Hello,
I get different results when I run PCA using Plink2.0 and 1.9. This is despite running it on the same dataset. The dataset has already been QC'd (MAF, LD, HWE). The results from 1.9 match closely what I get from Hail and bignspr. However for plink2.0 i get no correlation and produce different population structures. Am I missing something?
My code is the following:
QC
plink --vcf $FILE_VCF --make-bed --out $FILE ---mind 1 --geno 0.01 --maf 0.05 --hwe 0.001 --indep-pairwise 50 5 0.5
PCA
module load plink/1.90b6.24
plink --bfile $FILE --pca 10 --out $FILE_pca_1
plink --bfile $FILE --pca 10 --out $FILE_pca_2