Genetic Correlation

A genetic correlation is the correlation between the genetic effects on two different traits, estimated from the covariation of their breeding values in a population. It measures the extent to which the same genes, or linked genes, influence both traits. A positive genetic correlation means that selecting for one trait will produce a correlated response in the other; a negative genetic correlation means that improvement in one trait comes at a genetic cost to the other.

The concept is central to quantitative genetics and agricultural breeding, where unfavorable genetic correlations — such as the negative correlation between milk yield and fertility in dairy cattle — constrain the rate of genetic improvement. In human genetics, genetic correlations between traits like height and intelligence, or between schizophrenia and bipolar disorder, have become major research findings, though their interpretation is contested.

The statistical estimation of genetic correlations relies on the same variance-partitioning assumptions as heritability: additive gene action, random mating, and independence of genetic and environmental effects. When these assumptions are violated — by gene-environment interaction, epistasis, or population stratification — the estimated genetic correlation may be a statistical artifact rather than a biological reality. The field has been slow to confront this possibility because genetic correlations are computable, publishable, and narratively compelling, even when their biological meaning is uncertain.

Genetic correlation is a statistical shadow cast by the additive model. It tells us what would happen if genes acted additively in a randomly mating population with no environmental structure. Whether that shadow corresponds to any biological process is a question the method cannot answer.