_in_gene","CosmicTissue_tissue_types_affected","DNARepairGenes_Activity_linked_to_OMIM","DNARepairGenes_Chromosome_location_linked_to_NCBI_MapView","DNARepairGenes_Accession_number_linked_to_NCBI_Entrez","Familial_Cancer_Genes_Syndrome","Familial_Cancer_Genes_Synonym","Familial_Cancer_Genes_Reference","Gencode_XHGNC_hgnc_id","Gencode_XRefSeq_mRNA_id","Gencode_XRefSeq_prot_acc","HGNC_HGNC_ID","HGNC_Approved_Name","HGNC_Status","HGNC_Locus_Type","HGNC_Locus_Group","HGNC_Previous_Symbols","HGNC_Previous_Name","HGNC_Synonyms","HGNC_Name_Synonyms","HGNC_Chromosome","HGNC_Date_Modified","HGNC_Date_Symbol_Changed","HGNC_Date_Name_Changed","HGNC_Accession_Numbers","HGNC_Enzyme_IDs","HGNC_Entrez_Gene_ID","HGNC_Ensembl_Gene_ID","HGNC_Pubmed_IDs","HGNC_RefSeq_IDs","HGNC_Gene_Family_ID","HGNC_Gene_Family_Name","HGNC_CCDS_IDs","HGNC_Vega_ID","HGNC_Entrez_Gene_ID(supplied_by_NCBI)","HGNC_OMIM_ID(supplied_by_OMIM)","HGNC_RefSeq(supplied_by_NCBI)","HGNC_UniProt_ID(supplied_by_UniProt)","HGNC_Ensembl_ID(supplied_by_Ensembl)","HGNC_UCSC_ID(supplied_by_UCSC)","Oreganno_Build","Oreganno_ID","Oreganno_Values","Simple_Uniprot_uniprot_entry_name","Simple_Uniprot_DrugBank","Simple_Uniprot_alt_uniprot_accessions","Simple_Uniprot_uniprot_accession","Simple_Uniprot_GO_Biological_Process","Simple_Uniprot_GO_Cellular_Component","Simple_Uniprot_GO_Molecular_Function","dbSNP_ASP","dbSNP_ASS","dbSNP_CAF","dbSNP_CDA","dbSNP_CFL","dbSNP_COMMON","dbSNP_DSS","dbSNP_G5","dbSNP_G5A","dbSNP_GENEINFO","dbSNP_GNO","dbSNP_HD","dbSNP_INT","dbSNP_KGPhase1","dbSNP_KGPhase3","dbSNP_LSD","dbSNP_MTP","dbSNP_MUT","dbSNP_NOC","dbSNP_NOV","dbSNP_NSF","dbSNP_NSM","dbSNP_NSN","dbSNP_OM","dbSNP_OTH","dbSNP_PM","dbSNP_PMC","dbSNP_R3","dbSNP_R5","dbSNP_REF","dbSNP_RS","dbSNP_RSPOS","dbSNP_RV","dbSNP_S3D","dbSNP_SAO","dbSNP_SLO","dbSNP_SSR","dbSNP_SYN","dbSNP_TOPMED","dbSNP_TPA","dbSNP_U3","dbSNP_U5","dbSNP_VC","dbSNP_VLD","dbSNP_VP","dbSNP_WGT","dbSNP_WTD","dbSNP_dbSNPBuildID");
     private static final List<String> VCF_FIELDS_GENCODE_28_DS = Arrays.asList("Gencode_28_hugoSymbol","Gencode_28_ncbiBuild","Gencode_28_chromosome","Gencode_28_start","Gencode_28_end","Gencode_28_variantClassification","Gencode_28_variantType","Gencode_28_refAllele","Gencode_28_tumorSeqAllele1","Gencode_28_tumorSeqAllele2","Gencode_28_genomeChange","Gencode_28_annotationTranscript","Gencode_28_transcriptStrand","Gencode_28_transcriptExon","Gencode_28_transcriptPos","Gencode_28_cDnaChange","Gencode_28_codonChange","Gencode_28_proteinChange","Gencode_28_gcContent","Gencode_28_referenceContext","Gencode_28_otherTranscripts");//,"Achilles_Top_Genes","CGC_Name","CGC_GeneID","CGC_Chr","CGC_Chr_Band","CGC_Cancer_Somatic_Mut","CGC_Cancer_Germline_Mut","CGC_Tumour_Types__(Somatic_Mutations)","CGC_Tumour_Types_(Germline_Mutations)","CGC_Cancer_Syndrome","CGC_Tissue_Type","CGC_Cancer_Molecular_Genetics","CGC_Mutation_Type","CGC_Translocation_Partner","CGC_Other_Germline_Mut","CGC_Other_Syndrome/Disease","ClinVar_HGMD_ID","ClinVar_SYM","ClinVar_TYPE","ClinVar_ASSEMBLY","ClinVar_rs","Cosmic_overlapping_mutations","CosmicFusion_fusion_genes","CosmicFusion_fusion_id","CosmicTissue_total_alterations_in_gene","CosmicTissue_tissue_types_affected","DNARepairGenes_Activity_linked_to_OMIM","DNARepairGenes_Chromosome_location_linked_to_NCBI_MapView","DNARepairGenes_Accession_number_linked_to_NCBI_Entrez","Familial_Cancer_Genes_Syndrome","Familial_Cancer_Genes_Synonym","Familial_Cancer_Genes_Reference","Gencode_XHGNC_hgnc_id","Gencode_XRefSeq_mRNA_id","Gencode_XRefSeq_prot_acc","HGNC_HGNC_ID","HGNC_Approved_Name","HGNC_Status","HGNC_Locus_Type","HGNC_Locus_Group","HGNC_Previous_Symbols","HGNC_Previous_Name","HGNC_Synonyms","HGNC_Name_Synonyms","HGNC_Chromosome","HGNC_Date_Modified","HGNC_Date_Symbol_Changed","HGNC_Date_Name_Changed","HGNC_Accession_Numbers","HGNC_Enzyme_IDs","HGNC_Entrez_Gene_ID","HGNC_Ensembl_Gene_ID","HGNC_Pubmed_IDs","HGNC_RefSeq_IDs","HGNC_Gene_Family_ID","HGNC_Gene_Family_Name","HGNC_CCDS_IDs","HGNC_Vega_ID","HGNC_Entrez_Gene_ID(supplied_by_NCBI)","HGNC_OMIM_ID(supplied_by_OMIM)","HGNC_RefSeq(supplied_by_NCBI)","HGNC_UniProt_ID(supplied_by_UniProt)","HGNC_Ensembl_ID(supplied_by_Ensembl)","HGNC_UCSC_ID(supplied_by_UCSC)","Oreganno_Build","Oreganno_ID","Oreganno_Values","Simple_Uniprot_uniprot_entry_name","Simple_Uniprot_DrugBank","Simple_Uniprot_alt_uniprot_accessions","Simple_Uniprot_uniprot_accession","Simple_Uniprot_GO_Biological_Process","Simple_Uniprot_GO_Cellular_Component","Simple_Uniprot_GO_Molecular_Function","dbSNP_ASP","dbSNP_ASS","dbSNP_CAF","dbSNP_CDA","dbSNP_CFL","dbSNP_COMMON","dbSNP_DSS","dbSNP_G5","dbSNP_G5A","dbSNP_GENEINFO","dbSNP_GNO","dbSNP_HD","dbSNP_INT","dbSNP_KGPhase1","dbSNP_KGPhase3","dbSNP_LSD","dbSNP_MTP","dbSNP_MUT","dbSNP_NOC","dbSNP_NOV","dbSNP_NSF","dbSNP_NSM","dbSNP_NSN","dbSNP_OM","dbSNP_OTH","dbSNP_PM","dbSNP_PMC","dbSNP_R3","dbSNP_R5","dbSNP_REF","dbSNP_RS","dbSNP_RSPOS","dbSNP_RV","dbSNP_S3D","dbSNP_SAO","dbSNP_SLO","dbSNP_SSR","dbSNP_SYN","dbSNP_TOPMED","dbSNP_TPA","dbSNP_U3","dbSNP_U5","dbSNP_VC","dbSNP_VLD","dbSNP_VP","dbSNP_WGT","dbSNP_WTD","dbSNP_dbSNPBuildID");
     private static final List<String> MAF_FIELDS_GENCODE_DS = Arrays.asList(MafOutputRendererConstants.FieldName_Hugo_Symbol, MafOutputRendererConstants.FieldName_NCBI_Build, MafOutputRendererConstants.FieldName_Chromosome,
@@ -1677,5 +1680,57 @@ public void testEnsureDbSnpInMaf() {
                 .collect(Collectors.toList());
         Assert.assertEquals(guessDbSnpValStatus, gtDbSnpValStatus);
     }
+
+    @Test
+    public void testCanCreateNonLocatableFuncotations() {
+
+//        FuncotatorTestConstants.NON_TRIVIAL_DATA_VALIDATION_TEST_HG19_DATA_SET_1_EXPECTED_OUTPUT
+
+        final File outputFile = createTempFile(tmpOutDir + File.separator + NON_LOCATABLE_FUNCOTATED_INPUT_VCF + ".funcotator", ".vcf");
+
+        final ArgumentsBuilder arguments = createBaselineArgumentsForFuncotator(
+                NON_LOCATABLE_FUNCOTATED_INPUT_VCF,
+                outputFile,
+                b37Reference,
+                FuncotatorTestConstants.FUNCOTATOR_DATA_SOURCES_MAIN_FOLDER,
+                FuncotatorTestConstants.REFERENCE_VERSION_HG19,
+                FuncotatorArgumentDefinitions.OutputFormatType.VCF,
+                true);
+
+        // Run the tool with our args:
+        runCommandLine(arguments);
+
+        // ===============================
+
+        final Pair<VCFHeader, List<VariantContext>> vcfInfo
+                = VariantContextTestUtils.readEntireVCFIntoMemory(outputFile.getAbsolutePath());
+
+        final VCFInfoHeaderLine funcotationHeaderLine = vcfInfo.getLeft().getInfoHeaderLine(VcfOutputRenderer.FUNCOTATOR_VCF_FIELD_NAME);
+        final String[] funcotationFieldNames = FuncotatorUtils.extractFuncotatorKeysFromHeaderDescription(funcotationHeaderLine.getDescription());
+
+        final VariantContext variant = vcfInfo.getRight().get(0);
+
+        final Map<Allele, FuncotationMap> alleleToFuncotationMap =
+                FuncotatorUtils.createAlleleToFuncotationMapFromFuncotationVcfAttribute(
+                        funcotationFieldNames,
+                        variant,
+                        "Gencode_19_annotationTranscript",
+                        "TEST");
+
+        // Make sure we get the correct transcript here:
+        Assert.assertEquals(alleleToFuncotationMap.get(variant.getAlternateAllele(0)).getTranscriptList().size(), 1);
+
+        // Now get the transcript annotations:
+        final List<Funcotation> funcotations = alleleToFuncotationMap.get(variant.getAlternateAllele(0)).get("ENST00000379410.3");
+
+        // Now assert that we got what we should have gotten from a few HGNC (non-locatable data source) fields:
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_HGNC_ID"), "HGNC:25284");
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_Status"), "Approved");
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_Locus_Type"), "gene_%20_with_%20_protein_%20_product");
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_Locus_Group"), "protein-coding_%20_gene");
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_Previous_Name"), "\"pleckstrin_%20_homology_%20_domain_%20_containing_%2C__%20_family_%20_N_%20_member_%20_1\"");
+        Assert.assertEquals(funcotations.get(0).getField("HGNC_Synonyms"), "DKFZP434H2010");
+
+    }
 }
 

src/test/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/cosmic/CosmicFuncotationFactoryUnitTest.java

@@ -221,6 +221,11 @@ private Object[][] provideForTestCreateFuncotations() {
     //==================================================================================================================
     // Tests:
 
+    @Test
+    public void testRequiresFeatures() {
+        Assert.assertFalse(new CosmicFuncotationFactory(PATH_TO_TEST_DB).requiresFeatures());
+    }
+
     @Test(dataProvider = "provideDataForTestProteinPositionRegex")
     public void testPositionRegex(final Pattern regex, final List<String> dbPositions, final int expectedNumResults) {
 

src/test/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/gencode/GencodeFuncotationFactoryUnitTest.java

@@ -2503,4 +2503,9 @@ public void testCreateFuncotationsWithFlanks(final String expectedGeneName,
             Assert.assertEquals(funcotation.getHugoSymbol(), expectedGeneName, "Gene name not correct");
         }
     }
+
+    @Test
+    public void testRequiresFeatures() {
+        Assert.assertTrue(testMuc16SnpCreateFuncotationsFuncotationFactory.requiresFeatures());
+    }
 }

src/test/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/vcf/VcfFuncotationFactoryUnitTest.java

@@ -191,6 +191,11 @@ public void testGetAnnotationFeatureClass() {
         Assert.assertEquals(vcfFuncotationFactory.getAnnotationFeatureClass(), VariantContext.class);
     }
 
+    @Test
+    public void testRequiresFeatures() {
+        Assert.assertTrue(createVcfFuncotationFactory(FACTORY_NAME, FACTORY_VERSION, IOUtils.getPath(FuncotatorTestConstants.VARIANT_FILE_HG19_CHR3)).requiresFeatures());
+    }
+
     @Test
     public void testGetType() {
         final VcfFuncotationFactory vcfFuncotationFactory = createVcfFuncotationFactory(FACTORY_NAME, FACTORY_VERSION, IOUtils.getPath(FuncotatorTestConstants.VARIANT_FILE_HG19_CHR3));

src/test/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/xsv/LocatableXsvFuncotationFactoryUnitTest.java

@@ -248,6 +248,11 @@ private Object[][] provideForTestSetSupportedFuncotationFields() {
     //==================================================================================================================
     // Tests:
 
+    @Test
+    public void testRequiresFeatures() {
+        Assert.assertTrue(new LocatableXsvFuncotationFactory(LocatableXsvFuncotationFactory.DEFAULT_NAME, DataSourceFuncotationFactory.DEFAULT_VERSION_STRING, new LinkedHashMap<>(), null).requiresFeatures());
+    }
+
     @Test(dataProvider = "provideForTestGetName")
     public void testGetName(final String name, final String expected) {
         final LocatableXsvFuncotationFactory locatableXsvFuncotationFactory;

src/test/java/org/broadinstitute/hellbender/tools/funcotator/dataSources/xsv/SimpleKeyXsvFuncotationFactoryUnitTest.java

@@ -7,14 +7,14 @@
 import org.broadinstitute.hellbender.engine.ReferenceContext;
 import org.broadinstitute.hellbender.engine.ReferenceDataSource;
 import org.broadinstitute.hellbender.exceptions.GATKException;
+import org.broadinstitute.hellbender.testutils.FuncotatorReferenceTestUtils;
 import org.broadinstitute.hellbender.tools.funcotator.Funcotation;
 import org.broadinstitute.hellbender.tools.funcotator.FuncotatorTestConstants;
 import org.broadinstitute.hellbender.tools.funcotator.dataSources.TableFuncotation;
 import org.broadinstitute.hellbender.tools.funcotator.dataSources.gencode.GencodeFuncotation;
 import org.broadinstitute.hellbender.tools.funcotator.dataSources.gencode.GencodeFuncotationBuilder;
 import org.broadinstitute.hellbender.utils.SimpleInterval;
 import org.broadinstitute.hellbender.utils.io.IOUtils;
-import org.broadinstitute.hellbender.testutils.FuncotatorReferenceTestUtils;
 import org.testng.Assert;
 import org.testng.annotations.DataProvider;
 import org.testng.annotations.Test;
@@ -425,4 +425,18 @@ public void testCreateFuncotationsIgnoresTranscriptVersions() {
         final TableFuncotation tableFuncotation = (TableFuncotation)funcotation;
         Assert.assertEquals(tableFuncotation.get(defaultName + "_Beatle"), "Harrison", "Wrong value for the Beatle column in returned funcotation");
     }
+
+    @Test
+    public void testRequiresFeatures() {
+        final SimpleKeyXsvFuncotationFactory simpleKeyXsvFuncotationFactory = new SimpleKeyXsvFuncotationFactory(
+                defaultName,
+                IOUtils.getPath(FuncotatorTestConstants.XSV_CSV_FILE_PATH),
+                "VERSION",
+                ",",
+                0,
+                SimpleKeyXsvFuncotationFactory.XsvDataKeyType.GENE_NAME
+        );
+
+        Assert.assertFalse(simpleKeyXsvFuncotationFactory.requiresFeatures());
+    }
 }