This report summarizes a project of trying to further understand human complex diseases. WHY should this topic be of interest to gene-environment interactions? As we’ve often discussed in these GEITP pages, multifactorial traits include height, body mass index, type-2 diabetes, dementia, cancer, and even phenotypes such as drug toxicity and drug efficacy. Why do two workers respond differently when expsed to the same amount of an occupational hazardous chemical? Why do two patients respond differently to the same dose of a particular drug?
Linkage disequilibrium (LD) is defined as “the occurrence of some genes (or alleles, or DNA segment) near one another on the same chromosome, more often than would be expected by chance alone.” This “nonrandom association of alleles at different loci” is a sensitive indicator of population-genetic forces that structure a genome, that build the genetic architecture. Thus, the older the population, the less likely it will be –– that genes in the same neighborhood (along the same chromosome) will remain “linked” –– because of chromosomal cross-over events during meiosis of germ cells, i.e. which occur with each generation of offspring; two genes near one another in an older population would more likely show “low levels of LD,” and two genes near one another in a younger population would more likely show “high levels of LD.” Finally, “heritability” is defined as “an estimate of how much variation in a trait in a population is due to genetic variation between individuals in that population.”
Recent studies have implied LD-dependent genetic architecture of human complex traits, in which single-nucleotide variants (SNVs) of DNA with lower levels of LD have larger per-SNV heritability. Authors [see attached report] analyzed 56 complex traits (average N = 101,401) by extending stratified LD-score regression to continuous annotations. They determined that SNVs with Lower Levels of LD have significantly larger per-SNP heritability and that roughly half of this effect can be explained by functional annotations that are negatively correlated with Low LD, such as DNase I hypersensitivity sites. The remaining signal is largely driven by the authors’ finding that more recent common variants tend to have Lower Levels of LD and tend to explain more heritability (P = 2.4 × 10–104). Using the model of extreme discordance, authors found that the youngest 20% of common SNVs explain 3.9 times more heritability than the oldest 20% –– which is consistent with a mechanism of negative selection.
Nature Genetics Oct 2o17; 49: 1421–1427