Genetic Load¶
The proportional shortfall of a population's mean genotype-associated fitness below a declared optimal-genotype reference in a specified setting.
Core Idea¶
Genetic load measures how far a population's mean genotype-associated fitness falls below a stated optimal-genotype reference, as a proportion of that reference. If genotype frequencies are \(p_i\) and their comparable fitnesses are \(w_i\), then \(\bar w=\sum_i p_iw_i\) and \(L=(w_{\mathrm{ref}}-\bar w)/w_{\mathrm{ref}}\) for positive \(w_{\mathrm{ref}}\). The reference may be an actually available best genotype or a hypothetical optimum. Its choice, the fitness component and the environment must be stated for \(L\) to mean anything stable.[ref-5a11322f2269][ref-a56fb11ec996]
The measure is broader than mutational load. Recurrent harmful mutations can reduce mean fitness, but so can segregation when a fit heterozygote produces less-fit homozygotes. Crow explicitly separated the two; the redirected Mutational load candidate remains a narrower unresolved identity, not an alias of this entry.[^ref-5a11322f2269]
Scope of Application¶
The literal setting is population genetics: modeled or estimated genotype frequencies and fitnesses measured on one scale, plus a declared genetic benchmark. Crow's original mutation–selection and heterozygote-advantage cases both instantiate the index but have different causes. If fitness is relative to competitors, a numerical load alone does not imply a fixed number of nonreproducing individuals or extinction risk. Those claims need separate absolute-fitness and demographic assumptions.[ref-5a11322f2269][ref-ad85cdd9c8c2]
Clarity¶
The key questions are mean shortfall from what reference, under what fitness definition, and due to which mechanism? \(L\) answers the first only after a reference is chosen. A mutational cause must be established with a mutation–selection model; a segregation cause follows a different inheritance model. Thus “high load” is not a synonym for “many deleterious variants,” and the same number can have different causes.[^ref-5a11322f2269]
Manages Complexity¶
The weighted mean \(\bar w\) compresses many genotype frequencies and fitnesses into a single number. Normalization turns its difference from the optimum into a dimensionless proportion. This allows within-model comparison, but the compression loses the identities of contributing genotypes, their inheritance rules and the model's environmental assumptions. A total \(L\) cannot decompose itself.[ref-5a11322f2269][ref-a56fb11ec996]
Abstract Reasoning¶
In Crow's mutation model, frequencies \(p^2,2pq,q^2\) and fitnesses $1,1-hs,1-s$ give \(L=2pqhs+q^2s\) relative to reference $1$. Under the unlike overdominance model, fitnesses \(1-s,1,1-t\) give \(L=sp^2+tq^2\) against the hypothetical all-heterozygote reference. Both are load calculations; only the first assigns a recurrent-mutation cause. Their equilibrium formulas need additional assumptions and should not be treated as definitions.[^ref-5a11322f2269]
Knowledge Transfer¶
The index transfers literally between population-genetic models when their genotype distributions, fitness scales and references are restated. Beyond genetics, the portable mathematical skeleton is the proposed live parent Relative Change: a shortfall from a nonzero reference divided by that reference. An analogous economic or engineering ratio is not itself genetic load because its operands are not population genetic fitness.
[^ref-5a11322f2269]: James F. Crow, “Some Possibilities for Measuring Selection Intensities in Man”, Human Biology 30:1–13 (1958), original scan, printed pp. 768–771 (scan pp. 6–9); publisher archive. [^ref-ad85cdd9c8c2]: Yann Lesecque et al., “A Resolution of the Mutation Load Paradox in Humans”, Genetics 191:1321–1330 (2012), original research, Introduction and Discussion. [^ref-a56fb11ec996]: “Inbreeding Load in Finite Populations from Dominant and Overdominant Mutations”, original research (2025), Results §§(b)–©.
Relationships to Other Abstractions¶
Current abstraction Genetic Load Domain-specific
Parents (1) — more general patterns this builds on
-
Genetic Load is a kind of Relative Change Prime
Genetic load is a proportional difference between a fitness reference and a population mean.
Hierarchy path (1) — routes to 1 parentless root
- Genetic Load → Relative Change → Ratio → Comparison → Self Checking
Neighborhood in Abstraction Space¶
Genetic Load sits in a moderately populated region (57th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.
Family — Selection, Speciation & Experimental Evolution (22 abstractions)
Nearest neighbors
- Frequency-Dependent Selection — 0.88
- Underdominance — 0.86
- Complete mixing — 0.85
- Theil Index — 0.84
- Hardy-Weinberg Principle — 0.84
Computed from structural-signature embeddings · 2026-10-08