The P-value thresholds that maximise the power and AUC are more permissive than the usual thresholds for individual markers. This means that polygenic analyses can be powerful while still including many non-significant markers, so they will continue to be useful as long as individually associated markers remain to be discovered. Although larger samples are needed for useful risk prediction, polygenic scores have an ongoing current role in assessing the variance explained by marker panels and the genetic correlation between related traits and populations.
