Population Genetics II: Linkage, Recombination, and the Genomic Landscape of Selection

Manual: General · Subject: Evolutionary Biology

This lesson connects single-locus theory to multi-locus dynamics, linkage disequilibrium, selective sweeps, background selection, and genomic prediction.

Beyond independent loci

Why linkage matters

When loci are linked, selection at one site alters nearby neutral variation through hitchhiking and background selection. Recombination decouples loci and determines how local genomic architecture mediates adaptation. Multi-locus theory is essential for understanding trait architecture, ancestry mosaics, and polygenic selection.

Important terms

Linkage disequilibrium

Non-random association among alleles at different loci

Selective sweep

Rapid fixation of an advantageous allele reducing nearby diversity

Background selection

Purging deleterious variants that reduces linked neutral diversity

Hill-Robertson interference

Competition among linked selected sites that reduces efficacy of selection

Genomic signatures

Selective sweep

Local reduction in diversity
Skew in site-frequency spectrum
Extended haplotype homozygosity

Background selection

Broad reductions in diversity in low-recombination regions
Can mimic demography if not modeled carefully

Quantitative framework

Linkage disequilibrium can be summarized by $D = p_{AB} - p_A p_B$ . Recombination reduces $|D|$ each generation, while selection and drift can create or maintain associations. In polygenic adaptation, many loci of small effect can respond in parallel, often leaving subtle allele-frequency shifts rather than classic sweeps.

What is a hallmark of a classic hard selective sweep?

Genome-wide increase in diversity Local reduction in diversity around the selected site No effect on linked sites Only changes in phenotype, not genotype

Define linkage disequilibrium in one sentence.

Which process tends to restore linkage equilibrium most directly?

Recombination Selection Genetic drift Mutation

Detecting selection in genomic data

1

Identify a null model that accounts for demography and recombination.
2

Scan for outliers in diversity, haplotype structure, or allele-frequency spectra.
3

Test alternative explanations such as introgression or background selection.
4

Validate with simulations under matched demographic conditions.
5

Replicate signals across datasets or populations when possible.

Common statistics in genome scans

Statistic	Typical use
$F_{ST}$	Population differentiation
Tajima's $D$	Site-frequency spectrum shape
iHS	Recent selection on haplotypes
$\pi$	Nucleotide diversity

Why can background selection complicate sweeps scans?

Neuroanatomy Explorer

Memory Deck

Match the Pairs

Concept Constellation