The relationship between genotype- and phenotype-based estimates of genetic liability to psychiatric disorders, in practice and in theory
(2026) In American Journal of Human Genetics 113(1). p.184-201- 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... (More)
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.
(Less)
- author
- author collaboration
- organization
- publishing date
- 2026-01-08
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- family genetic risk scores, genetic liability, genetics, polygenic risk scores, psychiatric disorders
- in
- American Journal of Human Genetics
- volume
- 113
- issue
- 1
- pages
- 18 pages
- publisher
- Cell Press
- external identifiers
-
- scopus:105026720771
- pmid:41435841
- ISSN
- 0002-9297
- DOI
- 10.1016/j.ajhg.2025.11.016
- language
- English
- LU publication?
- yes
- id
- 3ae493bc-7489-4e19-bead-b04e642a78c6
- date added to LUP
- 2026-03-12 15:41:48
- date last changed
- 2026-06-06 06:36:50
@article{3ae493bc-7489-4e19-bead-b04e642a78c6,
abstract = {{<p>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.</p>}},
author = {{Dybdahl Krebs, M. and Appadurai, Vivek and Georgii Hellberg, Kajsa Lotta and Ohlsson, Henrik and Steinbach, Jette and Pedersen, Emil and Werge, Thomas and Sundquist, Jan and Sundquist, Kristina and Border, Richard and Cai, Na and Zaitlen, Noah and Dahl, Andy and Vilhjalmsson, Bjarni and Flint, Jonathan and Bacanu, Silviu Alin and Kendler, Kenneth S. and Schork, Andrew J.}},
issn = {{0002-9297}},
keywords = {{family genetic risk scores; genetic liability; genetics; polygenic risk scores; psychiatric disorders}},
language = {{eng}},
month = {{01}},
number = {{1}},
pages = {{184--201}},
publisher = {{Cell Press}},
series = {{American Journal of Human Genetics}},
title = {{The relationship between genotype- and phenotype-based estimates of genetic liability to psychiatric disorders, in practice and in theory}},
url = {{http://dx.doi.org/10.1016/j.ajhg.2025.11.016}},
doi = {{10.1016/j.ajhg.2025.11.016}},
volume = {{113}},
year = {{2026}},
}