Population Genetics (Evolutionary Biology)

Genetic Drift

A mechanism of microevolution involving random fluctuations in allele frequencies across generations due to chance, arising from random sampling of gametes in finite populations. Over time, drift can lead to the fixation or loss of alleles irrespective of their fitness effects.

Admixture

Gene flow resulting from interbreeding between two or more previously isolated populations, producing offspring with mixed ancestry. Over generations, this creates a population with a mosaic of genomic blocks inherited from distinct source populations.

Single Nucleotide Polymorphism (SNP)

A single nucleotide base-pair substitution (e.g., A→G) that occurs at a specific position in the genome and is present in at least 1% of a population. SNPs are the most common type of genetic variation and are widely used as molecular markers in association studies, population structure inference, and detecting signatures of natural selection.

Panmixia

A population state in which all individuals have an equal probability of mating with one another, irrespective of their genotype, phenotype, geographic location, or population of origin. In a panmictic population, there is no population subdivision or assortative mating, and allele frequencies are expected to follow Hardy–Weinberg equilibrium.

Principal Component Analysis (PCA)

A dimensionality-reduction technique that transforms large sets of correlated SNP genotypes into uncorrelated principal components (PCs), which capture the major axes of genetic variation in the data. When plotted (e.g., PC1 vs PC2), individuals with similar ancestry cluster together, revealing population structure, admixture patterns, and outliers.

Hardy-Weinberg Equilibrium (HWE)

A principle stating that in the absence of evolutionary forces (random mating, infinite/large population size, no selection, no mutation, no gene flow), allele frequencies (p and q) remain constant across generations, and genotype frequencies reach and remain at p²+2pq+q² =1 after a single generation of random mating.

Heterozygosity

A measure of genetic diversity at a locus, defined as the proportion of individuals in a population that carry two different alleles. Observed (H_O​) is directly counted from genotype data, while expected heterozygosity (H_E=1−∑p_i²​) is calculated from allele frequencies under HWE.

Acclimation

A reversible, non-heritable physiological adjustment by an individual organism in response to a change in its environment (e.g., temperature, altitude). It occurs within an organism's lifetime and does not involve genetic change.

Wright-Fisher Model

A discrete-generation mathematical model of allele frequency change in a finite population of constant size N, where each generation is formed by random sampling of 2N gametes from the previous generation. It serves as the foundational model for studying genetic drift and, when combined with mutation, predicts the equilibrium between mutation and drift (e.g., neutral theory).

Assortative Mating

Non-random mating in which individuals mate based on phenotypic similarity or dissimilarity. It can increase homozygosity and reduce heterozygosity at loci influencing the trait if individuals mate with individuals with similar phenotypes to them. It can increase heterozygosity at those loci, instead, if individuals mate with individuals with dissimilar phenotypes to them. Unlike inbreeding, it does not affect allele frequencies—only genotype frequencies.

Inbreeding

Mating between individuals that are more closely related than average members of the population. Inbreeding increases genome-wide homozygosity and decreases heterozygosity because offspring are more likely to inherit alleles that are identical by descent (IBD) from a common ancestor. This results in a deficit of heterozygotes relative to Hardy-Weinberg expectations.

Outbreeding

Mating between individuals that are less closely related than the average pair in the population (or between unrelated individuals). It increases heterozygosity across the genome and can lead to heterosis (hybrid vigor) when beneficial dominant alleles are masked from recessive deleterious ones.

Effective Population Size (Ne)

The size of an idealized, Wright-Fisher population (constant size, random mating, equal sex ratio) that would experience the same rate of genetic drift (or increase in inbreeding) as the actual census population. N_e​ is typically much smaller than the census population size due to demographic factors.

Pairwise Sequentially Markovian Coalescent (PSMC)

A coalescent-based method that uses the genome-wide distribution of heterozygous sites in a single diploid genome to infer historical changes in effective population size (N_e​) over time. It models the sequential coalescence of the two haplotypes along the chromosome using a Hidden Markov Model, with longer homozygous blocks reflecting more recent coalescence (and thus smaller N_e​ in the recent past).

Linkage Equilibrium

A state in which alleles at different loci are inherited independently of one another, such that the frequency of a haplotype is equal to the product of the frequencies of its constituent alleles (e.g., f_AB=f_A×f_B​). This is the null expectation under random recombination and no epistatic selection.

Linkage Disequilibrium

The non-random association of alleles at different loci, such that certain combinations of alleles (haplotypes) occur more or less frequently than expected by chance. While often caused by physical proximity on a chromosome (reduced recombination), it can also arise from population admixture, natural selection, or genetic drift.

Haplotypes

A specific combination of alleles at multiple linked loci on a single chromosome that are inherited together as a block from one parent. They can range in size from a few SNPs to entire chromosomal regions, but are progressively broken down by recombination over generations.

D

A measure of linkage disequilibrium defined as the difference between the observed frequency of a haplotype (AB) and its expected frequency under linkage equilibrium: P(AB)−P(A)P(B). A value of 0 indicates linkage equilibrium, while positive or negative values indicate an excess or deficit of that haplotype. However, it is frequency-dependent, so standardized measures are often preferred for comparisons.

r²

A standardized measure of linkage disequilibrium defined as the squared correlation coefficient between alleles at two loci: D²/ P(A)P(a)P(B)P(b)​. It ranges from 0 (linkage equilibrium) to 1 (perfect LD). It is sensitive to both recombination and allele frequency changes (drift/mutation), making it the preferred measure for GWAS power calculations.

Wahlund Effect

A reduction in observed heterozygosity (and corresponding excess of homozygotes) in a pooled population that is actually composed of two or more genetically distinct subpopulations, each of which may be in Hardy-Weinberg equilibrium. The deficit arises because allele frequencies differ between subpopulations, so when pooled, the expected heterozygosity under HWE is overestimated.

Hitchhiking Effect

The process by which a beneficial mutation rapidly increases to fixation under positive selection, carrying with it physically linked neutral or slightly deleterious alleles. This reduces genetic diversity (heterozygosity) in the surrounding chromosomal region, creating a characteristic 'trough' or 'valley' of low polymorphism—a pattern known as a selective sweep.

Epistatic Selection

Natural selection that acts on the fitness effects of interactions between alleles at different loci. The fitness contribution of an allele at one locus depends on the genotype present at another locus (i.e., non-additive fitness effects). This can maintain genetic polymorphism (e.g., through reciprocal selection) or drive the fixation of favorable multi-locus genotypes.

Runs of Homozygosity (ROH)

Continuous stretches of the genome (typically >1 Mb) in which an individual is homozygous at all genotyped sites. These regions indicate that the individual inherited the same haplotype from both parents from a common ancestor (identical by descent, IBD). The length of them is informative: long ones suggest recent inbreeding, while short ones reflect more distant shared ancestry.

Bottlenecked Population

A population that has experienced a sharp, temporary reduction in population size (e.g., due to a disaster, disease, or colonization). This causes a random loss of genetic diversity, particularly rare alleles, and a disproportionate reduction in heterozygosity. Even after the population recovers in size, the genetic signature of this persists for many generations.

Sub-Population

A smaller, partially isolated group of individuals (also called a deme) within a larger metapopulation. They experience genetic drift due to finite size, but are connected to others by low to moderate levels of gene flow (migration).

FST

The Fixation index, a measure of genetic differentiation that quantifies the proportion of total genetic diversity (heterozygosity) attributable to differences between sub-populations, rather than within them. It ranges from 0 (no differentiation; complete gene flow) to 1 (complete differentiation; no gene flow).

Stepping Stone Model

A model of population structure in which gene flow (migration) occurs only between adjacent or geographically neighboring sub-populations. This leads to an isolation-by-distance pattern, where genetic differentiation increases with geographic distance.

Island Model

A simplified model of population structure in which all sub-populations exchange migrants at an equal rate (m) with one another, irrespective of geographic distance. This model serves as a theoretical null model for studying the balance between genetic drift and gene flow.

Coalescent Theory

A retrospective mathematical model that traces the ancestry of a sample of alleles (or individuals) backwards in time to their Most Recent Common Ancestor (MRCA). It assumes neutral evolution and is widely used to infer demographic history (e.g., population size changes, divergence times) from genetic data.

Coalescent Event

A point in the genealogical history of a sample when two distinct ancestral lineages merge into a single common ancestral allele as one traces alleles backwards in time. Each one reduces the number of distinct lineages by one.

μ

The mutation rate, typically expressed as the number of new mutations per nucleotide site per generation. It quantifies the rate at which new genetic variation enters a population and is a fundamental parameter in neutral theory, where it determines the equilibrium between mutation and genetic drift.

Θ

The population mutation rate, a scaled parameter that quantifies the product of the effective population size (N_e​) and the per-generation mutation rate (μ). For a diploid population, it is equal to 4N_eμ (or 2N_eμ for haploids). It determines the expected level of neutral genetic diversity (heterozygosity) under mutation-drift equilibrium and is a central parameter in coalescent theory.

Directional Selection

A mode of natural selection in which one extreme phenotype is favored over all others, causing the mean value of a trait to shift consistently in one direction over generations. At the molecular level, this is equivalent to positive selection, where beneficial mutations increase in frequency and can sweep to fixation, reducing genetic diversity at linked sites.

Stabilising Selection

A mode of natural selection that favors intermediate phenotypes and selects against extremes, thereby reducing phenotypic variance while leaving the mean trait value unchanged. At the molecular level, this is equivalent to purifying (negative) selection, which removes deleterious mutations and maintains conserved functional sequences.

Disruptive Selection

A mode of natural selection that favors two or more extreme phenotypes over intermediate forms, often creating a bimodal distribution of traits. This can lead to phenotypic dimorphism and, if mating becomes assortative between the extremes, may drive sympatric speciation.

Positive Selection

A mode of natural selection in which advantageous alleles (whether new mutations or standing variation) increase in frequency and are driven toward fixation due to their fitness benefit. At the molecular level, this is often detected by an excess of nonsynonymous substitutions relative to synonymous changes.

Negative (purifying) Selection

A mode of natural selection that removes deleterious alleles from a population, thereby maintaining functionally important sequences and preventing their degradation. It is the most common form of selection acting on genomes, and its effects are strongest at conserved sites.

CADD Score

Combined Annotation Dependent Depletion — a computational tool that integrates diverse genomic annotations (e.g., conservation, regulatory potential, protein structure) to predict the deleteriousness or functional impact of a genetic variant. It outputs a Phred-like score: scores >10 indicate variants in the top 10% most deleterious, >20 in the top 1%, and >30 in the top 0.1%.

McDonald-Kreitman Test (MK test)

A comparative test for positive selection that contrasts the ratio of nonsynonymous to synonymous substitutions within species (polymorphisms) against the same ratio between species (divergence). Under neutral evolution, the two ratios are expected to be equal. An excess of nonsynonymous divergence suggests positive selection, while an excess of nonsynonymous polymorphisms suggests purifying selection.

α

It is the proportion of nonsynonymous substitutions between species that have been driven to fixation by positive selection, as estimated from the MK test. It is calculated as:

α=1−[(D_S x P_N) / (D_N x P_S)​​]

where D_N​ and D_S​ are nonsynonymous and synonymous divergence, and P_N and P_S are nonsynonymous and synonymous polymorphisms. Values close to 1 indicate that most amino-acid changes are adaptive; values near 0 suggest they are largely neutral.

Hard Sweep

A selective sweep resulting from a single new advantageous mutation that arises in a population and rapidly rises to fixation under positive selection. This drastically reduces genetic diversity (heterozygosity) in the surrounding linked genomic region, leaving a characteristic long haplotype block with low polymorphism.

Soft Sweep

A selective sweep where multiple haplotypes containing the favored variant increase in frequency, either because selection acted on standing genetic variation (pre-existing variants) or because the same beneficial mutation occurred independently on multiple genetic backgrounds. They leave a weaker genomic signature than hard sweeps—diversity is reduced but not as severely, and haplotype blocks are shorter.

Site Frequency Spectrum (SFS)

It is a summary statistic that describes the distribution of allele frequencies of polymorphisms within a sample of individuals. It is typically represented as a histogram or vector showing the number of variants at each frequency class. It can be folded (ignoring ancestral/derived states) or unfolded (using an outgroup to identify derived alleles) and is the basis for many neutrality tests.

Polygenic Adaptation

A mode of adaptation where selection acts simultaneously on many loci of small effect, causing subtle allele frequency shifts across the genome rather than strong sweeps at individual loci. This results in 'partial sweeps' where beneficial alleles increase in frequency but rarely reach fixation. It is common for complex quantitative traits and is more difficult to detect than classic hard sweeps.

Background Selection

The process by which purifying selection against deleterious mutations reduces the effective population size (N_e​) at physically linked neutral or weakly selected sites. As deleterious mutations are removed from the population, linked neutral variants are 'dragged' to extinction, reducing genetic diversity (heterozygosity) in those genomic regions—a pattern that can superficially resemble a selective sweep.

Population Branch Statistics (PBS)

A three-population F_ST​-based statistic that identifies loci where a specific lineage (population) has undergone accelerated evolution relative to two reference populations. It is calculated from pairwise F_ST values. High values indicate that a locus has experienced strong positive selection in the target population. It is particularly useful for detecting local adaptation in human populations.