What is linkage disequilibrium?

Linkage Disequilibrium: Definition and Significance in Genomics

Linkage disequilibrium (LD) is a nonrandom association or dependence of alleles at different loci in a given population, making the frequencies of the alleles deviate from the expected frequency if the alleles were independent. 1

Key Characteristics of Linkage Disequilibrium

• LD occurs when alleles at different loci are physically linked on a chromosome, leading to non-random coinheritance such that their frequencies in a population are correlated 1
• LD is a consequence of genetic distance, history of mutation events, and changes in population dynamics 1
• The strength of LD between loci decreases with increasing physical and genetic distance between them 2
• LD patterns vary significantly across different populations due to their unique demographic histories 3

Measurement and Quantification of LD

• Common statistical measures of LD include squared correlation coefficients (r²) for phased genotypes and a related measure for unphased genotypes 4
• At equilibrium, LD values are determined by sample size, recombination frequency, effective population size, and mating system 4
• Testing for LD when linkage phase is unknown requires statistical methods that account for the ambiguity of unobserved haplotypes 5
• The likelihood-ratio statistic is commonly used but can have biased type I error rates; composite statistics provide more reliable alternatives 5

Applications in Genomics and Medicine

• LD is used as a tool for gene mapping and estimation of effective population size 4
• In genome-wide association studies (GWAS), LD is crucial for understanding the relationship between identified variants and potential causal variants 1
• LD patterns help determine appropriate multiple testing thresholds in GWAS (typically p < 5×10⁻⁸), based on the estimated one million independent regions across the genome 1
• LD is exploited in transcriptome-wide association studies (TWAS) to identify gene-trait associations by integrating GWAS and gene expression datasets 1

Importance in Population Genetics

• LD serves as a sensitive indicator of population genetic forces that structure a genome 6
• Evolutionary biologists and human geneticists use LD to understand past evolutionary and demographic events 6
• LD patterns differ between isolated populations and more general populations, with isolated populations often showing higher levels of LD, particularly on sex chromosomes 3
• LD is widely distributed in anonymous regions of the human genome and may allow more accurate measurement of small genetic distances than standard linkage analysis 2

Limitations and Considerations

• LD patterns vary substantially across different ancestral populations, affecting the accuracy of imputation and genetic analyses 1
• Many genotyping platforms have been designed based on European LD patterns, potentially limiting their utility in non-European populations 1
• The non-independence of genetic variants due to LD means that standard multiple testing corrections (like Bonferroni) may be overly conservative in genomic studies 1
• LD can lead to spurious prioritization of non-causal genes in methods like TWAS, especially when using expression data from non-trait-related tissues 1

Clinical and Research Implications

• Understanding LD is essential for fine-mapping procedures that aim to identify causal genetic variations affecting traits of interest 1
• LD patterns influence the design of genotyping arrays and the selection of tagging SNPs for genetic studies 1
• The extent of LD in different populations is an important consideration when selecting ascertainment strategies for studies of complex diseases 3
• Accounting for LD is necessary when performing multiple testing adjustment in GWAS to avoid being overly conservative 1