Error in GATK HaplotypeCaller

Question

[Hyper_Odin]

I am getting the following error while running the following command:

gatk HaplotypeCaller --java-options '-Xms2g -Xmx31g -XX:ParallelGCThreads=4 -Djava.io.tmpdir=/u/rajeshpal93/scratch/ca/diva_test_final/temp/samtools' -R /ELS/els9/users/biosciences/references/ucsc/hg19/ucsc.hg19.fasta -I reads/recalibrated/DNA18-0067-R0001.dedup.recal.bam -O variant_calling/DNA18-0067-R0001.g.vcf.gz -ERC GVCF -L /ELS/els9/users/biosciences/references/ucsc/hg19/intervals/hg19_ncbi_refseq_sorted_merged.bed -ip 200 -G StandardAnnotation >& logs/gatk/HaplotypeCaller/DNA18-0067-R0001.genotype_info.log

the error

A USER ERROR has occurred: Badly formed genome unclippedLoc: Contig chr1_gl383516_fix given as location, but this contig isn't present in the Fasta sequence dictionary

I don't understand where is the problem. Is it the bam file or bed file?

Thanks in advance !!

[Amitm]

Hi, The error message is saying that there is a chromosome (in this case chr1_gl383516_fix) that is present in your input BAM (either in the header lines, or in the alignment, or both), that is not present in the genome fasta : ucsc.hg19.fasta you have provided. Check the header lines of your BAM using samtools view -H my.bam (or alternatively don't use the -H param and grep on the output for chr1_gl383516_fix to find the source of error)

[Amitm]

On second thoughts, the BED file that you are providing is also worth checking. As a rule, given a standard BAM file, the sequence names in the header lines of that should match with all other sequence data inputs (genome fasta, BED file etc.)

[Hyper_Odin]

Yess, found the culprit. It was the bed file indeed!! Thanks !!