`count` field in `ancestry_distribution` in `scores`

[pgscatalog]

`count` field in `ancestry_distribution` in `scores`

[ramiromagno]

Thank you so much @smlmbrt !

I was feeling unsure about the 0 indeed. Going to make NA then.

Edit: I could sum the sample size across all unique sample sets, but that would imply another request to a different endpoint. And I am trying to keep a one-to-one correspondence between the endpoints and PGS entities.

[smlmbrt]

The count field is really just meant to show the denominator for the percentages - we explain this on the website and think it's ok to have them refer to slightly different quantities as the purpose of these distributions is to get an overview of the diversity.

I'd recommend setting the n_sample_sets as NA for gwas and dev to remain consistent with the Catalog - PSS aren't assigned to those stages, so the 0 may be misleading. It is possible however, to derive the total sample size summed across all unique sample sets in the eval stage in the $ancestries table though!