Abstract
Genetics as a science has roots in studying phenotypes of relatives, but molecular approaches facilitate direct measurements of genomic variation between individuals. Agricultural and human biomedical research are both emphasizing genotype-based instruments, such as polygenic scores, but unlike in agriculture, there is an emerging consensus that family variables act nearly independently of genotypes in models of human disease. However, there is insufficient theoretical treatment of these scores, especially guiding our understanding of how and why scores derived from different sources of data may combine. To advance our understanding of this phenomenon, we use 2,066,057 family records of 99,645 genotyped probands from the Integrative Psychiatric Research (iPSYCH)2015 case-cohort study to show that state-of-the-field genotype- and phenotype-based genetic instruments explain largely independent components of liability to psychiatric disorders. We support these empirical results with theoretical analysis and simulations to describe, in a human biomedical context, parameters affecting current and future performance of the two approaches, their expected interrelationships, and consistency of observed results with expectations under simple additive, polygenic liability models of disease. We conclude, at least for psychiatric disorders, that the low correlation between current phenotype- and genotype-based genetic instruments is caused by both being noisy measures of additive genetic liability. We expect they should remain complementary over the near future and therefore expect approaches integrating both sources of information to achieve more power for genetic inference.