Missing data for mixed.gen.Burridge_Schneider_2017

[svengato]

For cowpea QTL data in v2/Vigna/unguiculata/genetic/mixed.gen.Burridge_Schneider_2017/

1. Markers listed in vigun.mixed.gen.Burridge_Schneider_2017.qtlmrk.tsv.gz are not found in Vigna/unguiculata/markers/IT97K-499-35.gnm1.mrk.Cowpea1MSelectedSNPs/vigun.IT97K-499-35.gnm1.mrk.Cowpea1MSelectedSNPs.gff3.gz

2. In vigun.mixed.gen.Burridge_Schneider_2017.obo.tsv.gz, some traits have no ontology code.

[sammyjava]

I put in the parent terms for stuff that's missing in the TO like "Root Tip Abundance" (RTA). Unless there's an ontology term for the Rochester Transit Authority.

[svengato]

Point 1 (missing Burridge-Schneider markers) is still unresolved.

[sammyjava]

@cann0010 is there a source for the missing Cowpea1MSelectedSNPs marker maps to vigun.IT97K-499-35.gnm1? Otherwise we'll just close this, we can't create data that don't exist. Here's what's in the qtlmrk file:

13772_1075
5084_519
4836_807
10811_937
2326_226
12501_343
139_439
5061_428
7102_965
1004_587
8969_1386
2227_693
9645_589
5428_339
3211_511
4749_1972
11138_624
14604_737
13848_735
11851_914
4245_136
2391_614

[StevenCannon-USDA]

Some sleuthing: It's not the IT97K-499-35.gnm1.div.Huynh_Ehlers_2018 diversity set, which has only 32k markers.
Skimming the paper ...
Burridge JD, Schneider HM, Huynh BL, Roberts PA, Bucksch A, Lynch JP. Genome-wide association mapping and agronomic impact of cowpea root architecture. Theor Appl Genet. 2017 Feb;130(2):419-431. doi: 10.1007/s00122-016-2823-y. Epub 2016 Nov 18. PMID: 27864597.
... the markers come from
Lucas MR, Diop N-N, Wanamaker S, Ehlers JD, Roberts PA, Close TJ
(2011) Cowpea–soybean synteny clarified through an improved
genetic map. Plant Genome 4:218–225
https://acsess.onlinelibrary.wiley.com/doi/full/10.3835/plantgenome2011.06.0019

... which traces back to Muchero et al, 2009
10.1073/pnas.0905886106

YES: Supplement for that Muchero et al paper (tsp file attached; derived from manuscript supplement 0905886106_SD1.xls)
0905886106_SD1.txt

[adf-ncgr]

thanks @cann0010! just glancing at the file it looks like at least some of the qtlmrk markers correspond to SNPs in the Cowpea1MSelectedSNPs, using the column 1 (SNP) as the match, e.g. 13772_1075 from the list above -> 1_0749 which is in the 1M set. Think it's legit to substitute the SNP ids for the "marker" ids in the qtlmrk file? Seems better to me to do so, but let me know if you agree with this change and I can make it so.

[sammyjava]

What matters is that the identifier in the qtlmarker file is the Name attribute of the marker in the GFF. That's how they're merged. Otherwise it'll be stranded markers with no genomic location.

[sammyjava]

(There are likely marker GFFs that will show up as issues because they don't have full-yuck ID attributes and short-name Name attributes, although I'll typically fix those myself.)

[adf-ncgr]

only 17 out of the 22 SNPs can be found in the 1M set; so we may have to get genomic locations for the others by mapping the sequences. Since they are not all in the 1M set, perhaps we should just treat them as their own little marker collection?

[sammyjava]

Fine with me, be sure to update README.genotyping_platform with your new name if/when you do.

[adf-ncgr]

OK, probably won't get to it for a while, but will re-assign to myself (though others are welcome to wrest it back)

[sammyjava]

ping @adf-ncgr

[sammyjava]

But also, @adf-ncgr you're really good at hijacking issues, which is not an instant messaging app. So please if you have a new issue like "what to do with marker sequences" create a new one rather than hijacking this one. Much appreciated. I'll do so with these posts. legumeinfo/datastore-specifications#24

[adf-ncgr]

thanks, you're right.

[sammyjava]

I think this got sorted out over time. Or we'll create a fresh issue but I've got VignaMine loaded and running.