error using bqsr and incompatible contigs

Question

[bioguy24]

Using the below command in GATK 4.0.6.0, I am getting:

A USER ERROR has occurred: Input files reference and features have incompatible contigs: No overlapping contigs found.

Code:

gatk BaseRecalibrator \ 
    -I NA12878_mark_duplicates.bam \ 
    -R /home/cmccabe/Desktop/fastq/resources/ucsc.hg19.fasta \
     --known-sites /home/cmccabe/Desktop/fastq/resources/dbsnp_138.b37.vcf \ 
     --known-sites /home/cmccabe/Desktop/fastq/resources/Mills_and_1000G_gold_standard.indels.hg19.vcf \
     -O /home/cmccabe/Desktop/fastq/NA12878_recal_data.table

I am using the resource bundle from the ftp as well as the fasta. Is there a way to ignore these errors or am I doing something wrong? I also aligned with the same fasta using bwa-mem. Thank you :).

[Ram]

Can you check the dbsnp file and ensure the contig names match those in the ucsc.hg19.fasta? I think the problem might lie there.

[h.mon]

The warnings is hinting you have different chromosome names somewhere. Examine the fasta, bam and vcf and check if they all use chr1, chr2, chr3, and so on; or they all use 1, 2, 3, and so on.

[bioguy24]

Thank you both very much :).

Answer 1

[h.mon]

Upon closer inspection, I see you are mixing hg19 and b37 references. Although they are largely the same, they are not identical, and there are more differences than just chromosome naming convention:

• differences in mtDNA sequence
• different REF bases at some SNPs

You should use the exactly same reference for all resources (reference genome for mapping and reference vcfs).

Further discussions:

What's the difference between b37 and hg19 resources?

problems in hg19 and b37 compatibility