I am running HC3.3-0 with the following options (e.g. GENOTYPE_GIVEN_ALLELES):
$java7 -Djava.io.tmpdir=tmp -Xmx3900m \
-jar $jar \
--analysis_type HaplotypeCaller \
--reference_sequence $ref \
--input_file $BAM \
--intervals $CHROM \
--dbsnp $dbSNP \
--out $out \
-stand_call_conf 0 \
-stand_emit_conf 0 \
-A Coverage -A FisherStrand -A HaplotypeScore -A MappingQualityRankSumTest -A QualByDepth -A RMSMappingQuality -A ReadPosRankSumTest \
-L $allelesVCF \
-L 20:60000-70000 \
--interval_set_rule INTERSECTION \
--genotyping_mode GENOTYPE_GIVEN_ALLELES \
--alleles $allelesVCF \
--emitRefConfidence NONE \
--output_mode EMIT_ALL_SITES \
The file $allelesVCF contains these neighbouring SNPs:
20 60807 . C T 118.96 .
20 60808 . G A 46.95 .
20 61270 . A C 2870.18 .
20 61271 . T A 233.60 .
I am unable to call these neighbouring SNPs; despite reads being present in the file $BAM, which shouldn't matter anyway. I also tried adding --interval_merging OVERLAPPING_ONLY to the command line, but that didn't solve the problem. What am I doing wrong? I should probably add GATK breaker/misuser to my CV...
Thank you as always.
P.S. The CommandLineGATK documentation does not say, what the default value for --interval_merging is.
P.P.S. Iterative testing a bit slow, because HC always has to do this step:
HCMappingQualityFilter - Filtering out reads with MAPQ < 20