Calculation of different populations

[labxYZY]

Regarding the calculation of PGS by subpopulation:
I am using your tool to calculate PGS scores for different populations in UK Biobank.
Here I am doing Quality control for each of the different populations, so I end up with different numbers of SNPs left in the different populations.
The scores in the PGS catalogue I'm utilising are all European, which seems to result in a very low match rate of just a few per cent for both Asian and black people when I'm calculating the scores. What should I do at this point? Is it better to keep going to the PGS scores obtained from the multiracial development?

Translated with DeepL.com (free version)