Abstract
Accurate estimation of genetic correlation requires large sample sizes and access to genetically informative data, which are not always available. Accordingly, phenotypic correlations are often assumed to reflect genotypic correlations in evolutionary biology. Cheverud’s conjecture asserts that the use of phenotypic correlations as proxies for genetic correlations is appropriate. Empirical evidence of the conjecture has been found across plant and animal species, with results suggesting that there is indeed a robust relationship between the two. Here, we investigate the conjecture in human populations, an analysis made possible by recent developments in availability of human genomic data and computing resources. A sample of 108,035 British European individuals from the UK Biobank was split equally into discovery and replication datasets. 17 traits were selected based on sample size, distribution and heritability. Genetic correlations were calculated using linkage disequilibrium score regression applied to the genome-wide association summary statistics of pairs of traits, and compared within and across datasets. Strong and significant correlations were found for the between-dataset comparison, suggesting that the genetic correlations from one independent sample were able to predict the phenotypic correlations from another independent sample within the same population. Designating the selected traits as morphological or non-morphological indicated little difference in correlation. The results of this study support the existence of a relationship between genetic and phenotypic correlations in humans. This finding is of specific interest in anthropological studies, which use measured phenotypic correlations to make inferences about the genetics of ancient human populations.
Introduction
In evolutionary biology and anthropology, a central challenge is understanding the degree to which traits share a common genetic foundation. This relationship, known as genetic correlation ($r_g$), is vital for predicting how traits co-evolve or how specific mutations might influence multiple systems. However, calculating $r_g$ is historically difficult, requiring massive genomic datasets and complex statistical power—resources rarely available for ancient human populations or under-studied groups.
In 1988, James Cheverud proposed a solution now known as "Cheverud’s Conjecture." He suggested that phenotypic correlations ($r_p$)—the observable relationships between physical measurements—could serve as reliable proxies for genetic correlations. While this "phenotypic gambit" has been validated in various plant and animal species, its applicability to human populations remained an untested assumption until this recent study.
Methodology
To test the conjecture in humans, researchers utilized a massive dataset from the UK Biobank, focusing on 108,035 British European individuals. The sample was split into two equal groups—discovery and replication—to ensure the findings were statistically robust and reproducible.
Linkage Disequilibrium Score Regression (LDSC)
The study employed LD Score Regression (LDSC) to estimate genetic correlations. Unlike traditional methods that require individual-level genotype data, LDSC utilizes summary statistics from Genome-Wide Association Studies (GWAS). This method is particularly effective as it accounts for "noise" caused by linkage disequilibrium—the tendency of nearby genes to be inherited together—allowing for efficient estimation of $r_g$ across a wide range of traits.
Trait Selection
The researchers selected 17 traits based on their heritability and sample size. These were categorized into two groups:
- Morphological Traits: Physical structures such as height, Body Mass Index (BMI), and grip strength.
- Non-Morphological Traits: Physiological or behavioral measures, including blood pressure and reaction time.
Key Results
The study found a near-perfect relationship between genetic and phenotypic correlations. In the discovery set, the Pearson correlation between $r_g$ and $r_p$ was 0.97, while the replication set yielded a correlation of 0.96.
Additional findings included:
- Minimal Disparity: The mean absolute difference between the genetic and phenotypic correlation values was only 0.06.
- Genetic Magnitude: On average, genetic correlations were slightly higher (by approximately 0.04) than their phenotypic counterparts.
- Heritability: The selected traits showed a mean SNP-heritability of 0.20, providing sufficient signal for the LDSC method to produce high-confidence results.
Interestingly, the conjecture held up across both trait categories. While non-morphological traits generally exhibited lower overall correlation magnitudes than physical ones, the relationship between their genetic and phenotypic structures remained consistent with Cheverud’s predictions.
Discussion: The "Phenotypic Gambit" in Humans
The high concordance between $r_g$ and $r_p$ suggests that environmental effects often align with genetic effects. If a genetic factor increases a trait's value, environmental factors (such as nutrition or lifestyle) often act through similar biological pathways, keeping the correlations in sync.
For anthropologists, these results provide a rigorous statistical foundation for using skeletal measurements to infer the genetic architecture of ancient populations. It confirms that when we observe the proportions of ancient remains, we are viewing a reliable mirror of their underlying genetics.
Limitations and Future Directions
While the results are definitive for the population studied, the authors acknowledge several limitations:
- Population Specificity: The study was limited to a relatively homogeneous group of British Europeans. Further research is needed to determine if these results hold in more genetically diverse populations or those living under different environmental pressures.
- Trait Spectrum: The 17 traits analyzed do not cover the entire human "phenome." Behavioral or complex life-history traits (such as age at first birth) may not adhere to the conjecture as strictly as morphological traits.
- Environmental Stability: The UK Biobank participants live in a modern, stable environment. In ancient populations facing extreme stressors like famine or disease, phenotypic plasticity might create a wider gap between genotype and phenotype.
Conclusion
This study provides long-awaited validation of Cheverud’s Conjecture in humans. By demonstrating that phenotypic correlations are excellent proxies for genetic correlations, the research supports the continued use of physical measurements to make meaningful inferences about human evolutionary history and genetic architecture.
Summary updated