How Do Polygenic Traits Create Continuous Variation?
Polygenic traits produce continuous phenotypic variation because they are governed by multiple genetic loci acting together with environmental influences. Rather than sorting individuals into discrete categories, the small additive effects of numerous alleles combine across the genome to generate a smooth spectrum of values. In quantitative biology, models such as the infinitesimal model and statistical distributions illustrate how individual Mendelian contributions accumulate into standard normal distributions for complex characteristics like human height.
The Additive Model of Polygenic Inheritance
Classical Mendelian inheritance focuses on monogenic traits, where variations at a single locus result in distinct phenotypic classes. In contrast, quantitative traits are polygenic, influenced by dozens to thousands of genetic loci.
Under the additive genetic framework formulated by Ronald Fisher in 1918, each contributing allele at each locus confers a small increment to the overall phenotype. If a trait is governed by \(n\) biallelic loci with alleles \(A\) and \(a\), where \(A\) adds an effect size of \(d\) and \(a\) has an effect size of \(0\), an individual's genetic value \(G\) is the sum of these contributions:
\[G = \sum_{i=1}^{n} (x_i \cdot d_i)\]
Here, \(x_i \in \{0, 1, 2\}\) represents the count of effect alleles at locus \(i\), and \(d_i\) represents the additive effect of that locus. As the number of contributing loci \(n\) increases, the number of possible discrete genotypic combinations grows exponentially according to \(3^n\), narrowing the phenotypic distance between adjacent genotypes.
The Central Limit Theorem and Phenotypic Distributions
The emergence of a smooth, bell-shaped distribution from discrete genetic inputs is an empirical manifestation of the Central Limit Theorem. When a biological outcome is the sum of many independent or weakly linked random variables, the resulting sum tends toward a normal distribution regardless of the underlying distributions of individual loci.
In quantitative biology, the phenotypic value \(P\) of an individual is partitioned into genetic and environmental components:
\[P = G + E\]
Where:
- \(P\) is the observed phenotypic measurement, such as total height in centimeters.
- \(G\) is the cumulative genetic value, which can be further decomposed into additive effects (\(A\)), dominance deviations (\(D\)), and epistatic interactions (\(I\)): \(G = A + D + I\).
- \(E\) represents environmental variance, including nutrition, childhood health, and non-genetic biological factors.
Even if genetic values \(G\) were discrete, environmental variation \(E\) is inherently continuous and acts as a biological smoothing filter over the genotypic classes. When both components combine, the total phenotypic variance \(V_P\) is expressed as:
\[V_P = V_G + V_E + 2\text{Cov}(G,E)\]
Assuming negligible gene-environment covariance, \(V_P = V_G + V_E\). The broad-sense heritability \(H^2\) is defined as \(V_G / V_P\), while narrow-sense heritability \(h^2\), which drives the resemblance between relatives, is defined as:
\[h^2 = \frac{V_A}{V_P}\]
For human height, narrow-sense heritability is estimated to be approximately \(0.80\), indicating that while the majority of phenotypic spread originates from additive genetic variation, environmental variance supplies the continuous buffer that blurs any remaining discrete genetic boundaries.
The Omnigenic Architecture of Human Height
Modern genome-wide association studies (GWAS) have confirmed that human height conforms to an omnigenic architecture. Instead of a few master genes determining stature, tens of thousands of variants across nearly the entire genome influence bone development, chondrocyte biology, extracellular matrix composition, and growth hormone pathways.
Most associated single-nucleotide polymorphisms (SNPs) have minute individual effect sizes, often shifting total adult height by fractions of a millimeter:
\[d_i < 1\text{ mm}\]
Because human populations carry varying mixtures of thousands of these small-effect variants, each individual inherits a unique aggregate score. When evaluated in large cohorts, the cumulative distribution of these genetic values produces the continuous, bell-shaped Gaussian distribution observed in quantitative physiology.