Missing heritability

Missing heritability is a concept in genetics that describes the significant gap between two different ways of measuring the genetic contribution to a trait or disease. For decades, scientists have used family and twin studies to estimate heritability—the proportion of variation in a trait (like height, intelligence, or risk for schizophrenia) that can be attributed to genetic differences. For many common traits, these studies suggested a large genetic component, with heritability estimates often exceeding 50-80%. This led to the expectation that a handful of genes with substantial effects would be responsible.

However, with the advent of genome-wide association studies (GWAS) in the mid-2000s, a different picture emerged. GWAS scan the genomes of thousands of individuals to find specific genetic variants, most often single-nucleotide polymorphisms (SNPs), that are associated with a particular trait. While these studies successfully identified hundreds or thousands of variants linked to complex traits, the individual effect of each variant was typically very small. When added together, all the identified variants could only account for a small fraction—often less than 10-20%—of the heritability estimated from twin studies. This puzzling shortfall is what geneticists termed the "missing heritability."

Context and Explanations

The problem of missing heritability sparked a major debate and catalyzed new avenues of research into the genetic architecture of complex traits. Several hypotheses have been proposed to explain the gap, and it is now widely accepted that a combination of factors is responsible:

• Polygenicity (Many Variants of Small Effect): The primary explanation is that most complex traits are not governed by a few genes of large effect, but by thousands, or even tens of thousands, of common genetic variants, each contributing an infinitesimally small amount. Early GWAS were not statistically powerful enough to detect these tiny signals, but as study sizes have grown into the millions, more of the heritability is being accounted for.

• Rare Variants: GWAS are designed to detect common variants (those present in >1% of the population). Rare variants, which are unique to individuals or small families, may have larger effects but are missed by standard GWAS. Whole-genome sequencing is now being used to investigate their contribution.

• Other Types of Genetic Variation: GWAS typically focus on SNPs, but other forms of variation, such as copy-number variations (deletions or duplications of DNA segments) and other structural variants, also contribute to traits and are not always well-captured.

• Complex Interactions: The models used in GWAS often assume that genes act independently. However, gene-gene interactions (epistasis) and gene-environment interactions may play important roles that are difficult to detect statistically.

• Overestimation of Heritability: Some researchers argue that classical twin studies may have overestimated heritability by not fully accounting for shared environmental factors that make identical twins more similar than fraternal twins.

The discrepancy between the heritability of a trait as estimated from family and twin studies and the proportion of that heritability that can be explained by currently identified genetic variants.

The concept of missing heritability has fundamentally shifted the scientific understanding of the genetic basis of common diseases and traits. It marked a move away from the search for a single "gene for" a condition towards a more complex, polygenic model. This has profound implications for medicine and public health. It explains why genetic prediction for conditions like heart disease, depression, or type 2 diabetes is probabilistic rather than deterministic.

This challenge has driven innovation, leading to the formation of massive international research consortia to increase statistical power and the development of new tools like polygenic risk scores (PRS). These scores aggregate the small effects of millions of genetic variants to provide a more comprehensive estimate of an individual's genetic predisposition to a disease. While much of the heritability is no longer considered "missing" but rather distributed across the genome in tiny fragments, the term remains a useful descriptor for the historical and ongoing challenge of fully mapping the genetic landscape of human traits.