VCF Tranche filtering in java (#4800)

broadinstitute · Jun 21, 2018 · fb4c7a1 · fb4c7a1
1 parent 9d8fdac
commit fb4c7a1
Show file tree

Hide file tree

Showing 10 changed files with 281 additions and 252 deletions.
diff --git a/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantTrain.java b/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantTrain.java
@@ -103,7 +103,7 @@ protected void onStartup() {
 
     @Override
     protected Object doWork() {
-        final Resource pythonScriptResource = new Resource("training.py", FilterVariantTranches.class);
+        final Resource pythonScriptResource = new Resource("training.py", CNNVariantTrain.class);
         List<String> arguments = new ArrayList<>(Arrays.asList(
                 "--data_dir", inputTensorDir,
                 "--output_dir", outputDir,

diff --git a/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantWriteTensors.java b/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantWriteTensors.java
@@ -121,7 +121,7 @@ protected void onStartup() {
 
     @Override
     protected Object doWork() {
-        final Resource pythonScriptResource = new Resource("training.py", FilterVariantTranches.class);
+        final Resource pythonScriptResource = new Resource("training.py", CNNVariantWriteTensors.class);
         List<String> arguments = new ArrayList<>(Arrays.asList(
                 "--reference_fasta", reference,
                 "--input_vcf", inputVcf,

diff --git a/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/FilterVariantTranches.java b/src/main/java/org/broadinstitute/hellbender/tools/walkers/vqsr/FilterVariantTranches.java
diff --git a/src/main/resources/org/broadinstitute/hellbender/tools/walkers/vqsr/tranches.py b/src/main/resources/org/broadinstitute/hellbender/tools/walkers/vqsr/tranches.py
diff --git a/src/test/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantPipelineTest.java b/src/test/java/org/broadinstitute/hellbender/tools/walkers/vqsr/CNNVariantPipelineTest.java
@@ -95,11 +95,10 @@ public void testTranches() {
         args.add("FilterVariantTranches")
                 .addArgument(StandardArgumentDefinitions.VARIANT_LONG_NAME, trancheVCF)
                 .addArgument(StandardArgumentDefinitions.OUTPUT_LONG_NAME, outputVCF.getAbsolutePath())
-                .addArgument("snp-truth-vcf", snpTruthVCF)
-                .addArgument("indel-truth-vcf", indelTruthVCF)
+                .addArgument("resource", snpTruthVCF)
+                .addArgument("resource", indelTruthVCF)
                 .addArgument("tranche", "99.0")
                 .addArgument("tranche", "95.0")
-                .addArgument("max-sites", "2000")
                 .addArgument("info-key", "VQSLOD");
 
         new Main().instanceMain(args.getArgsArray());

diff --git a/...rg/broadinstitute/hellbender/tools/walkers/vqsr/FilterVariantTranchesIntegrationTest.java b/...rg/broadinstitute/hellbender/tools/walkers/vqsr/FilterVariantTranchesIntegrationTest.java
@@ -0,0 +1,104 @@
+package org.broadinstitute.hellbender.tools.walkers.vqsr;
+
+import org.broadinstitute.hellbender.CommandLineProgramTest;
+import org.broadinstitute.hellbender.utils.test.ArgumentsBuilder;
+import org.broadinstitute.hellbender.utils.test.IntegrationTestSpec;
+import org.broadinstitute.hellbender.cmdline.StandardArgumentDefinitions;
+
+import org.testng.annotations.Test;
+
+import java.util.Arrays;
+import java.io.IOException;
+
+public class FilterVariantTranchesIntegrationTest  extends CommandLineProgramTest {
+
+
+    /**
+     * Run the tool on a small test VCF.
+     */
+    @Test
+    public void testTrancheFiltering() throws IOException {
+        final String trancheVCF = largeFileTestDir + "VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz";
+        final String indelTruthVCF = largeFileTestDir + "VQSR/ALL.wgs.indels_mills_devine_hg19_leftAligned_collapsed_double_hit.sites.20.1M-10M.vcf";
+        final String snpTruthVCF = largeFileTestDir + "VQSR/Omni25_sites_1525_samples.b37.20.1M-10M.vcf";
+
+        final ArgumentsBuilder argsBuilder = new ArgumentsBuilder();
+        argsBuilder.addArgument(StandardArgumentDefinitions.VARIANT_LONG_NAME, trancheVCF)
+                .addArgument(StandardArgumentDefinitions.OUTPUT_LONG_NAME, "%s")
+                .addArgument("resource", snpTruthVCF)
+                .addArgument("resource", indelTruthVCF)
+                .addArgument("tranche", "99.0")
+                .addArgument("info-key", "MIX_SMALL_2D_W_DROPOUT")
+                .addArgument(StandardArgumentDefinitions.ADD_OUTPUT_VCF_COMMANDLINE, "false");
+
+        final IntegrationTestSpec spec = new IntegrationTestSpec(argsBuilder.toString(),
+                    Arrays.asList(largeFileTestDir + "VQSR/expected/g94982_20_1m_10m_tranched_99.vcf"));
+        spec.executeTest("testTrancheFiltering", this);
+
+    }
+
+    @Test
+    public void testTrancheFilteringDuplicate() throws IOException {
+        final String trancheVCF = largeFileTestDir + "VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz";
+        final String indelTruthVCF = largeFileTestDir + "VQSR/ALL.wgs.indels_mills_devine_hg19_leftAligned_collapsed_double_hit.sites.20.1M-10M.vcf";
+        final String snpTruthVCF = largeFileTestDir + "VQSR/Omni25_sites_1525_samples.b37.20.1M-10M.vcf";
+
+        final ArgumentsBuilder argsBuilder = new ArgumentsBuilder();
+        argsBuilder.addArgument(StandardArgumentDefinitions.VARIANT_LONG_NAME, trancheVCF)
+                .addArgument(StandardArgumentDefinitions.OUTPUT_LONG_NAME, "%s")
+                .addArgument("resource", snpTruthVCF)
+                .addArgument("resource", indelTruthVCF)
+                .addArgument("tranche", "99.0")
+                .addArgument("tranche", "99.0")
+                .addArgument("info-key", "MIX_SMALL_2D_W_DROPOUT")
+                .addArgument(StandardArgumentDefinitions.ADD_OUTPUT_VCF_COMMANDLINE, "false");
+
+        final IntegrationTestSpec spec = new IntegrationTestSpec(argsBuilder.toString(),
+                Arrays.asList(largeFileTestDir + "VQSR/expected/g94982_20_1m_10m_tranched_99.vcf"));
+        spec.executeTest("testTrancheFilteringDuplicate", this);
+    }
+
+    @Test
+    public void testTrancheFilteringTranches() throws IOException {
+        final String trancheVCF = largeFileTestDir + "VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz";
+        final String indelTruthVCF = largeFileTestDir + "VQSR/ALL.wgs.indels_mills_devine_hg19_leftAligned_collapsed_double_hit.sites.20.1M-10M.vcf";
+        final String snpTruthVCF = largeFileTestDir + "VQSR/Omni25_sites_1525_samples.b37.20.1M-10M.vcf";
+
+        final ArgumentsBuilder argsBuilder = new ArgumentsBuilder();
+        argsBuilder.addArgument(StandardArgumentDefinitions.VARIANT_LONG_NAME, trancheVCF)
+                .addArgument(StandardArgumentDefinitions.OUTPUT_LONG_NAME, "%s")
+                .addArgument("resource", snpTruthVCF)
+                .addArgument("resource", indelTruthVCF)
+                .addArgument("tranche", "95.0")
+                .addArgument("tranche", "99.0")
+                .addArgument("tranche", "99.9")
+                .addArgument("info-key", "MIX_SMALL_2D_W_DROPOUT")
+                .addArgument(StandardArgumentDefinitions.ADD_OUTPUT_VCF_COMMANDLINE, "false");
+
+        final IntegrationTestSpec spec = new IntegrationTestSpec(argsBuilder.toString(),
+                    Arrays.asList(largeFileTestDir + "VQSR/expected/g94982_20_1m_10m_tranched_95_99_99.9.vcf"));
+        spec.executeTest("testTrancheFilteringTranches", this);
+    }
+
+    @Test
+    public void testTrancheFilteringTranchesOrder() throws IOException {
+        final String trancheVCF = largeFileTestDir + "VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz";
+        final String indelTruthVCF = largeFileTestDir + "VQSR/ALL.wgs.indels_mills_devine_hg19_leftAligned_collapsed_double_hit.sites.20.1M-10M.vcf";
+        final String snpTruthVCF = largeFileTestDir + "VQSR/Omni25_sites_1525_samples.b37.20.1M-10M.vcf";
+        final ArgumentsBuilder argsBuilder = new ArgumentsBuilder();
+        argsBuilder.addArgument(StandardArgumentDefinitions.VARIANT_LONG_NAME, trancheVCF)
+                .addArgument(StandardArgumentDefinitions.OUTPUT_LONG_NAME, "%s")
+                .addArgument("resource", snpTruthVCF)
+                .addArgument("resource", indelTruthVCF)
+                .addArgument("tranche", "99.9")
+                .addArgument("tranche", "95.0")
+                .addArgument("tranche", "99.0")
+                .addArgument("info-key", "MIX_SMALL_2D_W_DROPOUT")
+                .addArgument(StandardArgumentDefinitions.ADD_OUTPUT_VCF_COMMANDLINE, "false");
+
+        final IntegrationTestSpec spec = new IntegrationTestSpec(argsBuilder.toString(),
+                Arrays.asList(largeFileTestDir + "VQSR/expected/g94982_20_1m_10m_tranched_95_99_99.9.vcf"));
+        spec.executeTest("testTrancheFilteringTranchesOrder", this);
+    }
+
+}
diff --git a/src/test/resources/large/VQSR/expected/g94982_20_1m_10m_tranched_95_99_99.9.vcf b/src/test/resources/large/VQSR/expected/g94982_20_1m_10m_tranched_95_99_99.9.vcf
diff --git a/src/test/resources/large/VQSR/expected/g94982_20_1m_10m_tranched_99.vcf b/src/test/resources/large/VQSR/expected/g94982_20_1m_10m_tranched_99.vcf
diff --git a/src/test/resources/large/VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz b/src/test/resources/large/VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz
diff --git a/src/test/resources/large/VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz.tbi b/src/test/resources/large/VQSR/g94982_20_1m_10m_python_2dcnn.vcf.gz.tbi