Consanguinity and genomic sharing in human evolutionary inference

Published on March 18, 2015   51 min

You are viewing a talk that is a part of one of our comprehensive courses. Additional learning material: case studies, projects, workshops and recommended reading; multiple choice questions and suggested exam questions with model answers are available on application. Learn more

Other Talks in the Series: Human Population Genetics II

0:00
Hello, I'm Trevor Pemberton, a Professor of Biochemistry and Medical Genetics at the University of Manitoba. Today I'm going to talk about the influence of cultural and population processes on patterns of identity by descenting human populations and their importance in understanding human evolutionary history and phenotypic variation.
0:17
I will start by defining what identity by descent is and the population and cultural processes that give rise to it. Next, I will introduce the inbreeding coefficient as a measure of identity by descent levels in individual genomes, the pedigree and genomic estimators used to calculate it, and give an overview of its patterns in worldwide human populations, and how these reflect their different cultures and histories. Then I will introduce runs of homozygosity as an approach to detect identity by descent regions in individual genomes, and how inbreeding coefficient estimates based upon this approach correlate with those obtained with the genomic estimator. Finally, I'll review some recent findings on genomic patterns in runs of homozygosity in worldwide human populations, their utility for understanding human evolutionary history, and briefly outline their importance in human phenotypic variation.
1:02
For a pair of individuals a genomic region is said to be identical by state if they have an identical nucleotide sequence in that region. An identical by state region is identical by descent if both individuals have inherited it from a common ancestor, that is the region has the same ancestral origin in these individuals. Genomic regions that are identical by descent are identical by state by definition. but regions that are not identical by descent can still be identical by state due to the same mutations arising in different individuals, or a combination of events that change the ancestral origin of the segment without altering its sequence.
1:37
The basic principles that underlie identity by descent sharing in individual genomes are shown in this diagram. Chromosomes with different ancestral origins in the two founders are represented by different colors. At each parent/ offspring transmission there exists an opportunity for chromosomal crossover to occur during the formation of the parents' gametes, where recombination exchanges genetic material between homologous chromosomes. These recombinant chromosomes are then passed stochastically on to the offspring, with this process repeated in subsequent generations. For convenience, if we consider the chromosomal complement of the two founders to derive from different ancestral origins, then they share no genomic segments identical by descent. Their offspring would, however, be expected to share many long identical by descent segments since recombination has only had a single opportunity to disrupt the ancestral haplotypes they inherit from the founders, while in subsequent generations fewer and shorter segments will be expected as additional crossover events further fragment the ancestral haplotypes segregating in their lineages.
2:32
During meiosis segments of identity by descent are removed by recombination. Therefore, the expected number of identical by descent segments shared by a pair of individuals depends on the number of generations since their most recent common ancestor. Since at each meiosis the probability of transmitting an identical by descent segment is one half, the number of identity by descent segments they share is Poission distributed, and follows this equation in which r denotes the number of recombination events expected per parent offspring transmission. In humans this has the sex average rate of approximately 31 events per meiosis. However, it should be note that the rate of observed in females is much higher than the rate observed in males. c denotes the number of autosomal chromosomes, which for humans is 22. a denotes the number of common ancestors the pair of individuals share, which in most cases will be 2, and will only unequal 1 if their shared ancestry involves half siblings. d denotes the number of parent offspring transmissions in the path connecting the pair of individuals.
3:26
For siblings there are two parent offspring transmissions, one for each offspring. So d is equal to 2. Thus, for first cousins d is equal to 4, and for second cousins d is equal to 6. So for a pair of individuals from the same generation that share two common ancestors, d is simply twice the number of generations since their common ancestor.
3:46
If we look at the distributions of expected numbers of identity by descent segments in different generations, we find that siblings will on average share 84. As we move through subsequent generations the number of segments decreases markedly, down to an average 36 for first cousins, 13 for second cousins, and just 4 on average for third cousins.
4:06
During meiosis segments of identity by descent are broken up by recombination. Therefore, the expected lengths of identical by descent shared by a pair of individuals depend on the number of generations since their most recent common ancestor. The expected length of identity by descent segments is exponentially distributed following this equation, in which d again denotes the number of parent offspring transmissions in the path connecting the pair of individuals. The lengths provided by this equation are in centiMorgans, a unit for measuring genetic linkage that is defined as the distance between chromosome positions for which the expected rate of chromosomal crossovers in a single generation is 1%. In humans this is on average equivalent to 1.3 million base pairs, but it should be noted that this distance varies across the genome with a standard deviation of around 800,000 base pairs. And the distance observed in females is much higher than the distance observed in males.
4:54
If we look at the distributions of expected lengths of identity by descent segments in different generations, we see that siblings will share segments over a range of lengths, including many long segments. As we move through subsequent generations, the distribution of segment lengths is progressively shifted towards smaller lengths through first cousins, second cousins, and third cousins, with an average length of around 65 megabases in siblings, down to just over 16 megabases in third cousins.
Hide

Consanguinity and genomic sharing in human evolutionary inference

Embed in course/own notes