I have two vcf files - one hg19 and one hg38, analysing data from the same participants on two slightly different SNP platforms. Both files have been through the pre-imputation checks.
The header (and the first line) of the hg38 version looks like:
##fileformat=VCFv4.3
##FILTER=<ID=PASS,Description="All filters passed">
##fileDate=20230906
##source=PLINKv2.00
##contig=<ID=chr1,length=248917420>
##contig=<ID=chr2,length=242106609>
.. other contigs
##INFO=<ID=PR,Number=0,Type=Flag,Description="Provisional reference allele, may not be based on real reference genome">
##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
#CHROM <participant_IDs so removed>
chr22 15349056 22:15349056:C:T C T . . PR GT 0/0 0/0 0/0 0/0 0/0 ...
...
The hg19 version looks like:
##fileformat=VCFv4.3
##FILTER=<ID=PASS,Description="All filters passed">
##fileDate=20230906
##source=PLINKv2.00
##contig=<ID=1,length=249211619>
##contig=<ID=2,length=243048760>
.. other contigs
##INFO=<ID=PR,Number=0,Type=Flag,Description="Provisional reference allele, may not be based on real reference genome">
##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
#CHROM <participant_IDs so removed>
22 16590657 22:16590657:T:C T C . . PR GT 0/1 0/0 0/0 0/1 0/0 ....
...
Both files have the chromosome identifiers recommended by MIS (hg19: <X>, hg38: chr<X>), but both files are failing the initial validation with No valid chromosomes found!
What could be causing this issue? I could re-do the pre-imputation checks, but these have already been through them before they got to me.
I have ensured that the correct assembly is selected when trying to impute, and I have also tried both files with the "wrong" identifiers (chr<X> for hg19, <X> for hg38), but the jobs still fail this initial check.
Looks like you have correct format. You can try getting rid of alleles attached to SNP column (for e.g. change
22:16590657:T:C
into22:16590657
) if that is the problem OR try running pre-imputation check again.I've successfully run another dataset through MIS with identifiers that look similar <X>:<POS>:REF:ALT, so I don't think it's that