I am calculating PRS scores from the the PGS catalog, using their tool pascal_calc.
I have a dataset from 1000 genomes with many samples from different individuals in it and I have a dataset that I got from 23andme of one individual. Since they have a different number of variants, can I compare the score I get for the 23andme dataset with the scores I get from the large 1000genomes dataset to evaluate it and say if it is high or low? (With score I mean the actual sum of weighted variants divided by number of non missing genotypes used to calculate the score)
I am not sure if there is a flaw in my logic here, since they use a different number of variants when calculating the scores from 1000genomes and the 23andme one. Does dividing by the number of non missing genotypes for both scores make it ,so you can compare the 2 scores?
From what you mentioned, it seems like you want to calculate score in 1000G and 23and me, then compare them? My guess is you are trying to use 1000G as control and 23 and Me as cases? If that is the case, even if you can compare the scores, your results will only show difference between the 1000G and 23 and me, not the difference between case / control