New insights into the development of dog breeds

There's another fascinating paper out this week from the laboratory of the Doyen of Dogs, Dr Elaine Ostrander at the National Institutes of Health (Parker et al 2017). This study sorts through the genotypes of 161 breeds of dogs to establish genetic relationships and reveal the history of breed development through hybridization of existing breeds.

Below is the dendrogram that displays the genetic relationships among the breeds in this study. (If you don't know how to read a dendrogram, there's a little tutorial here.)

​

The breeds in this dataset sort themselves into 23 different "clades", groups of dogs that share common genetic history. By comparing blocks of DNA called "haplotypes" that tend to get inherited together, we are able to identify breeds that share unique genetic signatures that indicate "introgression", or crossbreeding, at some time in the past.

For example, the Rhodesian Ridgeback shares haplotypes with a number of breeds: the Airedale Terrier, Irish Terrier, Irish Wolfhound, Greyhound, Doberman, Rottweiler, Great Dane, Boerboel, Bullmastiff, and Bulldog (see the figure below). According to the history of this breed, European breeds were bred with the Hottentot hunting dog of the Khoisan in Southern Africa, from which the ridge was inherited. That dog is long extinct, however, so it can't appear in this illustration of shared haplotypes, but the Basenji does, which probably shares some ancestry with the Hottentot dog.

​

The origin of the Schipperke has long been disputed, with some believing it to be most closely related to the Pomeranian, while others argue that it was developed by breeding down in size from the Leuvenaar, a black shepherd that is the ancestor of the modern Belgian Shepherd or Groenendael. From the haplotype chart for the Schipperke (below), it is evident that there is DNA shared with a number of toy and spitz-type breeds, including the Pug, Brussels Griffon, Pomeranian, and a few others. But by golly, the Belgian Shepherd and Tervuren show up as well, so there appears to be some truth in both stories!

​

One thing you need to be careful about when interpreting these haplotype charts. The analyses identify breeds that share blocks of DNA, but there is no way to know if breed A was crossed into B or the other way around. For instance, the Schipperke could have been used in the development of the Keeshond, or the Keeshond might have already existed and was used to develop the Schipperke.
We are making much progress unraveling the genetic history of dog breeds, but still we have barely scratched the surface. I pulled out a data file of dog breeds of the world to compare with the impressive list of breeds in this study. My list was compiled from Desmond Morris's wonderful book "Dogs", which lists more than 1,000 breeds, including not just the 400 or so breeds "recognized" by kennel clubs, but hundreds of others that are lesser known to the dog fancy (and still it's not complete!).

Using these data, I've created a graph of the number of breeds by country and indicated the number with DNA information (below). For most countries, we have studied fewer than half the known breeds, and there are dozens of countries that are still completely off the research radar. Most studies to date have focused on the popular breeds registered by AKC and the UK Kennel Club, no doubt because those dogs are abundant and convenient. I personally think that these breeds, while familiar, are not representative of the great diversity of dog breeds elsewhere in the world. I hope that we can start including these less common breeds in future research, not only because they are novel, but also because most have not been subjected to the strong selection of breeding for show and we might find that they have very interesting genetic stories to tell.

Breeds by country

​

_breeds_by_country.png
Download File​

​

In the graph below, I have sorted the breeds by function, which reveals that companion dogs and earth dogs are overrepresented, the latter no doubt reflecting the strong bias towards breeds developed in the UK for vermin control. The largest category would appear to be scenthounds, but many of these breeds are from France, where color and size varieties are given different names. Again, it's clear we have only scratched the surface in terms of exploring the variation within functional types.

Breeds by function

​

2017-04-28_1648.png
Download File​

​

This is fascinating stuff. It's hard to believe that we can sort through hundreds of breeds of dogs with just a bit of DNA! There is lots more to talk about in this paper that I will save for another post, but you should download a copy of the paper (below) and browse through it yourself.

REFERENCES
Parker, HG, DL Dreger, M Rimbault, BW Davis, AB Mullen, G Carpintero-Ramirez, & EA Ostrander. 2017. Genomic analyses reveal the influence of geographic origin, migration, and hybridization on modern dog breed development. Cell Reports 19: 697-708. doi: 10.1016/j.celrep.2017.03.079. (Download pdf).

Check out
ICB's online courses

*******************

Join our Facebook Group
ICB Breeding for the Future
...the science of dog breeding
*******************

Visit our Facebook Page
ICB Institute of Canine Biology
...the latest canine news and research​

Continue reading...