Locus Drawer object in Credible Sets Widgets

[buniello]

The locus drawer will open from the number field in the credible set size column, for both GWAS and molQTLs widget. if credible sets size is 7, the drawer will contain 7 rows for 7 variants in the locus.

Credible sets widget ticket

Discussed with @DSuveges and Genetics Team.

Pending credible sets API deployment at time of drafting this, we have scoped this object with this schema:

"locus": [
    {
      "is95CredibleSet": true,
      "is99CredibleSet": true,
      "logBF": null,
      "posteriorProbability": 0.8281915098204469,
      "variantId": "3_50557710_C_T",
      "pValueMantissa": null,
      "pValueExponent": null,
      "beta": null,
      "standardError": 0.9999996011812708,
      "r2Overall": 1.0000000000000027
    },

TABLE/DRAWER:

1. HEADER: VARIANT ID - variantId
2. HEADER: P-VALUE - pValueMantissa & pValueExponent (Table will be sorted/sortable by this column)
3. HEADER: BETA - beta - Tooltip: Beta with respect to the ALT allele
4. HEADER: SE standardError-Tooltip: Standard Error: Estimate of the standard deviation of the sampling distribution of the beta
5. HEADER: LD (r2) - r2Overall - Tooltip: Linkage disequilibrium with the queried variant
6. HEADER: POSTERIOR PROBABILITY - posteriorProbability - Tooltip: Posterior inclusion Probability from fine-mapping that this variant is causal
7. HEADER: LOG(BF) - logBF
8. HEADER: IS IN 95% CREDIBLE SET - is95CredibleSet

[d0choa]

We are not considering representing as much data, but this is an equivalent view for inspiration: https://r11.finngen.fi/region/I9_HYPTENS/20:58876654-59376654

Correction: There is a difference in the fact we are representing a study-locus whereas they are representing a locus across all studies.

[d0choa]

A few comments from @addramir:

• There will be top-level information about the study-locus (credible set widgets) additional to the table
• is95CredibleSet column is not required in the table
• variantId and posteriorProbability are not nullable. Everything else is optional.

[d0choa]

As discussed on August 7th, we will make progress on the credible set API endpoint while @gjmcn is away, so it simplifies the locus drawer implementation

[buniello]

I would still keep the is95CredibleSet field in table (same as current OTG) as my understanding from discussion with Daniel S is that not all variants are in the 95% (but they all are in 99%).

[buniello]

@gjmcn - please pause implementation on this for now (more scoping happening in the background)

[buniello]

this task is not planned anymore as replaced by this