The comparisons made between datasets must be considered in light of several limitations of the current study. There are substantial differences among the examined samples in terms of age (CADD and MCTFR represent adolescent data, while SAGE was comprised of adults), sex composition (MCTFR and SAGE are split evenly by sex, while CADD has an overrepresentation of males due to the sampling scheme), and diversity of ancestry (MCTFR is less diverse than either CADD or SAGE, which each have different representations of non-Caucasian ancestry groups). The sampling schemes of CADD and SAGE aimed to increase power to detect effects by oversampling extreme phenotype individuals, whereas the MCTFR study is closer to community-representative.
