Some of this we can resolve:
Human SNPs (on the array) are skewed towards being rare; few SNPs are correlated cos we didn’t use imputations; null/realistic simulations show TarGene’s good performance academic.oup.com/biostatistic...
We think we know why most hits are protective & will update
Thanks!
academic.oup.com
Semiparametric efficient estimation of small genetic effects in large-scale population cohorts
Summary. Population genetics seeks to quantify DNA variant associations with traits or diseases, as well as interactions among variants and with environmen
