GWAS result file -- does it allow a range

[maxglycine]

Sometimes GWAS results are given as a range. In these cases, the authors say that marker1, marker2, marker3 are all part of a co-inherited block (linkage block) and as such are one genetic thing (i.e. gene or genes). When that happens in classic we depicted it in CMap GmGWAS as a bar between markers 1&3 as opposed to a series of individual markers (SNPs). I was looking at the Genus/species/gwas/result.md file for a couple of GWAS studies and it does not seem to have a way of depicting a range. Is that concept engineered into the spec? It is not clear if the same trait_name can be associated with more than one SNP from looking at a selection of records. I am assuming that concept was engineered into the GlycineMine data model by Sam.

We also need to add an explanation of how to make a GWAS entry in the protocols page (https://github.com/legumeinfo/datastore-specifications/tree/main/PROTOCOLS).
@StevenCannon-USDA @jd-campbell

[StevenCannon-USDA]

A range would be represented through a marker. The GWAS "result" file just reports an association (as a p-value) between a trait and a marker. Plenty of markers have lengths -- SSRs, for example.

I don't believe that any of our current GWAS studies have range markers. You would see the range in the "Markers Length" value, for example here:
https://mines.legumeinfo.org/glycinemine/genotypingplatform:Vuong_Sonah_2015

That none of our GWAS studies use range markers is probably just a reflection of what's published. They all seem to use SNPs.

We could make something like meta-GWAS collections that link markers (using marker sets with "compound markers"). I don't know what we would do with the p-value statistics.

In any case, I don't see a technical reason that it couldn't be done -- at least in terms of the technical representation.