If the disease is genetic (error in a gene) or genomic (error in the large scale structure of the genome), a very important thing to do would be to also sequence the parents (preferably WGS [Whole Genome Sequencing]). Doing such a trio sequencing experiment allows you to determine, for every gene and even base pair, from what parent it came and this allows you to assess what is normal (assuming the parents are not affected). This way you may find out if your partner is homozygous for an aberration both parents are heterozygous for. This could mean that is THE affected gene/chromosome. You may also find what we call "de novo" aberrations, these are aberrations that are not present in the parents at all and arose in you partner. These aberrations are very suspect (again, because both parents are not affected).
You may find something, you may not, a lot is unknown and regions outside of the genes may be affected and even the cause of the phenotype, but we still understand very little of this.
Depending on where you live, genomic counseling is free and trio sequencing is usually part of it.
This is not really my expertise (more in oncology) but feel free to ask more questions.
If you have BAM (or CRAM + reference genome) files for parents and your partner, you could download a trial of VarSeq [0] to do a more GUI based analysis of the results.
"Is it sufficient to identify variants by searching for the SNP string (e.g. "rsXXXXXX") in the VCF file?"
If the variants have been associated with the same phenotype as your partner's, then yes, it is interesting. If there is no phenotype, perhaps you can track down the source publication and try to talk to the authors.
There are probably groups online with people in the same situation, try to find them, they can probably help you a lot more.
Maybe to add, from a personal perspective, try not to loose yourself in this. I felt the same when my mother was diagnosed with cancer... I was the expert, I needed to save her. But my mother needed to have faith in the system and no second guessing. She trusted her oncologist. I backed off, they got her to the point of being cancer free for 3 years now against all odds.
Admittedly, your situation is different. Still, your partner may need you more as a supporting, fun, optimistic person rather than the miserable piece of human you can become from a bottomless rabbit hole like genomics, where the answer to your partner's problem may forever seem like "almost within your grasp".
A final thing to add, I didn't see anyone here mention trio WGS, which, I think is the way to go for unknown hereditary conditions. Perhaps an expert can confirm, but it will narrow your search by a huge factor.
You may find something, you may not, a lot is unknown and regions outside of the genes may be affected and even the cause of the phenotype, but we still understand very little of this.
Depending on where you live, genomic counseling is free and trio sequencing is usually part of it.
This is not really my expertise (more in oncology) but feel free to ask more questions.
If you have BAM (or CRAM + reference genome) files for parents and your partner, you could download a trial of VarSeq [0] to do a more GUI based analysis of the results.
"Is it sufficient to identify variants by searching for the SNP string (e.g. "rsXXXXXX") in the VCF file?" If the variants have been associated with the same phenotype as your partner's, then yes, it is interesting. If there is no phenotype, perhaps you can track down the source publication and try to talk to the authors.
There are probably groups online with people in the same situation, try to find them, they can probably help you a lot more.
[0]: https://www.goldenhelix.com/products/VarSeq/index.html