Merge branch 'main' into refactoring/output-format

pom.xml

@@ -152,7 +152,7 @@
 		<dependency>
 			<groupId>com.github.samtools</groupId>
 			<artifactId>htsjdk</artifactId>
-			<version>2.21.3</version>
+			<version>3.0.5</version>
 		</dependency>
 
 		<dependency>

src/main/java/genepi/imputationserver/steps/QualityControlCommand.java

@@ -235,8 +235,9 @@ private boolean analyzeFiles() {
 		double sampleCallrate = panel.getQcFilterByKey("sampleCallrate");
 		double mixedGenotypesChrX = panel.getQcFilterByKey("mixedGenotypeschrX");
 		int strandFlips = (int) (panel.getQcFilterByKey("strandFlips"));
+		int alleleSwitches = (int) (panel.getQcFilterByKey("alleleSwitches"));
 		String ranges = panel.getRange();
-
+		
 		if (ranges != null) {
 			HashSet<RangeEntry> rangeEntries = new HashSet<RangeEntry>();
 
@@ -379,6 +380,15 @@ else if (task.getStrandFlipSimple() + task.getStrandFlipAndAlleleSwitch() > stra
 			return false;
 		}
 
+		// Check if too many allele switches are detected
+		else if (task.getAlleleSwitch() + task.getStrandFlipAndAlleleSwitch() > alleleSwitches) {
+			text.append("<br><b>Error:</b> More than " + alleleSwitches
+					+ " allele switches have been detected. Imputation cannot be started!");
+			context.error(text.toString());
+
+			return false;
+		}
+
 		else if (task.isChrXMissingRate()) {
 			text.add(
 					"\n<b>Error:</b> Chromosome X nonPAR region includes > 10 % mixed genotypes. Imputation cannot be started!");

src/main/java/genepi/imputationserver/steps/fastqc/StatisticsTask.java

@@ -83,7 +83,7 @@ public class StatisticsTask implements ITask {
 
 	private boolean chrXMissingRate = false;
 	private boolean chrXPloidyError = false;
-
+ 
 	// chunk results
 	private int removedChunksSnps;
 	private int removedChunksOverlap;
@@ -474,12 +474,11 @@ else if (GenomicTools.alleleSwitch(snp, refSnp)) {
 				if (insideChunk) {
 
 					alleleSwitch++;
-					/*
-					 * logWriter.write("Allele switch" + snp.getID() + "\t" + chr + ":"+
-					 * snp.getStart() + "\t" + "ref: " + legendRef + "/" + legendAlt + "; data: " +
-					 * studyRef + "/" + studyAlt + ")");
-					 */
+					filtered++;
+					excludedSnpsWriter.write(snp + "\t" + "Allele switch" + "\t" + "Ref:" + legendRef + "/" + legendAlt);
+
 				}
+				return;
 
 			}
 

src/main/java/genepi/imputationserver/util/RefPanel.java

@@ -15,6 +15,7 @@
 
 public class RefPanel {
 
+	public static final String ALLELE_SWITCHES = String.valueOf(Integer.MAX_VALUE);
 	public static final String STRAND_FLIPS = "100";
 	public static final String SAMPLE_CALL_RATE = "0.5";
 	public static final String MIN_SNPS = "3";
@@ -57,6 +58,7 @@ public RefPanel() {
 		defaultQcFilter.put("sampleCallrate", SAMPLE_CALL_RATE);
 		defaultQcFilter.put("mixedGenotypeschrX", CHR_X_MIXED_GENOTYPES);
 		defaultQcFilter.put("strandFlips", STRAND_FLIPS);
+		defaultQcFilter.put("alleleSwitches", ALLELE_SWITCHES);
 	}
 
 	public String getId() {
@@ -127,7 +129,7 @@ public int getSamplesByPopulation(String population) {
 		if (population == null) {
 			return 0;
 		}
-		
+
 		RefPanelPopulation panelPopulation = getPopulation(population);
 		if (panelPopulation != null) {
 			return panelPopulation.getSamples();
@@ -147,7 +149,7 @@ public List<RefPanelPopulation> getPopulations() {
 	public boolean supportsPopulation(String population) {
 		if (population == null || population.equals("")) {
 			return true;
-		}		
+		}
 		return (getPopulation(population) != null);
 	}
 
@@ -169,15 +171,24 @@ public Map<String, String> getQcFilter() {
 	}
 
 	public double getQcFilterByKey(String key) {
+
 		if (qcFilter == null) {
 			qcFilter = defaultQcFilter;
 		}
-		String n = qcFilter.get(key);
-		if (n != null) {
-			return Double.parseDouble(n);
-		} else {
+		Object n = qcFilter.get(key);
+
+		if (n == null) {
 			return Double.parseDouble(defaultQcFilter.get(key));
 		}
+
+		if (n instanceof String) {
+			return Double.parseDouble((String) n);
+		} else if (n instanceof Number) {
+			return ((Number) n).doubleValue();
+		} else {
+			throw new IllegalArgumentException("Value associated with key is neither a String nor a Number: " + n);
+		}
+
 	}
 
 	public void setQcFilter(Map<String, String> qcFilter) {
@@ -257,7 +268,7 @@ public static RefPanel fromProperties(Object properties) throws IOException {
 		if (map.get("reference_build") != null) {
 			panel.setBuild(map.get("reference_build").toString());
 		}
-		
+
 		if (map.get("build") != null) {
 			panel.setBuild(map.get("build").toString());
 		}

src/test/java/genepi/imputationserver/steps/QualityControlCommandTest.java

@@ -443,6 +443,27 @@ public void testQcStatisticsDontAllowStrandFlips() throws Exception {
 
 	}
 
+	@Test
+	public void testQcStatisticsDontAllowAlleleSwitches() throws Exception {
+
+		String panels = "test-data/configs/hapmap-3chr/panels.txt";
+		String inputFolder = "test-data/data/simulated-chip-3chr-imputation-switches";
+
+		// create workflow context
+		RefPanel panel = RefPanel.loadFromYamlFile(panels, "hapmap2-qcfilter-alleleswitches");
+
+		QualityControlCommand command = buildCommand(inputFolder);
+		command.setRefPanel(panel);
+		assertEquals(-1, (int) command.call());
+
+		CloudgeneReport log = new CloudgeneReport(CLOUDGENE_LOG);
+
+		// check statistics
+		assertTrue(log.hasInMemory("Excluded sites in total: 121,176"));
+		assertTrue(log.hasInMemory("Allele switch: 118,209"));
+		assertTrue(log.hasInMemory("No chunks passed the QC step. Imputation cannot be started!"));
+	}
+
 	@Test
 	public void testQcStatisticsFilterOverlap() throws Exception {
 

test-data/configs/hapmap-3chr/panels.txt

@@ -64,6 +64,21 @@
       strandFlips: 100
       minSnps: 1000
 
+  - id: hapmap2-qcfilter-alleleswitches
+    genotypes: ${app_local_folder}/ref-panels/hapmap_r22.chr$chr.CEU.hg19.msav
+    legend: ${app_local_folder}/ref-panels/hapmap_r22.chr$chr.CEU.hg19_impute.legend.gz
+    mapEagle: ${app_local_folder}/ref-panels/genetic_map_hg19_chr1.txt
+    refEagle: ${app_local_folder}/ref-panels/hapmap_r22.eagle/hapmap_r22.chr$chr.CEU.hg19.recode.bcf
+    populations:
+      - id: eur
+        name: EUR
+        samples: 60
+      - id: mixed
+        name: Mixed
+        samples: -1
+    qcFilter:
+      alleleSwitches: 33  
+
   - id: hapmap2-qcfilter-low-callrate
     reference_build: hg19
     genotypes: ${app_local_folder}/ref-panels/hapmap_r22.chr$chr.CEU.hg19.m3vcf.gz