Missing views for Cancer SVs

[ivadym]

They may be some missing fields for the SV views:

• [x] Cancer observations (both somatic and germline) are not appearing in the SV view in Scout. However, they are visible for SNVs! It seems that the follwing if statement should also consider cases when variant.category == "cancer_sv":

  https://github.com/Clinical-Genomics/scout/blob/dc0ed342f22d4f61448262894b2ed2f6f771ab9b/scout/server/blueprints/variant/templates/variant/variant_details.html#L219-L232

  Fixed in https://github.com/Clinical-Genomics/scout/pull/4342

• [x] The Observed column only shows the Local tag when it contains RD local observations, not taking into account the ones from Cancer. A good example is this variant with only Cancer observations: https://scout-stage.scilifelab.se/cust059/G1A1471p10_Balsamic/cancer/becc87e78804ce7da198f49a14dcd70f?cancer=yes Fixed: https://github.com/Clinical-Genomics/scout/pull/4348

• [x] The Pop Freq column is empty for Cancer Clinical Somatic Structural Variants: https://scout-stage.scilifelab.se/cust059/G1A1471p10_Balsamic/sv/variants/1f8f1489ff2b16b65489ce1d5173043d Fixed: https://github.com/Clinical-Genomics/scout/pull/4353

I think it's also worth considering if this issue extends beyond these cases, as it seems that cancer_sv is not considered as a proper category to display some data. We should verify if we're expecting to display information also for SVs when the variant category is just cancer.

[northwestwitch]

Hi @ivadym are you working on this? Otherwise I can help!

[northwestwitch]

Hi again. Are you sure the cases described in checkboxes 2 and 3 will not be resolved by changing the settings relative to cust059?

You need to select the loqusdb instances you want stats from in the settings:

Don't forget to save the settings by clicking on "save settings" at the end of the page

[ivadym]

Yes, but that setting seems to only affect what's shown in the Local observations component in the variants page:

I'm talking about this one, which misses to report archive observations for Cancer:

I've checked a bit the code and I think we need to add something like this for the observations (update the observed_cell_general macro, which currently only shows local_obs_old):

Thanks for looking into this! I can work on it once I get access to cg-vm1, but if you have time please go ahead. I could also help with some testing :)

[northwestwitch]

Ok, but then we can mark the second checkbox as done, right? The third one might have to do with the variant parsing.

[ivadym]

Sorry, I'm not being clear enough.

In point 2, the variant I mentioned lacks the expected Local tag in the Observed column on the cancer/variants page:

However, on the specific variant page, you'll see we're reporting cancer variants:

I think we need to extend this part to include local_obs_cancer_somatic_old and local_obs_cancer_germline_old:

https://github.com/Clinical-Genomics/scout/blob/835dac88c81895a15249d02532ab927da95bd7b6/scout/server/blueprints/variants/templates/variants/components.html#L184-L186

[northwestwitch]

Ah I see, you are on the variantS page, not the variant one. Understood!

[northwestwitch]

Ok, point 2 should be fixed now. We merged the fix yesterday. The latest main branch is now deployed on stage