v2.8.1: diploid input, haploid output

[dmjask]

Hi,
I'm running cactus-pangenome with diploid data as input: two files per individual (14 fasta files), named as described in the docs. For 6 of the 7 individuals the vcf output is diploid, but one individual is entirely haploid. This individual is homozygous, but is not the only homozygous in the data.

And idea or suggestion what may be causing this?

[glennhickey]

If you are giving only 14 fasta files, one must be the reference for the vcf. that leaves 13 sample columns (6 diploid + 1 haploid). Could this be what's happening, or are you giving 7 diploid + 1 reference genome?

[dmjask]

Yes, sorry about the lack of clarity, I'm providing 14 samples and an additional reference for a total of 15 files listed in the seqfile.

[glennhickey]

Something that would cause this is that missing haplotype not aligning anywhere. If it's not in the graph it will show up as haploid in the VCF. You might want to verify with vg paths that it's in the graph as expected.