GATK: Remove known variants from the sample vcf file

Question

[ravast]

I have a test sample vcf file, from which I have to select the unique variants. As an initial step I want to remove the know variants from my sample. I used SelectVariants walker from GATK.. I got this error

Input files /home///Mouse_ref/mgp.v3.snps.rsIDdbSNPv137.vcf and reference have incompatible contigs: Relative ordering of overlapping contigs differs, which is unsafe.
##### ERROR   /home/Mouse_ref/mgp.v3.snps.rsIDdbSNPv137.vcf contigs = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, X]
##### ERROR   reference contigs = [1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 2, 3, 4, 5, 6, 7, 8, 9, MT, X, Y, JH584295.1, JH584292.1, GL456368.1, GL456396.1, GL456359.1, GL456382.1, GL456392.1, GL456394.1, GL456390.1, GL456387.1, GL456381.1, GL456370.1, GL456372.1, GL456389.1, GL456378.1, GL456360.1, GL456385.1, GL456383.1, GL456213.1, GL456239.1, GL456367.1, GL456366.1, GL456393.1, GL456216.1, GL456379.1, JH584304.1, GL456212.1, JH584302.1, JH584303.1, GL456210.1, GL456219.1, JH584300.1, JH584298.1, JH584294.1, GL456354.1, JH584296.1, JH584297.1, GL456221.1, JH584293.1, GL456350.1, GL456211.1, JH584301.1, GL456233.1, JH584299.1]

[norhanmahfouz]

I didn't quite get if the problem was solved ... I get the same error as Ravast and it's not about the order in case of mouse genome - it's actually that those chromosomes/contigs are absent all together from the vcf file header of already known SNPs compiled by the Sanger institute!!!! I tried to circumvent it by modifying the header (adding the missing contigs in the header) but I get the same error .... :(

Answer 1

[Ashutosh Pandey]

This is pretty common error. If you would have searched a little on this forum or online you could have got the answers. GATK requires order of chromosomes to be the same in both the files. Elaborately described here. I think you just used cat chr*.fa command to concatenate the individual fasta files (chromosomes) to make the reference file and that messed up the order. You are not wrong but this is how GATK works.

[ravast]

Thank you ashutosh..

[Biomonika (Noolean)]

This should be comment rather than an answer :-) Also, if the answer solved your problem, you should "accept" it so that your question will be stored as solved.

[ravast]

Ashutosh..could you please let me know how I should rectify the error.

[Ashutosh Pandey]

Karyotypically Ordered Hg19. You need to sort your reference fasta file or create a new reference fasta from scratch. Make sure it has all the chromosomes present in your vcf file in the same order ie. 1,2,3,4...X.