colour pathological/non pathological mutants differntly #2249
Comments
ValWood
changed the title
That should be no problem and it will make the display look more interesting. What's the rationale for only colouring the single amino acid mutants? |
|
We could do both. I was biased toward point mutations because I was thinking about survivorship bias and this image: 
|
Should this term be red or grey? It's viable but abnormal? |
|
This has some aspect of pathology so this would be red. |
So when should an allele be grey? Only when it's viable and also normal? Are these the terms we need?: |
|
If it is inviable (Cell or population) OR abnormal it should be red. All others should be grey. So they would be grey if they have ONLY viable (cell or population) OR normal |
So inviable cell (FYPO:0000049) or inviable cell population (FYPO:0002059) or abnormal phenotype (FYPO:0001985)? |
|
red if all others grey |
|
yes sorry posts crossed, i was looking them up and got side tracked |
|
First attempt:
|
|
We can do the same for partial deletions if you think that's useful:
|
|
An example with both:
|
|
This will be really useful! Could you zoom in on this a bit? 
|
Yes, I think we should |
It's here if you want for have a look: |
|
Yep, I just looked: So this one is the same allele separating normal and abnormal phenotypes. I envisaged that we would represent each allele once but colour it red even if it had some normal phenotypes, and long as it had some abnormal phenotypes. The rationale is that every allele will have lots of normal things, depending on what is tested. This is just a way to focus people attention on the alleles that have at least some identified detrimental effects. @PCarme what do you think? |
|
They look different to me. Same positions but the substitutions are different (S572A vs S572C, etc.) |
|
doh, ignore me! |
|
Are there any changes to make before I release this? |
|
No go for it! |
|
Those changes are on pombase.org now. |
|
This is really useful. perhaps And we should maybe use the same colour scheme for both aa subs and partial deletion |
|
It was a challenge to find an example gene that has inviable, abnormal and normal alleles. Inviable is quite a lot less common than abnormal - I guess that makes sense. I tried pale pink and it was difficult to see the difference between the inviable and the abnormal alleles. So here's how it looks with a more contrasting colour: https://desktop.kmr.nz/gene_protein_features/SPAC23C4.18c
|
|
That's a good idea to distinguish between abnormal phenotypes and inviable, but I would have chosen an in-between color for abnormal. Orange maybe ? |
Here are a few other examples: https://desktop.kmr.nz/gene_protein_features/SPBC2F12.08c |
|
Here's how it looks with orange:
|
|
I think Orange is a good plan. Can you use a stronger orange (as strong as you can go without it merging with the red) Hex #F06631 Hex #FD963D Hex #FD5602 Hex #F5883F Hex #FF8000 Hex #E78400 Hex #FF6E00 Hex #FF6700 Hex #FF9B00 Hex #FE5000 Hex #FF6E4A Hex #FF7518 Hex #F76806 Hex #FF9916 Hex #FFA500 Hex #F96815 Hex #F05E23 |
|
maybe most of these are too close to the red? |
|
perhaps Hex #FF9B00 or Hex #FF9916 which red shade are we using ? |
I was wondering if we could 'up the contrast' by moving the red shade towards one which is very slightly pink...we can try that once we have a stronger orange which works well. |
|
The contrast varies between monitors so perhaps we can try a few options live on the next Zoom call. Whatever colours we decide on I think it's good to make the inviable alleles clear because they are the uncommon case. Point mutations are the trickiest case. For example: |
|
OK, we can do that, there's no hurry. |
It's |
Sorry, I got that wrong. We configure |
Also tweak the modification colours to be less green. Refs pombase/website#2249
|
I've committed those changes so it should be on pombase.org on Wednesday morning. |
|
I forgot to update documentation for the modification colours here: https://www.pombase.org/documentation/gene-page-protein-features-widget |
I think it would be really useful if we coloured the pathological single amino acid mutants differently from the non pathological.
I.e if a single amino acid mutant is either
abnormal or inviable phenotype colour them red
if viable or normal colour them grey
(some may have both abnormal and normal phenotypes, in these cases abnormal would trump.
Rationale.
I though of this when reviewing the InterPro website changes. They report/dispaly only pathogenic mutants (which is what most users will be interested. in).
We are unusual in that we also report non-pathological.
We report the non-pathological for information and completion, but it would be great if, when looking at this view you could see clearly which residues were the most important ones.
The text was updated successfully, but these errors were encountered: