{
  "id": "genetic-dominance",
  "title": "Genetic Dominance",
  "lang": "en",
  "source_lang": "en",
  "summary": "Genetic dominance describes how the phenotype of a heterozygote relates to those of the corresponding homozygotes at a genetic locus.",
  "category": "science",
  "subcategory": "biology",
  "type": "concept",
  "aliases": [
    "Dominance in genetics",
    "Allelic dominance"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/genetic-dominance/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/genetic-dominance/",
      "json": "https://aiwiki.page/zh/genetic-dominance.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Genetic dominance is a relationship between alleles defined by their effects on a specified phenotype. In a diploid organism, which normally carries two copies of each autosomal gene, an individual with two different alleles is a heterozygote. If its phenotype is indistinguishable from that of an individual carrying two copies of one allele, that allele is described as dominant to the other for the phenotype being measured. The other allele is recessive. Dominance does not mean that an allele is more common, more beneficial, more strongly inherited, or necessarily more highly expressed.\n\nThe concept originated in nineteenth-century breeding experiments and became central to genetics, but its modern meaning extends well beyond simple dominant–recessive classifications. Heterozygotes can have intermediate phenotypes, simultaneously display distinguishable products of both alleles, or lie outside the range of the two homozygotes. The same allelic pair can also show different dominance relationships for different measurements, tissues, or environments.\n\nDominance connects molecular processes with inheritance, disease, breeding, and evolution. Its explanation requires both an account of how gene products function and a precise definition of the trait under consideration. It is therefore best understood as a property of an allele pair in a biological context, rather than as an intrinsic label attached permanently to a gene."
    },
    {
      "heading": "Definition, notation, and scope",
      "text": "For a locus with two alleles, conventionally written \\(A\\) and \\(a\\), the three possible diploid genotypes are \\(AA\\), \\(Aa\\), and \\(aa\\). The two homozygotes carry identical alleles, while the heterozygote carries different alleles. Complete dominance of \\(A\\) means that \\(AA\\) and \\(Aa\\) have the same phenotype under the conditions used to classify them, whereas \\(aa\\) differs. The relevant comparison is between these three genotypes, not between the physical strength or abundance of two DNA sequences.\n\nDominance must always refer to a particular measurement. A heterozygote may resemble one homozygote in visible appearance but have an intermediate concentration of an enzyme. A biochemical difference that can be measured in a laboratory need not produce a visible or clinical difference. Consequently, statements such as “the allele is dominant” are shorthand for “the allele is dominant relative to this alternative allele for this phenotype under these conditions.” A gene with several alleles can participate in several different relationships, with no single dominant form.\n\nUppercase and lowercase letters are convenient teaching conventions, not universal biological rules. Formal nomenclature varies among organisms and research communities. Human gene symbols, for example, do not generally encode dominance simply through capitalization. Nor does the reference sequence used to describe a mutation necessarily represent a dominant, common, ancestral, or functionally normal allele. A reference sequence is primarily a coordinate and comparison standard.\n\nDominance is distinct from the processes that determine which allele enters a gamete. The separation of homologous chromosomes during meiosis ordinarily gives the two alleles of a heterozygote equal opportunities for transmission. A dominant allele does not normally displace its recessive counterpart from reproductive cells. Recessive alleles can therefore pass through generations of heterozygotes without producing the phenotype by which they were originally identified.\n\nThe concept applies most directly when two alleles occur together in the same organism. In haploid organisms or haploid life-cycle stages, a single-copy allele usually has no homologous partner against which ordinary dominance can be assessed. Bacteria can nevertheless be studied using partially diploid experimental arrangements. In organisms with more than two chromosome sets, allele dosage and the number of distinct copies make simple diploid categories less adequate: one, two, or several functional copies may be required to cross a phenotypic threshold."
    },
    {
      "heading": "Mendel’s experiments and the development of the concept",
      "text": "Gregor Mendel presented his pea experiments in 1865 and published them in 1866 in Experiments on Plant Hybridization. He crossed lines differing in readily classified characteristics, including seed shape, seed color, and flower color. For the contrasting forms he studied, first-generation hybrids generally displayed one parental form rather than a visible blend. In the next generation, the alternative form reappeared. Mendel called the characters expressed in the hybrids dominant and those that remained latent recessive.\n\nFor a simple cross between true-breeding \\(AA\\) and \\(aa\\) parents, all first-generation offspring are \\(Aa\\). Self-fertilization of those heterozygotes produces an expected genotype ratio of \\(1\\,AA:2\\,Aa:1\\,aa\\). With complete dominance, that becomes a phenotype ratio of three dominant-form individuals to one recessive-form individual. Mendel’s reported total for round and wrinkled seeds in one major experiment was 7,324 seeds: 5,474 round and 1,850 wrinkled. The importance of the result was not an exact numerical fit, but its support for discrete hereditary factors that remained distinguishable across generations.\n\nThese experiments separated inheritance from blending. An intermediate-looking offspring does not necessarily carry physically blended hereditary material, and a recessive phenotype can reappear because its underlying allele has persisted. Mendel did not know the molecular nature of DNA, and his hereditary factors should not be read as a complete anticipation of modern molecular genes. His achievement was to establish a quantitative framework for following hereditary differences through crosses.\n\nMendel’s work gained broad recognition around 1900 through the work of Hugo de Vries, Carl Correns, and Erich von Tschermak. During the following decades, the chromosome theory of inheritance linked hereditary factors to chromosomes, while experiments in plants and animals revealed many cases that did not fit complete dominance. Incomplete dominance and codominance did not overturn segregation; instead, they showed that the relationship between genotype and phenotype was more varied than the simplest pea examples suggested.\n\nThe explanation of dominance became an evolutionary and biochemical problem in the twentieth century. R. A. Fisher proposed in 1928 that selection on modifying genes could help explain the frequent recessiveness of harmful mutations. Sewall Wright emphasized physiological relationships between gene activity and organismal function, including the possibility that half the usual enzyme activity could sustain nearly normal performance. Later work in metabolic control, notably by Henrik Kacser and James Burns, developed quantitative explanations for why changes in a single enzyme often have relatively small effects on overall pathway flux."
    },
    {
      "heading": "Complete dominance, incomplete dominance, and codominance",
      "text": "Complete dominance describes equality, or practical indistinguishability, between the heterozygote and one homozygote for a specified phenotype. Mendel’s round-versus-wrinkled pea seeds provide a familiar example at the level of mature seed appearance. This does not imply that the underlying biochemical states are identical. The classical wrinkled-seed allele affects starch-branching enzyme activity; changes in carbohydrate composition and water relations during development contribute to the wrinkled appearance after drying. A sufficient amount of functional product in heterozygotes can preserve a round-seed classification.\n\nIn incomplete dominance, the heterozygote has a phenotype between those of the two homozygotes. Crosses between particular red-flowered and white-flowered snapdragon lines can produce pink-flowered heterozygotes. When those heterozygotes are crossed, the expected phenotypic ratio is \\(1\\) red:\\(2\\) pink:\\(1\\) white, corresponding directly to the underlying genotype ratio. The visible intermediate does not mean that the alleles merge: the original homozygous forms can reappear in the next generation.\n\nAn intermediate phenotype need not be exactly halfway between the homozygotes. Pigment synthesis, light absorption, tissue structure, and visual perception can all introduce nonlinear relationships. A heterozygote that produces half as much pigment as one homozygote may not look halfway between that homozygote and a pigment-free individual. In quantitative work, the exact midpoint is a special case called additivity at the locus, rather than a necessary feature of every phenotype described loosely as intermediate.\n\nIn codominance, distinguishable effects of both alleles are detectable in the heterozygote. The human ABO blood group system is a standard example. The A and B alleles encode glycosyltransferases with different substrate specificities, producing different carbohydrate structures on cell surfaces. Individuals with an A allele and a B allele can display both A and B antigens. Common O alleles do not produce functional transferase activity of the corresponding kind; in the usual simplified model, A and B are each dominant to O for blood-group classification.\n\nIncomplete dominance and codominance refer to different observational patterns, but neither is a universal molecular label. Two alleles may be codominant when their protein products are distinguished experimentally, yet one may be dominant for an organism-level phenotype. The heterozygote associated with sickle hemoglobin, for example, produces both major hemoglobin forms, although the severe clinical phenotype of sickle cell anemia is ordinarily associated with two relevant disease-causing beta-globin alleles. Classification changes with the trait being described, not with the underlying inheritance of the alleles."
    },
    {
      "heading": "Molecular mechanisms underlying dominance",
      "text": "Many recessive mutations reduce or eliminate the activity of a gene product. Such loss-of-function variants can alter a coding sequence, disrupt RNA processing, reduce transcription, or cause instability of a protein. If one functional copy supplies enough activity for the measured phenotype, a heterozygote will resemble the functional homozygote. This condition is called haplosufficiency. It does not require the functional allele to silence the altered allele; adequate output from one copy can be sufficient.\n\nThe converse, haploinsufficiency, occurs when one functional copy does not provide enough activity for normal function. A heterozygous deletion or inactivating variant can then produce a dominant phenotype. Genes controlling development are often sensitive to dosage because their products regulate many downstream targets, interact with limiting partners, or operate within narrow concentration ranges. The resulting phenotype reflects insufficient total function, not an actively harmful product from the altered allele.\n\nA dominant-negative mutation produces a gene product that interferes with the function of the normal product or its molecular complex. This is especially important for proteins that assemble into multimers or structural networks. If a complex requires several correctly functioning subunits, mutant subunits can compromise complexes that also contain normal subunits. Some collagen disorders illustrate this principle: an altered chain may disrupt assembly or stability rather than merely reduce the total amount of collagen. The severity depends on the affected chain, variant, assembly process, and tissue.\n\nOther dominant variants increase activity, prevent normal regulation, or introduce a new function. These are commonly grouped as gain-of-function mechanisms. A receptor can signal too strongly, remain active without its usual ligand, or respond in an inappropriate tissue. Achondroplasia provides a well-established example involving activating variants in the FGFR3 receptor, whose signaling influences cartilage growth. Toxic new functions form another category: certain altered proteins accumulate or disrupt cellular processes even though a normal allele remains present.\n\nRegulatory changes can also be dominant. A variant in a promoter or another control element can increase expression, shift its developmental timing, or produce expression in the wrong cells. Changes in RNA splicing may generate both normal and abnormal transcripts from the same allele. These mechanisms explain why different variants in one gene can have different inheritance patterns: eliminating a product, increasing its activity, and changing where it is produced are biologically distinct events."
    },
    {
      "heading": "Physiological context, penetrance, and interactions between genes",
      "text": "Dominance depends on the route from molecular activity to a measured trait. For a simple enzyme-catalyzed reaction, reaction rate may vary with enzyme amount when substrate conditions are held fixed. An intact metabolic pathway is more complicated: many reactions, transport processes, substrate pools, and regulatory mechanisms jointly determine its output. A reduction in one enzyme can therefore have little effect on total flux until its activity falls below a consequential range. This helps explain why a heterozygote can have substantially reduced enzyme activity yet show an apparently normal organism-level phenotype.\n\nThe relationship also depends on feedback and compensation. A cell may increase transcription from a remaining functional allele, alter protein turnover, redirect metabolites through alternative pathways, or change substrate supply. Such responses can buffer gene dosage, but they are not universal. Their effectiveness can differ among tissues and developmental stages. A dosage reduction tolerated in one cell type may be consequential in another because of differences in demand, compensating genes, or the abundance of interacting proteins.\n\nPenetrance is the proportion of individuals with a specified genotype who display a defined phenotype. It is distinct from dominance: a dominant disease-associated genotype need not produce the recognized condition in every carrier. Variable expressivity describes differences in the form or severity of a phenotype among individuals with the same relevant genotype. Age, genetic background, environment, and random developmental variation can affect both. An apparently unaffected parent therefore does not, by itself, establish that an allele is recessive.\n\nDominance concerns alleles at the same locus, whereas epistasis concerns interactions between different loci. In Labrador retrievers, for example, one locus influences whether eumelanin is black or brown, while variation at another locus can prevent the usual deposition of that pigment in the coat, producing yellow animals. The effect of the black-versus-brown locus is then obscured by the genotype at the second locus. This is not the same phenomenon as one allele being dominant to another at a single locus.\n\nParent-of-origin effects and chromosome-wide regulation require further distinctions. In genomic imprinting, expression can depend on whether an allele was inherited from the mother or father. In many female mammals, X-chromosome inactivation produces cellular populations that express different parental X chromosomes. These processes can alter the appearance of inheritance and generate tissue mosaics, but neither is simply ordinary dominance. A useful analysis separates allele function, expression state, interactions among genes, and the final phenotypic threshold."
    },
    {
      "heading": "Segregation ratios and patterns of transmission",
      "text": "For a single autosomal locus with complete dominance, the cross \\(Aa \\times Aa\\) gives each offspring an expected probability of \\(1/4\\) for \\(AA\\), \\(1/2\\) for \\(Aa\\), and \\(1/4\\) for \\(aa\\). The corresponding phenotype probabilities are \\(3/4\\) dominant and \\(1/4\\) recessive. These are probabilities for individual reproductive events, not a guarantee that every group of four offspring contains exactly three of one phenotype and one of the other. Sampling variation can be considerable in small families.\n\nA test cross pairs an individual displaying the dominant phenotype with a recessive homozygote. If the first parent is \\(AA\\), all offspring are expected to be \\(Aa\\); if it is \\(Aa\\), the expected offspring ratio is \\(1\\,Aa:1\\,aa\\). This was historically important because phenotype alone could not distinguish the two dominant-phenotype genotypes. Modern molecular assays can often identify alleles directly, but the logic of a test cross remains useful in experimental genetics and breeding.\n\nAutosomal dominant inheritance usually means that one copy of a particular allele is sufficient to produce the phenotype under discussion. A heterozygous parent and a parent lacking that allele have a one-half transmission probability per offspring in the standard model. Autosomal recessive inheritance usually requires relevant variants in both copies of a gene. The two variants need not be identical: an individual carrying two different disease-causing variants in the same gene is a compound heterozygote.\n\nX-linked inheritance differs because most males have one X chromosome and one Y chromosome, while most females have two X chromosomes. For most X-linked genes outside the pseudoautosomal regions, males are hemizygous rather than homozygous or heterozygous. A phenotype in a male carrying one altered X-linked allele does not by itself show that the allele is dominant. Father-to-son transmission is absent for these X-linked loci because fathers ordinarily transmit a Y chromosome to sons.\n\nObserved ratios can depart from simple expectations for reasons unrelated to a change in dominance. Genotypes may differ in survival before scoring, alleles may be transmitted unequally, or phenotypes may be misclassified. A classic example is the dominant yellow allele at the mouse agouti locus: yellow heterozygotes can survive, whereas homozygosity for that particular allele is embryonically lethal. A cross between yellow heterozygotes consequently yields an expected two yellow to one nonyellow ratio among surviving offspring, rather than the usual three-to-one ratio. The deviation reflects viability as well as inheritance."
    },
    {
      "heading": "Dominance in quantitative genetics",
      "text": "For continuously measured traits, dominance is quantified as the deviation of the heterozygote from the midpoint of the homozygotes. Assign the genotypic values\n\n\\[G_{AA}=+a,\\qquad G_{Aa}=d,\\qquad G_{aa}=-a,\\]\n\nafter centering the two homozygotes around zero. Here \\(a\\) is half their difference, and \\(d\\) is the dominance deviation. If \\(d=0\\), the heterozygote is exactly intermediate on the chosen measurement scale. If \\(d=a\\), it equals \\(AA\\); if \\(d=-a\\), it equals \\(aa\\). Values outside the homozygote range indicate overdominance or underdominance for that measurement.\n\nThe ratio \\(d/a\\), when \\(a\\neq 0\\), is sometimes called the degree of dominance. It provides a compact description but is not a molecular explanation. It is also scale-dependent. A relationship that is intermediate for a raw concentration may not remain intermediate after logarithmic transformation, and a continuous molecular measurement can become a dominant–recessive classification when divided by a diagnostic or developmental threshold.\n\nPopulation-level effects depend on allele frequencies as well as genotype values. Let \\(p\\) be the frequency of \\(A\\) and \\(q=1-p\\) the frequency of \\(a\\). Under Hardy–Weinberg equilibrium, the expected genotype frequencies are \\(p^2\\), \\(2pq\\), and \\(q^2\\). With the genotype coding above, the average effect of substituting \\(A\\) for \\(a\\) is\n\n\\[\\alpha=a+d(q-p).\\]\n\nThus, the statistical average effect is not necessarily identical to the difference attributable to an allele in a fixed biochemical setting. It changes with the population in which substitutions are averaged.\n\nUnder the same single-locus assumptions, the additive and dominance contributions to genetic variance are\n\n\\[V_A=2pq\\alpha^2\\]\n\nand\n\n\\[V_D=(2pqd)^2.\\]\n\nThese expressions illustrate an important distinction: dominance at a locus does not imply that dominance variance accounts for most variation in a population. A locus with complete dominance can still contribute substantially to additive variance. Conversely, a strong genotype-level dominance effect may contribute little observable population variance if one allele is very rare.\n\nFor traits influenced by many loci, quantitative genetics partitions variation into components associated with additive effects, dominance deviations, interactions between loci, and environmental influences. The partition depends on population structure and modeling assumptions. Narrow-sense heritability concerns additive genetic variance relative to total phenotypic variance; broad-sense heritability includes genetic variance more generally. Dominance matters for predicting similarities among relatives and crossbred performance, but it should not be treated as the opposite of heritability or as evidence that a trait cannot respond to selection."
    },
    {
      "heading": "Dominance in evolution and population genetics",
      "text": "Dominance affects how efficiently natural selection acts on alleles because selection operates through the phenotypes of individuals carrying them. A newly arisen allele in a large diploid population is initially found almost entirely in heterozygotes. If its effects are recessive, those effects may be weakly expressed or undetectable while the allele remains rare. If its effects are dominant, selection can act on them immediately in heterozygotes. Whether selection favors or opposes the allele depends on its consequences for reproductive success, not on dominance itself.\n\nA common model assigns relative fitnesses\n\n\\[w_{AA}=1,\\qquad w_{Aa}=1-hs,\\qquad w_{aa}=1-s,\\]\n\nwhere \\(a\\) is deleterious, \\(s\\) measures its fitness disadvantage in homozygotes, and \\(h\\) describes its expression in heterozygotes. Here \\(h=0\\) represents complete recessivity of the deleterious effect, \\(h=1/2\\) an intermediate fitness effect, and \\(h=1\\) complete dominance. This \\(h\\) convention concerns fitness and should not be confused with the \\(d/a\\) convention for measured trait values. The same allele may have different apparent dominance for a biochemical trait and for total reproductive fitness.\n\nRecessive deleterious alleles can persist at low frequency because most copies occur in heterozygotes, where selection removes them less effectively. In a simple mutation–selection balance model with complete recessivity, the equilibrium frequency is approximately\n\n\\[q\\approx\\sqrt{\\frac{\\mu}{s}},\\]\n\nwhere \\(\\mu\\) is the forward mutation rate. If the deleterious effect is appreciably expressed in heterozygotes and the allele remains rare, the corresponding approximation is \\(q\\approx\\mu/(hs)\\). These results assume a large, randomly mating population and omit complications such as migration and fluctuating selection.\n\nWhen heterozygotes have greater fitness than either homozygote, heterozygote advantage can maintain both alleles. The association between sickle hemoglobin and protection against severe falciparum malaria is a well-established example of environment-dependent selection. This does not make the allele universally advantageous: the balance depends on malaria exposure and on the consequences of the different genotypes. More generally, trait-level overdominance and fitness overdominance must be distinguished.\n\nInbreeding increases homozygosity and can expose recessive deleterious alleles, contributing to inbreeding depression. Crosses between genetically differentiated lines may restore heterozygosity and mask some of these effects, one contributor to hybrid vigor. These processes interact with genetic drift, population size, and mating structure. The long-standing question of why many loss-of-function mutations are recessive therefore has multiple levels of explanation: biochemical buffering can generate recessivity, while selection and population history shape which variants and modifiers remain in populations."
    },
    {
      "heading": "Human inheritance and disease-associated variants",
      "text": "In human genetics, dominant and recessive usually describe patterns associated with particular variants and clinical phenotypes, rather than permanent properties of whole genes. Different variants in one gene can cause different conditions, sometimes through different inheritance modes. A variant that destroys protein function may behave recessively if one functional copy is sufficient, whereas an activating variant in the same gene may act dominantly. Accurate interpretation therefore depends on the specific molecular change and the phenotype being considered.\n\nCystic fibrosis is a well-established autosomal recessive condition associated with disease-causing variants in both copies of CFTR, which encodes an epithelial anion channel. Affected individuals can carry two identical variants or two different ones. By contrast, Huntington disease is an autosomal dominant disorder associated with an expanded CAG repeat in HTT. Its age-dependent manifestation illustrates why an individual’s phenotype at one age does not necessarily reveal the eventual expression of a genotype.\n\nFamilial hypercholesterolemia illustrates the limitations of rigid binary categories. Many cases involve variants affecting LDLR, and heterozygotes can have markedly elevated low-density lipoprotein cholesterol, while individuals with two severely impairing alleles commonly have a more severe phenotype. Such relationships may be described as semidominant or codominant in some contexts because gene dosage affects severity, even though a clinically recognized phenotype can occur with one altered copy. Other genes can also cause familial hypercholesterolemia, so the condition is not reducible to a single allelic relationship.\n\nCancer-predisposition syndromes provide another distinction between organism-level inheritance and cell-level mechanism. A pathogenic germline variant in a tumor-suppressor gene can confer an autosomal dominant predisposition, even when a tumor cell generally requires loss or inactivation of the remaining functional copy. Alfred Knudson’s 1971 analysis of retinoblastoma helped establish the two-hit framework. The inherited phenotype is increased susceptibility; the cellular transformation process can require additional somatic changes. This is not a contradiction, because the two descriptions concern different biological levels.\n\nMedical interpretation also separates transmission probability, penetrance, severity, and age of onset. A one-half probability of transmitting a heterozygous autosomal variant is not automatically a one-half probability of a particular disease outcome. Some dominant conditions arise through new mutations in individuals whose parents do not carry the variant in the tissues tested; parental germline mosaicism can further complicate family patterns. These observations make pedigrees informative but not sufficient on their own to establish variant function, inheritance mechanism, or the expected expression of a condition."
    },
    {
      "heading": "Experimental identification, breeding, and genomic analysis",
      "text": "Dominance is established by comparing appropriately characterized genotypes. Classical experiments use controlled crosses, inbred lines, and repeated phenotypic measurements. To attribute a difference to one locus, researchers must reduce or account for variation elsewhere in the genome. Near-isogenic lines, segregating families, and reciprocal crosses help distinguish a focal allele’s effect from genetic background, maternal effects, or parent-of-origin effects. Environmental replication is important when temperature, nutrition, or developmental conditions alter the measured phenotype.\n\nMolecular methods can test the proposed mechanism rather than only classify the inheritance pattern. DNA sequencing identifies candidate variants; measurements of RNA, protein abundance, localization, and enzyme activity characterize their consequences. Introducing a functional copy can test whether additional normal product rescues a phenotype. Comparing a deletion with a missense variant can help distinguish simple loss of function from interference by an abnormal protein. No single rescue experiment is conclusive in every setting, because transgene dosage and expression patterns may differ from those of the native locus.\n\nGenome editing allows researchers to construct matched \\(AA\\), \\(Aa\\), and \\(aa\\) genotypes and compare them in a common background. In cultured cells, such experiments can be especially useful for separating dosage effects from dominant-negative or activating mechanisms. Their interpretation still depends on the biological model: a phenotype in an immortalized cell line may not reproduce a developmental process involving multiple tissues, and an artificial expression system may produce protein concentrations that change complex assembly or toxicity.\n\nIn animal and plant breeding, dominance contributes to the performance of crosses but does not make selection straightforward. A favorable dominant phenotype can conceal whether an individual is homozygous or heterozygous. Progeny testing and molecular markers help distinguish those cases. Heterosis, or superior performance of a hybrid relative to an appropriate parental comparison, can involve masking of deleterious recessive alleles, locus-specific heterozygote advantage, and interactions among loci. Apparent overdominance can also arise from linked loci carrying favorable and unfavorable alleles in different combinations, a situation often called pseudo-overdominance.\n\nLarge genomic datasets extend these approaches to many loci. A genome-wide association study may initially use an additive dosage model because it is statistically efficient and often captures substantial variation, not because complete additivity is a biological certainty. Dominance testing requires adequate numbers of heterozygotes and both homozygotes, making rare variants particularly difficult to evaluate. Statistical associations can also be influenced by linkage, population structure, and phenotype definition. Functional validation, careful genotype grouping, and replication across relevant biological conditions are therefore central to determining whether an observed nonadditive pattern represents a genuine dominance relationship."
    }
  ],
  "links": [
    "allele",
    "phenotype",
    "gene",
    "genetics",
    "evolution",
    "enzyme",
    "mutation",
    "meiosis",
    "bacteria",
    "gregor-mendel",
    "genotype",
    "dna",
    "chromosome",
    "incomplete-dominance",
    "abo-blood-group-system",
    "protein",
    "dominant-negative-mutation",
    "promoter-biology",
    "rna-splicing",
    "penetrance",
    "epistasis",
    "genomic-imprinting",
    "x-chromosome-inactivation",
    "x-chromosome",
    "hardy-weinberg-equilibrium",
    "variance",
    "quantitative-genetics",
    "heritability",
    "natural-selection",
    "inbreeding",
    "genetic-drift",
    "mosaicism",
    "dna-sequencing",
    "genome-editing",
    "heterosis",
    "genome-wide-association-study"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}