Inbreeding and deleterious variation in South Asian populations

Consanguinity (or inbreeding) refers to the offspring of unions between closely related individuals. It is associated with a wide range of adverse outcomes, including increased mortality and disease burden. More than a billion people live in communities where consanguineous marriages are common, and these communities generally have high rates of rare, recessive diseases. Here, we estimate the degree of inbreeding in a wide range of groups, primarily from South Asia, and highlight shortcomings of previous inbreeding estimation methods. The estimated proportion of consanguineous individuals varies widely by country and group, ranging from 9.3% in Bangladesh to 42.8% in Pakistan for medical cohorts, and from 0 to 100% in Indian indigenous (Adivasi) groups. Within specific subgroups, country-of-residence influences consanguinity proportions, with Indian subgroups showing more consanguinity than their counterparts in other countries. Isolated groups identified from contemporary observations to have reduced population sizes (e.g., Jarawa, Onge, Nicobarese, Aeta, Toda and Toto) all display both increased levels of inbreeding and decreased estimates of current effective population size, in line with expectations. Finally, we quantify the genetic 'cost' of inbreeding by tabulating the numbers of presumed deleterious mutations across runs of homozygosity (ROH) in inbred individuals. We find a greater density of these deleterious mutations in long ROH, consistent with standard evolutionary models of purifying selection, and find evidence for the reduced efficacy of purifying selection in small, isolated populations.