zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #1318 from broadinstitute/rhl_va_hrun_exception 

Set --reference_window_stop if homopolymer is greater than  window size
This commit is contained in:
Ron Levine 2016-03-14 18:22:40 -04:00
parent 2baf7d8c4e 6d7ec7377c
commit 4efde50225
2 changed files with 31 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/annotator/HomopolymerRun.java
View File

@ -51,6 +51,7 @@
package org.broadinstitute.gatk.tools.walkers.annotator;
import org.apache.log4j.Logger;
import org.broadinstitute.gatk.utils.contexts.AlignmentContext;
import org.broadinstitute.gatk.utils.contexts.ReferenceContext;
import org.broadinstitute.gatk.utils.refdata.RefMetaDataTracker;
@ -83,7 +84,11 @@ import java.util.Map;
*/
public class HomopolymerRun extends InfoFieldAnnotation implements ExperimentalAnnotation {
private final static Logger logger = Logger.getLogger(HardyWeinberg.class);
private boolean ANNOTATE_INDELS = true;
private final static int BAD_RUN_LENGTH = -1;
public Map<String, Object> annotate(final RefMetaDataTracker tracker,
final AnnotatorCompatible walker,
@ -103,6 +108,9 @@ public class HomopolymerRun extends InfoFieldAnnotation implements ExperimentalA
} else {
return null;
}
if ( run == BAD_RUN_LENGTH )
return null;
Map<String, Object> map = new HashMap<>();
map.put(getKeyNames().get(0), String.format("%d", run));
@ -149,15 +157,13 @@ public class HomopolymerRun extends InfoFieldAnnotation implements ExperimentalA
GenomeLoc window = ref.getWindow();
int refBasePos = (int) (locus.getStart() - window.getStart())+1;
if ( vc.isSimpleDeletion() ) {
// check that deleted bases are the same
byte dBase = bases[refBasePos];
for ( int i = 0; i < vc.getReference().length(); i ++ ) {
if ( bases[refBasePos+i] != dBase ) {
return 0;
}
if ( refBasePos + vc.getReference().length() - 1 >= bases.length ) {
logger.warn("Encountered a homopolymer at " + locus.toString() + " longer than the tool's default window size, so the position was skipped. " +
"To process this position, add --reference_window_stop to your command with a value equal or greater than " + vc.getReference().length());
return BAD_RUN_LENGTH;
}
return computeHomopolymerRun(dBase, ref);
return computeHomopolymerRun(bases[refBasePos], ref);
} else {
// check that inserted bases are the same
byte insBase = vc.getAlternateAllele(0).getBases()[0];

18
protected/gatk-tools-protected/src/test/java/org/broadinstitute/gatk/tools/walkers/annotator/VariantAnnotatorIntegrationTest.java
View File

@ -505,4 +505,22 @@ public class VariantAnnotatorIntegrationTest extends WalkerTest {
Assert.assertFalse(lineIterator.hasNext());
Assert.assertFalse(lineIteratorAnn.hasNext());
}
@Test
public void testHomopolymerRunWindow() {
WalkerTestSpec spec = new WalkerTestSpec(
"-T VariantAnnotator -R " + hg19ReferenceWithChrPrefixInChromosomeNames + " -A HomopolymerRun --variant:vcf " + privateTestDir + "problem_del.vcf " +
"-U ALLOW_SEQ_DICT_INCOMPATIBILITY -L chr18:44382010-44384010 --reference_window_stop 59 --no_cmdline_in_header -o %s", 1,
Arrays.asList("bda55495578147b2390d850d7fb25a12"));
executeTest("Testing testHomopolymerRunWindow", spec);
}
@Test
public void testHomopolymerRunTooBig() {
WalkerTestSpec spec = new WalkerTestSpec(
"-T VariantAnnotator -R " + hg19ReferenceWithChrPrefixInChromosomeNames + " -A HomopolymerRun --variant:vcf " + privateTestDir + "problem_del.vcf " +
"-U ALLOW_SEQ_DICT_INCOMPATIBILITY -L chr18:44382010-44384010 --no_cmdline_in_header -o %s", 1,
Arrays.asList("e20b12fd45f37a7bb31a2f2e91983477"));
executeTest("Testing HomopolymerRunTooBig", spec);
}
}