search

pombase / website

PomBase website v2
MIT License
6 stars 1 forks source link

Display of promoter info on gene pages, and genotype pages #2161

Closed ValWood closed 6 months ago

ValWood commented 7 months ago

Collect all genotypes which differ only by promoter onto the same genotype page (this is analogous to background)

Rationale: Most users will want to see all of the associated phenotypes.

How it looks on Gene Page Term page Publication page

All alleles of a gene Allele page

Also need to think about multi locus

kimrutherford commented 7 months ago

Part of this is implemented already, by accident. Genotypes with and without promoters are already combined on the same page. For example mad1+[Overexpression] has two versions:

But there is only one genotype page:

The promoter is getting lost when the the genotypes are merged but the annotation is kept.

One of the genotype comments is getting lost too. Maybe the comments need to be included in the annotation table like we do for the background and we plan to do with the promoters?

ValWood commented 7 months ago

Some of the genotype comments are really long, and I think many of them need to be migrated to allele comments. Can I a list of genotype comments I can view? I think most will be relevant to any genotype, or even a few examples (I can't think of any genotypes with comments although there are loads)

kimrutherford commented 7 months ago

Here are all the genotype comments, with links to the genotype pages: genotype_comments.txt

Here are some examples:

kimrutherford commented 7 months ago

Most of the code for this is finished now but the display is a bit messy. Let's chat about it on the next call. Possibly this case is worse than most because there are promoters and a genotype background.

This is the publication page:

image

The genotype page is tidier:

image

kimrutherford commented 7 months ago

The genotype page is tidier:

I forgot about the plan to move the promoter into its own column in the annotation table. That's done now:

image

image

kimrutherford commented 7 months ago

I've just changed this:

image

to display as this instead to make it clear about exogenous promoters:

image

Do you have any suggestions to improve this?

ValWood commented 7 months ago

I think the label is a bit confusing because this is a manipulated s. pombe promoter. As far as I remember the only reason we have the ability for people to select the pombe name when its an exact swap is to control the contents (we can't control the contents of the free text one). So I don't think we need to display "exogenous promoter" on the website. I'm not even sure that we even need to link to the gene when its a swap?

CC @PCarme

PCarme commented 7 months ago

Linking to the gene to which the promoter is endogenously associated doesn't seem very useful to me. Aside from the expression data, most of the information about this gene are irrelevant to the case of a promoter swap anyway...

kimrutherford commented 7 months ago

As far as I remember the only reason we have the ability for people to select the pombe name when its an exact swap is to control the contents

I don't follow that. What do you mean by control the contents? We could just have a "Promoter swap (S. pombe)" checkbox if that's all that's needed. We could remove the ability for people to select the pombe gene if that detail isn't necessary.

kimrutherford commented 7 months ago

So I don't think we need to display "exogenous promoter" on the website.

So remove the text "exogenous promoter"?

ValWood commented 7 months ago

I don't follow that. What do you mean by control the contents? We could just have a "Promoter swap (S. pombe)" checkbox if that's all that's needed. We could remove the ability for people to select the pombe gene if that detail isn't necessary.

I can't remember the original rationale. We want the gene, but there isn't a big reason to separate exogenous and endogenous (IMHO).

We could just always have a single field and make them type the promoter. I think it is better to discuss this on a call.

ValWood commented 7 months ago

Summary

  1. Compress the promoter capture to be free text, just "promoter(optional)"

  2. Check whether the promoters we already captured in this way are not loaded as genes associated with a session

kimrutherford commented 6 months ago

Check whether the promoters we already captured in this way are not loaded as genes associated with a session

I'll remove that code when I change the loading code:

kimrutherford commented 6 months ago

I changed the web site code to handle having just one sort of promoter to display. I'll push the changes at the weekend once I upgrade the Canto database to the new schema (pombase/canto#2823).

Branch: pombase-chado-issue-567-go-ev-code-graph

kimrutherford commented 6 months ago

I'll push the changes at the weekend once I upgrade the Canto database to the new schema

That's done now. I'll check the website on Sunday.

kimrutherford commented 6 months ago

I'll push the changes at the weekend once I upgrade the Canto database to the new schema

All seems OK except that I found a case where the promoters aren't appearing in the annotation table on the gene page.

For mad2+[Knockdown] slp1+[Knockdown] the promoters are on the genotype page

https://pombase.org/genotype/mad2%2B%28wild_type%29-expression-knockdown__slp1%2B%28wild_type%29-expression-knockdown

image

And on the term page: https://pombase.org/term/FYPO:0003762

image

But they are missing from the gene page: https://pombase.org/gene/SPAC821.08c

kimrutherford commented 6 months ago

All seems OK except that I found a case where the promoters aren't appearing in the annotation table on the gene page.

Fixed. I'll close this because it seems mostly done and there are a lot of comments.