search

NCATSTranslator / Feedback

A repo for tracking gaps in Translator data and finding ways to fill them.
7 stars 0 forks source link

Is there anything that can be changed to have the full disease name displayed in the paths? #356

Open sstemann opened 1 year ago

sstemann commented 1 year ago

i thought these two paths were exact dupes:

image

But if you hover one is 'maturity-onset diabetes of the young' image

and the other is 'maturity-onset diabetes of the young 1' image

maybe users wont care but i struggle bc before i realized the end nodes were different, i was toggle back and forth on the evidence to see how they were different.

dnsmith124 commented 1 year ago

This is a very interesting issue and I totally understand the frustration! I think it's one that users will run into as well.

That said, given the wide range of lengths for node names I'm not sure how we could programmatically display the entire name for any entity while using the current design, since we'd need to accommodate a range of character counts from something as short as SNTB1 in the case of a chemical/gene query, all the way up to something as long as (1s,7s)-1-[3-[3,4-dihydroxy-4-[[(e)-8-hydroxy-2,6-dimethyloct-2-enoyl]oxymethyl]oxolan-2-yl]oxy-4,5-dihydroxy-6-(hydroxymethyl)oxan-2-yl]oxy-7-hydroxy-7-methyl-4a,5,6,7a-tetrahydro-1h-cyclopenta[c]pyran-4-carboxylic Acid in the case of a gene/chemical query.

We could reduce the font size based on the number of characters up to a certain limit which would likely solve the problem in some cases, but it's not an ideal solution. We could also try wrapping long names onto a single line, but that would cause issues with the design and not look great overall in my opinion. Hopefully it's not an incredibly widespread problem, but I can't say either way.

@desiyopp do you have any thoughts on how we could resolve this?

cbizon commented 1 year ago

If it were true that most of the differences between similar things were at the beginning and the end of the names, you could ellipse the middle rather than the end? So it would look like Maturity-Onset ... Young Maturity-Onset ... Type 1 or something like that?

dnsmith124 commented 1 year ago

@cbizon excellent idea, I'll look into how we could possibly implement that!

sierra-moxon commented 1 year ago

@dnsmith124 - do you think this might be on the roadmap for September? :)

dnsmith124 commented 1 year ago

@sierra-moxon I will tentatively say yes

dnsmith124 commented 1 year ago

After looking into what it would take to implement this, I'm pushing it to post-september. The amount of work is not worth the benefit right now, considering the amount of time left until the relay.

Will circle back after.

sstemann commented 8 months ago

@dnsmith124 would this be considered for Lobster1 - front end?

dnsmith124 commented 8 months ago

@sstemann we are not planning to implement this for the current cycle, I think it warrants further discussion and research given the wide-reaching implications.

sstemann commented 4 months ago

is this being considered for Eel? @dnsmith124

dnsmith124 commented 4 months ago

@sstemann we currently don't have this listed among our priorities for Eel, but I'm going to create a task in Clickup and share it with Reese to get her thoughts on how we could best resolve it. A resolution here is overdue, I think.