when i do SNP calling using GATK (HaplotypeCaller), there is a smaller output file. And i check the log file, found:

03:00:49.697 INFO  VectorLoglessPairHMM - Time spent in setup for JNI call : 38.567675937000004
03:00:49.698 INFO  PairHMM - Total compute time in PairHMM computeLogLikelihoods() : 12384.776844409002
03:00:49.698 INFO  SmithWatermanAligner - Total compute time in java Smith-Waterman : 306401.51 sec
03:00:49.698 INFO  HaplotypeCaller - **Shutting down engine**

[2021年3月22日 上午03时00分49秒] org.broadinstitute.hellbender.tools.walkers.haplotypecaller.HaplotypeCaller done. Elapsed time: 6,176.21 minutes.
Runtime.totalMemory()=1092616192

**java.lang.NullPointerException**
    at java.util.stream.ReferencePipeline.forEachWithCancel(ReferencePipeline.java:126)
    at java.util.stream.AbstractPipeline.copyIntoWithCancel(AbstractPipeline.java:499)
    ...
    at org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.BaseGraph.getReferenceSinkVertex(BaseGraph.java:201)
    at org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.BaseGraph.cleanNonRefPaths(BaseGraph.java:430)
    at org.broadinstitute.hellbender.tools.walkers.haplotypecaller.readthreading.ReadThreadingAssembler.getAssemblyResult(ReadThreadingAssembler.java:715)

...

the code is below: (runing in a conda environment)

for i in XS*.dedup.bam
do
nohup gatk --java-options "-Xms1g -Xmx3g -Djava.io.tmpdir=/mnt/10T/tmp" HaplotypeCaller \
                          -R /mnt/10T/BWA/Dsag.fa \
                          -I $i \
                          -ERC GVCF \
                          -O /mnt/10T/Dipus_sagitta/Bwa_result/gatk_result/${i%%.*}_gatk.g.vcf \
                          -stand-call-conf 30.0 \
                          --pcr-indel-model CONSERVATIVE \
                          --sample-ploidy 2 \
                          --kmer-size 10 --kmer-size 25 \
                          1>/mnt/10T/Dipus_sagitta/Bwa_result/gatk_result/${i%%.*}_gatk.log 2>&1 &
done