clarify weight of vine traits #76
Conversation
|
No, in this case I just did the edits in Atom as I wasn't adding new traits. But it does load and save fine in Protege after the edits. |
There was a problem hiding this comment.
Hi Bryan, It's not clear from your comments here what exactly you did. It looks like you modified the name of two sweetpotato vine weight traits (CO_331:0000248 and CO_331:0000252), but then removed (obsoleted?) them and modified the variable terms, with new terms for the methods and scales?
It also looks like thee is some confusion on the "Namespace" field as some of the variables are labeled "namespace: SweetpotatoTrait" .
Also, the term mass and weight are used interchangeably which is also a bit confusing. If its measured using a scale, it is more appropriate to use weight.
|
Thank you Laurel. I think I managed to confuse everyone including myself with this edit. It's a lot tricker since I don't have firsthand knowledge of how these are used. Here's a hopefully more clear explanation of the changes I made, courtesy of Jolien. But now I'm thinking this edit is not a good approach and it'd be better to just improve the definitons of the existing terms. OLD CO_331:1000007 quality traits
CO_331:1000008 agronomic traits
CO_331:1000008 agronomic traits
With regard to the namespace, that's a consequence of the way Sweetpotatobase uses the ontology - it loads only the traits and variables, and to do so they have to be in a single namespace. Thats currently the pattern in the whole file. But perhaps the master version of the ontology should maintain them in separate namespaces. And true, I didn't change the conflicting references to vine mass and vine weight. Hopefully Jolien can correct me if I'm wrong but I assume all these measurements are made using scales, so we could change all terms to refer to weight. |
|
Hi Bryan and others, nice to see my interpretation was correct. My advice would be to keep the field trait (kg/plot) as part of the agronomic traits and the lab traits (g) as part of the quality traits. But you can still keep the structure more or less: CO_331:1000008 agronomic traits
This structure will also be helpful when applying to the root weight. |
|
Thanks Jolien, good to know. I'll update the branch accordingly The main reason the database does not load methods and scales is that they're not strictly necessary for storing measurements (the variable should contain all the required info), and there have been problems with duplication in those namespaces. For example multiple terms named '9pt scale' or 'kg/plot'. But the duplication issues have already been fixed for the most part in the sweetpotato ontology, so yes we could try to fix the last few cases and load them. |
bellerbrock
force-pushed
the
topic/weight_of_vines
branch
from
23d843e to
a778e15
Compare
|
@bellerbrock aside from the problem with the name duplications, we also have an issue with methods and scales having a different namespace. we currently cannot display on the ontology browser cross-namespace relationships. |
|
Good points Naama. I hadn't thought about the limitations on cross-namespace relationships. I agree that the ideal solution would be separate ontologies. It won't be part of this branch but I'd like to come up with a test of that solution at some point to see how well it works |
|
Ok here's the updated set of changes for the vine terms. Same adjustments have also been applied to the root weight terms. Somebody please review and merge! OLD CO_331:1000007 quality traits
CO_331:1000008 agronomic traits
CO_331:1000007 quality traits
CO_331:1000008 agronomic traits
|
Description
closes #64
Checklist