Bugs & Feature Requests
Known issues and requested enhancements


This page lists the issues tagged "bug" in the GATK4 development repository. If you have reported a bug and cannot find it in this list, let us know in the thread where you reported it. Note that we are no longer listing or processing issues in older versions of GATK (3.x and prior).


Showing open



Possible race condition in GenomicsDBImport multi-interval mode.
CRAMs generated by GATK don't always have M5/UR attributes in the sequence dictionary
HaplotypeCaller should use realigned reads for ref conf calc in flanking regions
GATK4 VariantAnnotator does not handle PED file that GATK3 can
GenotypeGVCFs hangs in 4.0.5.1
Debug state is not properly passed to the EventMap in HaplotypeCaller
Mutect2 and haplotypecaller require UCSC chromosome names
Inconsistent unmapped but placed query results
Create a MAFCodec for eliminating possible conflicts
SV breakpoints location off when surrounding homology are of different length, and SVLEN off when duplicated copies are of different length
Malformed VCF headers in HaplotypeCallerSpark
The combination of ReadsDownsamplingIterator + ReservoirDownsampler does not work correctly
GenomicsDB errors discovered in the wild
Picard version doesn't print correctly when running a Picard tool from GATK
barclay fails to set collection arguments initialized with an immutable collection
BwaSpark and bam order
Users reporting unreasonable memory usage in GenotypeGVCFs
`__pthread_tpp_change_priority` error in GenotypeGVCFs (GATK 4.0.0.0)
GATK4 GenomicsDB error with GenomicsDB versions 0.9.0 and 0.9.2
GenotypeGVCFs takes a long time on small interval
Error running HaplotypeCallerSpark in GVCF mode
# in file names converted to %23 resulting in file not found
Spark tools have problems with umlauts
Funcotator - Problems with variant alleles that span exon, coding sequence start/end boundaries
MannWhitneyU shouldn't do a get(double)
Problem with spark.yarn.executor.memoryOverhead on DataProc
Crash with IndexFeatureFile on VCF with header but no variants
ReadsPipelineSpark: null point exception during MannWhitney permutation test
HaplotypeCallerSpark empty collection error on regions with no reads