Been reading recently about methods to determine how inbred a particular person is by measuring their "runs of homozygosity." Loosly, these are defined as regions of the genome (typically >1Mb) where the person is homozygous for all polymorphisms, suggesting their parents were identical for this region. If a person has 'many' of these regions, it suggests their parents were closely related.

A study in 2015 [1] found some evidence that people with more regions of homozygosity had higher risk of heart disease.

My question firstly relates to the amount of homozygosity you might expect in a 'normal' out-bred person, compared to an in-bred population (e.g. small community), and then to 'extreme' examples of relatedness (parent/child offspring). In my searching I didn't see any cut-offs, if they exist, or whether this is even the right way to be thinking about it. And secondly, just how bad is it to have these runs?

  1. Christofidou P, et al. Runs of Homozygosity: Association with Coronary Artery Disease and Gene Expression in Monocytes and Macrophages. Am J Hum Genet. 2015; 97:228–237. PMID: 26166477.
  • $\begingroup$ FIRST QUESTION: It all depends on the allele frequency in the population you consider. Under the Hardy-Weinberg assumption, the probability of two randomly picked allele to be identical by descent is $\sum_i x_i^2$, where $x_i$ is the frequency of the $i^{th}$ allele. If you look at a region that is under strong purifying selection in a large population, then most SNPs will have very low minor allele frequency and very low heterozygosity and therefore you would see very long runs of homozygosity. $\endgroup$ – Remi.b Apr 5 '16 at 17:09
  • $\begingroup$ SECOND QUESTION: The article you cite is probably one of the best place to go to ask "how bad are those runs". They probably report an effect size at some point. $\endgroup$ – Remi.b Apr 5 '16 at 17:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.