Capturing phase status for AR variants

[wrightmw]

Request for two changes in the GCI for genetic data for individuals with autosomal recessive disease inheritance:

(1) Update the text next to the checkbox in the "Individual - Variant(s) and Score(s) segregating with Proband" section from "Check here if there are 2 variants AND they are both located in trans with respect to one another" to "Check here if there are 2 variants that are proven or suspected to be in trans with respect to one another. If you wish to document information on variants that are not in trans, please add details to the non-scorable evidence section of the PMID summary." Here, add a link back to GDM landing page.

(2) Below this checkbox in the same section of the modal, add three additional checkboxes with the following text: "Please specify the phase status of the two variants: " O proven in trans O suspected in trans O unknown "O" is meant to represent a checkbox

https://broadinstitute.atlassian.net/browse/CGSP-578

[wrightmw]

@ikeseler @courtneythaxton I started a ticket... see above... please feel free to edit....

[ikeseler]

@courtneythaxton please see my changes to the ticket and make corrections/changes if needed. Also, for part (2) of the ticket, does this information need to be displayed in the evidence summary tables and on the website?

[wrightmw]

@courtneythaxton Do you want the logic that if they select unknown then they can't score the variant?

[wrightmw]

@ikeseler For your question above, I don't think it needs to be an update to evidence summary since, if it's unknown it will go into the evidence summary under the non-scorable evidence table. Only the proven or suspected trans ones should be scored.

[bmpbowen]

Mock-up: https://docs.google.com/presentation/d/1yAZBlGuQjeTTdRa5qC-D3QHM_mwDR-8w/edit#slide=id.p1

[gcheung-SF]

1. The options for specify the phase section are checkbox(allow multi-selection) or radio button (only one selection)?
2. If "Unknown" is selected, cannot add variants? Or allow adding variants but just cannot score/review/contradicts?
3. Same logic apply to MOI = Semidominant inheritance and proband is = "Biallelic compound heterozygous" which is treated as AUTOSOMAL_RECESSIVE?

[bmpbowen]

1. Radio button
2. Allow adding variants, but cannot score
3. The original request was that this logic only apply to AR GDMs, but I will ask the gene curation small group on Monday to confirm whether or not they want to extend this to semidominant GDMs with biallelic compound heterozygous observations

[bmpbowen]

@gcheung-SF I followed up with the gene curation small group and they thank you for the catch. Please extend the same phase options for semidominant GDMs as for AR GDMs.

[gcheung-SF]

A couple more questions:

1. Should the phase status selection be required before allowing to add variants/scores?
2. Are the phase status options available if individual is not proband?

[courtneythaxton]

@gcheung-SF I think it makes sense to ask them to add phase status before allow variant scores, that way we can get this information and it prompts the user to think about this when scoring.

I think there is no harm in adding the phase status for individuals not the proband.

[bmpbowen]

Agree with Courtney.

[bmpbowen]

Tested out this feature on the test site. It looks great and mostly works as expected. The menu of phasing options appears as expected when the two variants box is checked.

1 small change needed: After adding two variants, if I then uncheck the two variants option, an error pops up which contains a typo, "varaint(s)":

Instead, it should read "Please clear the variant(s) before changing phase status to Unknown"

This error also pops up when attempting to switch from Proven to Unknown or Suspected to Unknown

I don't think we actually need to remove the variants just because the type of phase may have changed. I would remove the error message in this case.

1 additional suggestion: I think if a user checks the two variants box, then they should not be allowed to save until they provide both variants. I was able to save after only adding one variant, even though the two variants box was checked.

[gcheung-SF]

Here's the reason for the messages:

1. The message is to make sure user knows when unselecting two trans means the variants will be deleted. And this is shown no matter what the phase status is. We can omit the message if ok to just delete the variants.
2. This message is necessary because the added variants are different when individual is proband. For Proven to Unknown or Suspected to Unknown, add variantScores which includes score. For Unknown, it's just variant and no score. No message when individual is not proband. We can omit message if ok to just delete the variants.
3. We can add that.

[bmpbowen]

1. Agree with error message use in 1. above.
2. Seems easier to just keep the error message then.
3. Thanks!

[bmpbowen]

We received an additional request to add phase status to the preview evidence scored table: Please add 1 column for phase status after the variants column in preview evidence scored. The cells should be collapsed by proband (similar to proband sex). Please ensure that adding more than two variants per individual in the GCI won’t break the phase functionality. If it does, restrict addition of variants to two variants per individual. No need to show phase status on the classification matrix, just the preview evidence summary.

Mock-up:

[gcheung-SF]

@bmpbowen,

In the form, we have options "Proven in trans" and "Suspected in trans". But in the evidence summary, the options become "Confirmed Trans" and "Suspected Trans". So which values should be stored in database and pass down to downstream? Thanks.

[bmpbowen]

Ah, that's my error Gloria. Please keep the same values across both forms - "Proven in trans" and "Suspected in trans" are the correct terms.