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

Permit empty fields in INFO and FORMAT structures - not fully tested yet but at least failing cases before now pass. Also, corrected a bug where in case we were reading 3.3 VCF's, or VCFs with no original allele encodings, we'd always print 2 bases per allele. 

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@3698 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
delangel 2010-07-01 01:56:07 +00:00
parent 15a6be5d50
commit dc4715c9c6
1 changed files with 47 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

67
java/src/org/broadinstitute/sting/utils/genotype/vcf/VCFWriter.java
View File

@ -290,10 +290,22 @@ public class VCFWriter {
} }
} else { } else {
// no original Allele information: add one common base to all alleles (reference at this location) // no original Allele information: see first if all alleles have a base encoding (ie no deletions)
trailingBases = new String(refBases); // if so, add one common base to all alleles (reference at this location)
numTrailingBases = 1; boolean hasBasesInAllAlleles = true;
position--; for ( Allele a : vc.getAlleles() ) {
String alleleString = new String(a.getBases());
if (alleleString.length()==0) {
hasBasesInAllAlleles = false;
break;
}
}
if (!hasBasesInAllAlleles) {
trailingBases = new String(refBases);
numTrailingBases = 1;
position--;
}
} }
for ( Allele a : vc.getAlleles() ) { for ( Allele a : vc.getAlleles() ) {
@ -370,24 +382,33 @@ public class VCFWriter {
} */ } */
VCFFormatHeaderLine.FORMAT_TYPE formatType = typeUsedForFormatString.get(key);
Object newVal; Object newVal;
if (!val.getClass().equals(String.class)) if (typeUsedForFormatString.containsKey(key)) {
newVal = formatType.convert(String.valueOf(val)); VCFFormatHeaderLine.FORMAT_TYPE formatType = typeUsedForFormatString.get(key);
else if (!val.getClass().equals(String.class))
newVal = val; newVal = formatType.convert(String.valueOf(val));
else
newVal = val;
if (numberUsedForFormatFields.get(key)>1 && val.equals(MISSING_GENOTYPE_FIELD)) {
// If we have a missing field but multiple values are expected, we need to construct new string with all fields.
// for example for Number =2, string has to be ".,."
StringBuilder v = new StringBuilder(MISSING_GENOTYPE_FIELD);
for ( int i = 1; i < numberUsedForFormatFields.get(key); i++ ) {
v.append(",");
v.append(MISSING_GENOTYPE_FIELD);
}
newVal = v.toString();
} }
else {
newVal = val;
}
if (numberUsedForFormatFields.containsKey(key)){
int numInFormatField = numberUsedForFormatFields.get(key);
if (numInFormatField>1 && val.equals(MISSING_GENOTYPE_FIELD)) {
// If we have a missing field but multiple values are expected, we need to construct new string with all fields.
// for example for Number =2, string has to be ".,."
StringBuilder v = new StringBuilder(MISSING_GENOTYPE_FIELD);
for ( int i = 1; i < numInFormatField; i++ ) {
v.append(",");
v.append(MISSING_GENOTYPE_FIELD);
}
newVal = v.toString();
}
}
// assume that if key is absent, given string encoding suffices.
String outputValue = formatVCFField(key, newVal); String outputValue = formatVCFField(key, newVal);
@ -531,11 +552,17 @@ public class VCFWriter {
isFirst = false; isFirst = false;
else else
info.append(INFO_FIELD_SEPARATOR); info.append(INFO_FIELD_SEPARATOR);
info.append(entry.getKey()); info.append(entry.getKey());
if ( entry.getValue() != null && !entry.getValue().equals("") ) { if ( entry.getValue() != null && !entry.getValue().equals("") ) {
int numVals = 1; int numVals = 1;
if (this.writingVCF40Format) { if (this.writingVCF40Format) {
numVals = numberUsedForInfoFields.get(entry.getKey()); String key = entry.getKey();
if (numberUsedForInfoFields.containsKey(key)) {
numVals = numberUsedForInfoFields.get(key);
}
// take care of unbounded encoding // take care of unbounded encoding
if (numVals == VCFInfoHeaderLine.UNBOUNDED) if (numVals == VCFInfoHeaderLine.UNBOUNDED)
numVals = 1; numVals = 1;