Genome-Wide Association Study

A genome-wide association study (GWAS) is a statistical scan of the entire genome to identify genetic variants — typically single nucleotide polymorphisms (SNPs) — that occur more frequently in individuals with a particular trait or disease than in those without it. By testing millions of variants simultaneously in large sample sizes, GWAS has identified thousands of genetic associations for complex traits ranging from height and intelligence to schizophrenia and diabetes.

The method's power comes from scale. Where QTL mapping in experimental crosses had limited resolution and required controlled breeding, GWAS exploits the natural recombination history of human populations to map associations to narrow genomic regions. Sample sizes in major GWAS now exceed a million individuals, producing association signals that are statistically robust even after severe multiple-testing correction.

But GWAS has not delivered the biological mechanism that its promoters promised. The associated variants typically explain only a small fraction of trait variance — the 'missing heritability' problem — and the vast majority fall in non-coding regions whose functional interpretation is uncertain. GWAS finds correlations between DNA markers and phenotypes; it does not find genes, pathways, or developmental processes. The jump from association to mechanism requires experimental work that GWAS does not provide and that the GWAS literature increasingly neglects.

The deeper issue is conceptual. GWAS assumes that complex traits are the sum of many independent genetic effects, each acting additively on a scalar phenotype. This is the quantitative genetics model applied to molecular data, and it inherits all the limitations of that model. Gene-environment interaction, epistasis, and developmental dynamics are treated as nuisances to be controlled rather than as central features of the traits being studied.

GWAS is the most expensive correlation matrix in the history of science. Its outputs are real, but their interpretation requires a theory of development that GWAS does not possess and that the field has shown little interest in building.