A gene is a stretch of DNA that carries instructions for building a specific protein or functional RNA molecule, while an allele is one particular version of that gene. Think of it this way: “eye color” is the gene (the category of instruction), but “brown” and “blue” are alleles (specific variants of that instruction). Every person inherits two copies of most genes, one from each parent, and those two copies can be the same allele or different alleles. The distinction matters because it is the alleles you carry, not just the genes themselves, that determine what your body actually does with those instructions.
What a Gene Is and What It Does
At the simplest level, a gene is a unit of hereditary information encoded in DNA. Most genes provide the blueprint for a protein, though some produce functional RNA molecules instead. In humans, roughly 20,000 protein-coding genes are scattered across 23 pairs of chromosomes. Each gene occupies a fixed address on a chromosome, called a locus, so when geneticists talk about “the gene for” something, they mean the DNA at that particular location.
The reality of what counts as a gene is messier than the textbook version. In complex organisms, the DNA sequence that ultimately encodes a protein is often split into pieces separated by non-coding stretches. The cell stitches together the final instruction set only after the DNA has been copied into an RNA transcript, and this stitching can happen in different ways, meaning a single gene can produce more than one protein. One analysis of this process argued that the gene, in a strict biochemical sense, does not fully exist at the DNA level in many cases but instead emerges as a complete sequence only at the messenger-RNA stage, just before the cell reads it to build a protein.1Europe PMC. Gene and genon concept: coding versus regulation That is a fairly radical framing, but it underscores an important point: genes are not neat, self-contained beads on a string the way early geneticists imagined.
How an Allele Differs from a Gene
If a gene is the location and general category of instruction, an allele is the specific spelling at that location. DNA is written in a four-letter chemical alphabet, and for any given gene, most of those letters are identical from person to person. But at certain positions, the letters can differ. A change of even one letter can alter the protein the gene produces, or change how much of that protein the cell makes, or sometimes do nothing at all. Each distinct version created by those differences is a separate allele.
Because humans carry two copies of each chromosome (one inherited from each parent), you have two alleles for every gene. If both copies are the same allele, you are homozygous at that locus. If they differ, you are heterozygous. The combination of alleles you carry is what ultimately shapes traits ranging from blood type to susceptibility to certain diseases.
Some genes have only two known alleles in the human population. Others have dozens or even hundreds. The ABO blood-type gene, for instance, has three common alleles that combine to produce the familiar A, B, AB, and O blood types. Immune-system genes involved in tissue compatibility can have thousands of alleles circulating worldwide, which is one reason finding a matching organ donor can be so difficult.
Where New Alleles Come From
Every allele started as a mutation, a change in the DNA sequence that was then passed to the next generation. Some mutations swap one letter for another. Others insert or delete letters, or rearrange larger chunks of DNA. A study of a particularly unstable region of human DNA found that new allele lengths arise primarily through changes in repeat copy number, sometimes involving the transfer of repeated units between the two copies of a chromosome.2PubMed. Mutation rate heterogeneity and the generation of allele diversity at the human minisatellite MS205 (D16S309) At this particular locus, the rearrangements were complex, not just simple one-letter swaps, and they occurred almost exclusively in reproductive cells rather than in ordinary body cells.
Larger-scale structural changes also contribute to allelic diversity. Segments of DNA can be duplicated or deleted entirely, creating what are called copy number variants. These are common features of the human genome, playing a role in evolution and population diversity, and sometimes contributing to disease.3PubMed Central. DNA copy number variation: Main characteristics, evolutionary significance, and pathological aspects So alleles are not always minor spelling differences. Some represent substantial restructuring of the DNA at a particular locus.
Natural selection, genetic drift, and migration then determine which alleles become common and which stay rare. An allele that confers a survival advantage in a certain environment tends to increase in frequency over generations. One that causes a severe disease in childhood tends to stay rare, unless it also provides a benefit in some other context. The sickle-cell allele is the classic example: carrying one copy offers some protection against malaria, which has kept the allele relatively common in regions where malaria is endemic, even though carrying two copies causes serious illness.
Dominant and Recessive Alleles
When you carry two different alleles for the same gene, one often has more influence on the visible trait than the other. The allele whose effect shows up is called dominant; the one whose effect is hidden (unless present in two copies) is called recessive. But that simple framing conceals a lot of biological complexity. Dominance is not a fixed property of an allele itself. It describes a relationship between two alleles, and the mechanisms that produce dominance vary enormously.
A review of how dominance works at the molecular level identified at least eight distinct categories, including situations where one allele simply does not make enough protein, situations where the mutant protein actively interferes with the normal one, and situations where a mutation creates an entirely new function that the normal allele never had.4PubMed Central. The molecular basis of genetic dominance So “dominant” does not mean “stronger” or “better” in any universal sense. A dominant allele for a genetic disease is dominant because of the specific way its protein product disrupts cell function, not because dominance signals some inherent superiority.
For proteins that work by assembling into multi-part complexes, the interaction between the products of two different alleles can be particularly telling. When both copies of a gene contribute subunits to the same complex, one allele’s product can drag down or alter the function of the other’s. Research on this kind of molecular dominance has shown that the physical interactions between the protein products of different alleles are themselves a key mechanism for determining which allele’s effect wins out.5PubMed Central. Phenotypic dominance emerges from activity fitness functions and molecular interactions
A striking example of dominance in action comes from butterfly wing-pattern research. In one species, a cluster of tightly linked genes controls which wing pattern an individual displays. When researchers compared gene activity in butterflies carrying two different alleles at this cluster, the expression profile in the heterozygote closely resembled that of the dominant homozygote across the region, while transcript expression outside that region showed no such bias.6PubMed Central. Dominance mechanisms in supergene alleles controlling butterfly wing pattern variation Dominance, in other words, was being enforced at the level of how much RNA was produced from each allele, and the mechanism differed across sub-regions of the gene cluster.
Alleles Outside Protein-Coding Genes
The conversation about genes and alleles often centers on protein-coding sequences, but the vast majority of your DNA does not code for proteins. Non-coding DNA includes sequences that get transcribed into regulatory RNA molecules and untranscribed stretches that control when, where, and how much a nearby gene is turned on.7PubMed Central. Non-coding regulatory elements: Potential roles in disease and the case of epilepsy These regulatory regions can also carry allelic variation, and those variants can have real biological consequences even though they do not change a protein’s structure.
A concrete example comes from the gene ACE2, which encodes a protein involved in blood-pressure regulation that also happens to be the entry point for certain viruses. Researchers catalogued variants in the non-coding regions surrounding ACE2 and found that specific single-letter changes in its promoter region, the DNA stretch that controls how actively the gene is read, significantly boosted promoter activity in laboratory tests. That means these variants could increase the amount of ACE2 protein a cell produces, even though they sit outside the gene’s protein-coding sequence entirely.8PubMed. Exploring non-coding genetic variability in ACE2: Functional annotation and in vitro validation of regulatory variants
This matters because it expands the scope of what “allele” means in practice. When a genetics report flags a variant, it might be in a coding region and directly alter a protein, or it might be in a regulatory region and change how much of a perfectly normal protein gets made. Both are allelic variants. Both can affect health. But they operate through different mechanisms, and a reader who equates “allele” only with “different protein” will miss an important piece of the picture.
Why the Distinction Matters in Medicine
In clinical genetics, the difference between gene and allele is not just academic vocabulary. Knowing which gene is involved tells a doctor what biological pathway is disrupted. Knowing which specific alleles a patient carries often determines how severe the disease will be and which treatments are likely to work.
Cystic fibrosis illustrates this clearly. The condition is caused by mutations in a single gene, CFTR, which encodes a chloride channel in cell membranes. But more than 2,000 different alleles of CFTR have been identified, and they do not all produce the same severity of disease. Clinical studies have grouped these alleles into classes based on how badly they impair the protein’s function. Patients carrying two copies of the most severely disruptive alleles (classes I through III) tend to have worse outcomes, with more liver involvement, lower sodium levels, and poorer growth, compared to patients whose allele combinations include at least one milder variant.9PubMed Central. Diagnosis of cystic fibrosis: a high heterogeneity of symptoms and genotypes in a Brazil population Separate research has confirmed that patients with “mild” genotypes generally experience a more favorable disease course than those with “severe” genotypes.10Zaporozhye Medical Journal. Heterogeneity of phenotypic manifestations of cystic fibrosis in children and predictors of the disease severity
This is why genetic testing in conditions like cystic fibrosis goes beyond simply confirming which gene is affected. The specific allele combination guides prognosis and, increasingly, treatment selection. Newer drugs called CFTR modulators are designed to work on specific allele classes, so two patients with the same disease, caused by the same gene, may receive entirely different medications depending on which alleles they carry.
When the Same Allele Behaves Differently
One complication that trips people up is that carrying a particular allele does not always produce the same outcome. Epigenetic modifications, chemical tags added to DNA or its packaging proteins, can dial a gene’s activity up or down without changing the underlying allele at all. Epigenetic information is plastic during development but can be stably maintained across cell divisions afterward, creating a layer of regulation that sits on top of the genetic sequence.11PubMed Central. Allele-specific gene expression and epigenetic modifications and their application to understanding inheritance and cancer In some cases, only one of your two alleles is active in a given tissue, not because the other is defective but because the cell has chemically silenced it.
This allele-specific silencing means two people carrying the exact same pair of alleles could differ in which allele is actually expressed. Genomic imprinting, where the copy inherited from one parent is silenced while the copy from the other stays active, is a well-known example. The allele itself has not changed, but its behavior in the cell has.
More intriguingly, the genetic sequence at one locus can influence epigenetic modification at another. Research on uterine fibroids found that specific mutations in the MED12 gene were associated with increased DNA methylation levels in fibroids, with mutated fibroids showing on average a 4.6-unit increase in methylation compared to non-mutated ones.12PubMed Central. Genetic–Epigenetic Interactions in Uterine Leiomyomas: MED12 Mutations as Predictors of Aberrant DNA Methylation So carrying a particular allele of one gene can reshape the epigenetic landscape around other genes. Genetics and epigenetics do not operate in separate worlds; they interact, and the alleles you carry influence how that interaction plays out.
Sex Chromosomes and the One-Allele Situation
Most of this discussion assumes you carry two alleles for each gene, but that is not always the case. In humans, biological males have one X chromosome and one Y chromosome, which means genes located on the X chromosome are present in only a single copy. There is no second allele to compensate if the one you have is faulty. This is why X-linked conditions like red-green color blindness and hemophilia are far more common in males: a single recessive allele on the X is enough to produce the trait because there is no matching allele on the Y to mask it.
The Y chromosome itself carries relatively few genes, and those it does carry are present in only one copy (hemizygous). Forensic geneticists use short repeating sequences on the Y chromosome as identification markers, taking advantage of the fact that each male has only one allele at each Y-linked locus rather than the usual two.13PubMed Central. Allele frequency distribution and haplotype of eleven hemizygous short tandem repeats in Jordanians Those markers pass from father to son essentially unchanged (barring rare mutations), making them useful for tracing paternal lineage.
Organisms with different chromosome arrangements complicate things further. Many plant species are polyploid, carrying three, four, or even six copies of each chromosome. In those species, a single gene might have four or more alleles within one individual, and the interactions among them create far more complex patterns of inheritance than the two-allele system familiar from human genetics.
Gene Editing That Targets a Single Allele
The gene-versus-allele distinction has become directly relevant to a new generation of medical therapies. CRISPR-based gene editing can, in principle, be precise enough to distinguish between two alleles that differ by just a single DNA letter. This opens the door to selectively disabling a disease-causing allele while leaving the healthy copy untouched.14PubMed Central. Allele-specific genome targeting in the development of precision medicine
For diseases caused by a dominant allele, this approach is especially appealing. If one allele is producing a toxic or malfunctioning protein, you do not need to fix the mutation. You just need to silence that one allele. Researchers have demonstrated this strategy can achieve complete allele discrimination by designing the CRISPR guide to recognize a natural variant on the chromosome carrying the disease mutation, essentially using a harmless genetic difference as an address label to direct the edit to the right copy.15Molecular Therapy. Mutation-Independent Allele-Specific CRISPR Gene Editing for Autosomal Dominant Disease
The precision is impressive but imperfect. In one study targeting a collagen gene variant that causes a form of muscular dystrophy, the guide RNAs successfully introduced edits in roughly 20 to 38 percent of reads from the disease allele, but one of the two guides also edited about 21 percent of reads from the normal allele in patient cells.16Molecular Therapy – Nucleic Acids. Allele-specific CRISPR-Cas9 editing inactivates a single nucleotide variant associated with collagen VI muscular dystrophy Off-target editing of the healthy allele is the central safety concern with this approach. A second guide RNA tested in the same study performed better, with only about 2 percent off-target activity on the normal copy, demonstrating that careful guide design can dramatically improve specificity.
The broader point is that these therapies only make sense because genes and alleles are not the same thing. If the two copies of a gene were always identical, there would be nothing to selectively target. The entire strategy depends on the fact that alleles differ, even slightly, and those differences can be exploited to treat disease at the molecular level.
Where the Words Themselves Come From
The terms “gene” and “allele” were coined during the early twentieth century, when the basic principles of inheritance were just being rediscovered. William Bateson, one of the most energetic promoters of Gregor Mendel’s rediscovered work, coined the word “genetics” to describe the new field and introduced the term “allelomorph” to describe paired trait variants, which was later shortened to “allele.”17Comptes Rendus Biologies. From Mendel to epigenetics: History of genetics The vocabulary Bateson and his contemporaries developed, including “homozygote” and “heterozygote,” baked in the assumption that each individual carries exactly two copies of each hereditary element, an idea Mendel himself had not explicitly articulated.
The word “gene” arrived slightly later, coined in 1909 by the Danish botanist Wilhelm Johannsen, who wanted a term for the abstract unit of heredity without committing to any particular physical model of what it was made of. That deliberate vagueness turned out to be useful. As molecular biology advanced, the physical reality of genes turned out to be far more complex than anyone in 1909 could have imagined, but the term was flexible enough to survive. “Allele” has proven equally durable, though today it covers a vastly wider range of variation than Bateson’s original paired trait differences in pea plants and poultry.