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

BCF2 is now officially zero-based open-interval, and that's how the GATK does it now

This commit is contained in:
Mark DePristo 2012-07-01 14:59:27 -04:00
parent 9b87dcda4f
commit 480b32e759
2 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
public/java/src/org/broadinstitute/sting/utils/codecs/bcf2/BCF2Codec.java
View File

@ -239,10 +239,10 @@ public final class BCF2Codec implements FeatureCodec<VariantContext>, ReferenceD
final String contig = lookupContigName(contigOffset);
builder.chr(contig);
this.pos = decoder.decodeInt(BCF2Type.INT32);
this.pos = decoder.decodeInt(BCF2Type.INT32) + 1; // GATK is one based, BCF2 is zero-based
final int refLength = decoder.decodeInt(BCF2Type.INT32);
builder.start((long)pos);
builder.stop((long)(pos + refLength - 1)); // minus one because of our open intervals
builder.stop((long)(pos + refLength - 1)); // minus one because GATK has closed intervals but BCF2 is open
}
/**

8
public/java/src/org/broadinstitute/sting/utils/variantcontext/writer/BCF2Writer.java
View File

@ -84,7 +84,6 @@ import java.util.*;
*/
class BCF2Writer extends IndexingVariantContextWriter {
final protected static Logger logger = Logger.getLogger(BCF2Writer.class);
final private static List<Allele> MISSING_GENOTYPE = Arrays.asList(Allele.NO_CALL, Allele.NO_CALL);
final private static boolean ALLOW_MISSING_CONTIG_LINES = false;
private final OutputStream outputStream; // Note: do not flush until completely done writing, to avoid issues with eventual BGZF support
@ -203,10 +202,11 @@ class BCF2Writer extends IndexingVariantContextWriter {
// note use of encodeRawValue to not insert the typing byte
encoder.encodeRawValue(contigIndex, BCF2Type.INT32);
// pos
encoder.encodeRawValue(vc.getStart(), BCF2Type.INT32);
// pos. GATK is 1 based, BCF2 is 0 based
encoder.encodeRawValue(vc.getStart() - 1, BCF2Type.INT32);
// ref length
// ref length. GATK is closed, but BCF2 is open so the ref length is GATK end - GATK start + 1
// for example, a SNP is in GATK at 1:10-10, which has ref length 10 - 10 + 1 = 1
encoder.encodeRawValue(vc.getEnd() - vc.getStart() + 1, BCF2Type.INT32);
// qual