What is the estimated relative distribution of gnomAD ancestry groups in the global population?

I understand how the gnomAD 4.1 ancestry group designations are assigned, but I’m curious to know if there’s any estimate of the proportion of the global population that would be assigned to each category using the method. For example, while EAS and SAS represent a small portion of the samples, they should represent 50% of the global population (assuming that they correlate very well with persons of east asian or south asian ancestry).

Is there any good estimate of the parent population size of each group?

There are census studies about population sizes around the world that can be used to estimate this. For example, the Martin et al. Nature Genetics 2019 paper had this figure in it.

I’m not sure what they used to make the global bar on the right but it should give you an overall sense of the estimate of the parent population sizes.

The question of what fraction of individuals would be sorted into particular genetic ancestry groups by our approach is hard to determine.