Genomic insights into inbreeding depression

Whole-genome sequencing provides unprecedented resolution in measuring inbreeding in natural populations of conservation concern. Following functional extinction in the 1960’s, a single pair of wolves recolonized the Scandinavian peninsula in the 1980’s. We sequenced the genomes of 97 individuals to document in detail the genomic consequences of inbreeding in the population.

Go to the profile of Marty Kardos
Nov 20, 2017
2
2
Upvote 2 Comment

The paper in Nature Ecology & Evolution is here: http://go.nature.com/2iDrFab

Whole-genome sequencing provides unprecedented resolution in measuring inbreeding in natural populations of conservation concern. Following functional extinction in the 1960’s, a single pair of wolves recolonized the Scandinavian peninsula in the 1980’s. We sequenced the genomes of 97 individuals to document in detail the genomic consequences of inbreeding in the population.

Gray wolves (Canis lupus) were eliminated from much of their historic range across the Northern hemisphere by the mid 20th century by hunting and trapping. Wolves became functionally extinct on the Scandinavian Peninsula in the mid 1960s. Following legal protection in Sweden (1966) and Norway (1972), a single breeding pair arrived in Sweden in the early 1980s, marking the beginning of a recovery of wolves on the Peninsula. However, small population size and the arrival of only a handful of additional reproductively successful immigrants since recolonization meant that inbreeding (mating between relatives) has been frequent in the population.

Intensive monitoring of the population resulted in a complete pedigree, and molecular genetic marker (microsatellite) data reaching back to the founding of the population. These data have yielded invaluable insights into inbreeding, alleviation of inbreeding depression via immigration (i.e., genetic rescue), and many other questions relevant to Scandinavian wolf ecology, behavior, and, management.

While pedigrees and genetic markers have been the mainstay of inbreeding depression studies for decades, they provide a limited picture of inbreeding. Pedigrees underestimate inbreeding when common ancestors of parents are unaccounted for. Additionally, randomness in the mechanisms of genomic inheritance (Mendelian segregation and recombination) can cause pedigree estimates of inbreeding to vary substantially from realized genomic inbreeding.

Genetic marker-based measures of inbreeding have the advantage of directly measuring genomic variation and are thus unhindered by the stochastic nature of genomic inheritance. However, markers sample only a tiny fraction of the genome, which means they can also be highly imprecise. Geneticists have recently been quite interested in just how much of the genome must be characterized to reliably measure individual inbreeding, and in how useful pedigrees really are in studies of inbreeding and inbreeding depression.

The historical lack of large-scale genomic data has left fundamental questions related to inbreeding essentially unanswered in many study systems. For example, what genomic regions contribute to inbreeding depression? How inbred are population founders and immigrants? Just how much of the genome needs to be characterized to reliably measure individual inbreeding?

We developed a collaboration among geneticists and ecologists from across Scandinavia to address some of these questions. Our main objectives were to 1) comprehensively measure inbreeding in the population using genome resequencing (genome sequence data aligned to the dog reference genome); and 2) determine how reliably genomic inbreeding is measured with the pedigree versus smaller numbers of loci across the genome.

The wolf genomes revealed clear molecular signatures of inbreeding in the form of numerous long ‘runs of homozygosity’ (ROH). This was most striking in several very highly inbred individuals, where whole chromosomes were often completely homozygous. Observing such extreme signatures of inbreeding in genome sequences demonstrates the power of modern genomics to enhance our understanding of inbreeding and genetic variation in natural populations to an extent that has only recently become possible.

Our analyses also showed that the Scandinavian wolf pedigree actually did quite a good job at predicting realized genomic inbreeding (pedigree inbreeding explained nearly 90% of the variance in realized genomic inbreeding). Such a strong relationship between pedigree and genomic inbreeding is mostly due to the relatively high variance in inbreeding in the population, and the large number of chromosomes (38 autosomes) in the wolf genome. Pedigree estimates of inbreeding are expected to have higher precision in species with a large number of chromosomes because of reduced ‘noise’ in the inheritance of the genome via Mendelian segregation. The high precision of pedigree inbreeding is comforting because it shows that results from pedigree analyses of inbreeding are likely to be highly reliable in wolves and other species with many chromosomes, at least in populations where the variance of inbreeding is quite high.

We also showed that as few as 500 genetic markers (single nucleotide polymorphisms) more precisely estimated realized genomic inbreeding than the pedigree did. This is good news because as it shows empirically that only a modest number of genetic markers, which are inexpensive and do not require a reference genome, are needed to reliably measure individual inbreeding and inbreeding depression in populations where inbreeding has high variance. Previous studies have used simulations to evaluate the precision of pedigree- and marker-based measures of inbreeding. But empirical comparisons of the precision of markers and pedigrees at estimating inbreeding measured from the entire genome in natural populations has been lacking.

While pedigrees and genetic markers will often be efficient at detecting inbreeding depression in natural populations, there are some questions that will require large-scale genomic data such as whole-genome sequences.  There are indeed many exciting opportunities to further our understanding of inbreeding depression in the future. I anticipate that there will be many additional forthcoming studies that begin to unravel the genetic basis of inbreeding depression. For example, what fitness components and biochemical pathways are most frequently involved with inbreeding depression? How many genes are involved with inbreeding depression? Are genes with large effects on inbreeding depression common? How does the genomic basis of inbreeding depression affect population viability? Our analyses identified regions with very few ROH that are likely to contain loci that contributed to inbreeding depression. Additional work is needed to confirm effects of these regions, and to unravel the genetic architecture of inbreeding depression.

 

Go to the profile of Marty Kardos

Marty Kardos

Postdoctoral reaearcher, University of Montana

2 Comments

Go to the profile of Gidon Eshel
Gidon Eshel 24 days ago

Such nice, important, thoughtful work. Thank you!


Go to the profile of Marty Kardos
Marty Kardos 16 days ago

Thanks very much, Gidon!