Quantifying risks of consanguineous mating in humans

The efficacy of selection in purging a deleterious mutation from a randomly mating population depends on numerous factors, including dominance effects of alleles – see my previous posts. Simplistically, most new mutations are expected to be heterozygotic, and be purged less effectively if they are recessive, than dominant alleles in homozygotes. In other words, recessive mutations segregate at higher frequencies (in heterozygotes), than dominant mutant homozygotes, especially so for Mendelian traits.

In inbreeding populations, thus, there’s a greater reduction of population fitness due to the accumulation of deleterious alleles in homozygotes, a phenomenon commonly termed “inbreeding depression”. Modern humans are a classic example of repeated non-random, consanguineous mating, resulting in localized population structure. Understanding the effects of non-random mating on the accumulation of recessive deleterious mutations is of great interest to disease genetics, anthropology, and the social sciences at large.

A Hutterite family – Image courtesy: National Geographic Channel – http://channel.nationalgeographic.com/american-colony-meet-the-hutterites/galleries/barnburners/at/king-colony-family-portrait-54169/

In a recent publication, Gao et al. (2015) attempt to quantify the fitness effects of deleterious (particularly autosomal, recessive lethal – i.e. inviable homozygote) mutations, using a founder population of Hutterites with well documented pedigrees, since 1950. The Hutterites are a small community of South Dakotans founded by 64 ancestors in the 18th-19th centuries, that migrated over to the United States and established 3 communal farms, and have since married consanguineously.

In short, Gao et al. (2015) (a) simulated Mendelian inheritance at 14 autosomal recessive disease genes along small and large pedigrees of the Hutterite population, (b) quantified loss or manifestation of mutant alleles (that were present in the founder population), and (c) use a Bayesian approach for point estimates (and intervals) of the number of autosomal recessive lethal mutations in a haploid human genome. Their study finds that 57% of recessive mutations from the founding population were lost prior to 1950 (after which extensive disease and pedigree records exist), 19% of the surviving alleles were present in homozygotes, and that 8.1% of the founder alleles will be expected to manifest in the present generation. This is also equivalent to 0.29 recessive deleterious mutant alleles on an average per haploid human genome. More importantly,

(this study) suggests that the risk of autosomal recessive disorders that manifest after birth should be increased by 0.29/16 = 1.8% in offspring of first-cousin couples (assuming no difference in environmental factors).

Here’s some recent press coverage on this study: http://www.sciencedaily.com/releases/2015/04/150408100522.htm



Gao, Z., Waggoner, D., Stephens, M., Ober, C. & Przeworski, M. Genetics 199, 1243–1254(2015). DOI: http://dx.doi.org/10.1534/genetics.114.173351


About Arun Sethuraman

I am a computational biologist, and I build statistical models and tools for population genetics. I am particularly interested in studying the dynamics of structured populations, genetic admixture, and ancestral demography.
This entry was posted in genomics, mutation, pedigree, population genetics, selection, societal structure and tagged , , , , . Bookmark the permalink.