Detecting Selective Sweeps in Genomic Data

112 concepts85 papers1 proposal

When a beneficial genetic mutation arises in a population, evolutionary forces do not let it drift quietly. Instead, natural or artificial selection drives the advantageous allele rapidly toward fixation—the point at which every individual in the population carries it.

However, genes are not inherited in isolation; they are bound together on chromosomes. Because a strong selective event forces an allele to spread through a population far faster than meiotic recombination can break up the surrounding DNA, the favored mutation drags its neighboring neutral variants along with it. This phenomenon, known as genetic hitchhiking, leaves a distinct, localized "scar" across the genome.

By developing statistical tools to detect these selective sweeps, researchers can look backward in time to identify exactly which genomic regions have undergone intense evolutionary pressure, mapping the genetic architecture of survival, adaptation, and domestication.

The Three Signals of a Sweep

To pinpoint where a selective sweep has occurred, population geneticists scan genomic data for three primary signatures left behind by genetic hitchhiking: the loss of local diversity, extreme population divergence, and extended haplotype blocks.

1. Localized Reductions in Genetic Diversity

Before a selective sweep, a given genomic region contains a diverse mosaic of variations across different individuals. When a sweeping allele takes over, it replaces that mosaic with a uniform block of DNA. This creates a sharp, localized drop in nucleotide diversity ($\pi$) centered precisely on the target gene.

As demonstrated by Annegret Werzner et al. in Selective Sweep in the Flotillin-2 Region of European Drosophila melanogaster, analyzing deviations in the site frequency spectrum allows researchers to isolate these localized diversity drops from background chromosomal averages. In agricultural contexts, this same signature reveals the history of human intervention. For example, Xiang Luo et al. mapped these diversity valleys in Domestication and selection footprints in Persian walnuts (Juglans regia) to identify the specific transcription factors early farmers selected to thin out the internal diaphragms of walnuts, making the kernels easier to extract.

2. Extreme Population Differentiation ($F_{ST}$)

If a selective pressure acts intensely on one population but is absent in another, the allele frequencies at the target locus will diverge rapidly between the two groups. Geneticists quantify this divergence using the fixation index ($F_{ST}$), which measures the proportion of genetic variance attributable to population structure.

While random genetic drift causes minor, unpredictable fluctuations in $F_{ST}$ across the entire genome, a selective sweep forces a dramatic spike in differentiation at a single locus. When a genomic window exhibits an $F_{ST}$ value that towers over the neutral background, it is flagged as an outlier. Derong Hao et al. applied this outlier detection strategy in Identification of Genetic Differentiation between Waxy and Common Maize by SNP Genotyping to successfully isolate 14 distinct genomic hotspots that underwent intense positive selection during the divergence of specialized waxy maize from common varieties.

3. Extended Haplotype Homozygosity (EHH)

Over long evolutionary periods, recombination acts like a blender, breaking up combinations of alleles (haplotypes) on a chromosome. However, because a selective sweep happens so rapidly, the favored allele reaches high frequency before recombination has a chance to disrupt its genetic neighborhood.

This creates a signature of Extended Haplotype Homozygosity (EHH): a long, unbroken, highly uniform stretch of DNA shared by many individuals in a population. Statistical metrics like the Integrated Haplotype Score (iHS) measure the decay of linkage disequilibrium to capture these ongoing or recent sweeps. In Genetic Diversity and Selection Signatures Within Diannan Small-Ear Pigs Revealed by Next-Generation Sequencing, Fen Wu et al. leveraged iHS to locate clusters of homozygous segments—known as "Runs of Homozygosity" (ROH) islands—pinpointing genomic regions actively favored by breeders for meat quality and environmental resilience.

Distinguishing Selection from Demographic Noise

The central challenge in selective sweep detection is that neutral demographic history can mimic the signatures of selection. A severe population bottleneck or a founder effect can purge genetic diversity and alter allele frequencies across an entire population, generating thousands of false positives.

The key to overcoming this obstacle lies in scale and distribution: demographic events affect the entire genome globally, whereas selective sweeps affect specific loci locally. Modern frameworks isolate true adaptive signals by establishing robust genomic baselines that account for population structure.

For instance, single-author work by Guo‐Bo Chen in EigenGWAS: finding loci under selection through genome-wide association studies of eigenvectors in structured populations introduced a method to adjust association statistics using the largest eigenvalues of a genetic relationship matrix. By normalizing statistics against the background genomic inflation factor, this approach separates the signals of localized selective pressure from random genetic drift and ancestral stratification.

Similarly, Jing Guo demonstrated in Global genetic differentiation of complex traits shaped by natural selection in humans that natural selection leaves a distinct signature on the variation of linkage disequilibrium (LD) scores across populations. Measuring the coefficient of variation for these LD scores provides a robust method for detecting selection that remains independent of simple allele frequency shifts caused by demographic history.

The Power of Multi-Metric Scans

Because different statistical tests capture different evolutionary timescales and facets of a sweep, relying on a single metric can leave blind spots. To maximize accuracy, contemporary studies deploy multi-metric pipelines that look for the intersection of independent signals.

[   Genomic Window Scan   ]
         │
         ├──► Metric 1: High Population Differentiation (F_ST Outlier)
         └──► Metric 2: Severe Drop in Local Diversity (Low π)
                 │
                 ▼
     [ Intersecting Thresholds ] ──► Confirmed Selective Sweep

By requiring a genomic window to pass multiple filters simultaneously—such as exhibiting both an extreme $F_{ST}$ outlier spike and a severe reduction in nucleotide diversity ($\pi$)—researchers can drastically reduce false-positive rates. Zhuoxian Weng et al. utilized this dual-threshold approach in Genomic variations and signatures of selection in Wuhua yellow chicken to reliably isolate genes governing immune response and plumage coloration.

When applied at a genome-wide scale, combining haplotype-based and frequency-based metrics provides a comprehensive window into complex adaptations. In Genome-Wide Analysis of Cold Adaptation in Indigenous Siberian Populations, Alexia Cardona et al. combined iHS, Cross-Population Extended Haplotype Homozygosity (XP-EHH), and the Population Branch Statistic (PBS). This multi-pronged strategy allowed the authors to capture selection signals across different evolutionary layers, successfully mapping the metabolic and vascular adaptations that enable indigenous populations to survive extreme arctic environments.

History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 86 papers
  2. Version 1 · 2026-10-04 · 61 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_selective_sweep_detection_c9c40c59,
  title = {selective sweep detection},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/selective-sweep-detection/rdr_c9c40c5916a84455a9f44dd1959d3d62},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 2, frozen text):

@misc{lacuna_selective_sweep_detection_c9c40c59_v2,
  title = {selective sweep detection},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/selective-sweep-detection/rdr_c9c40c5916a84455a9f44dd1959d3d62/v/2},
  note = {Version 2, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake