REVIEW: Complete Mitochondrial Genomes of Ancient Canids Suggest a European Origin of Domestic Dogs



Domestication is an event that has occurred in many different species due to human interference. Thalmann et al. was particularly interested in the domestication of dogs and wanted to determine the region modern dogs originated.  The oldest dog remains have been found in Western Europe and Siberia and are 15,000 to 36,000 years old, while the oldest remains from the Middle East and East Asia are 13,000 years old maximum.  Using DNA capture and high throughput sequencing mitochondrial DNA was sequenced from prehistoric canids, and modern wolves from America and Eurasia.  The team compared these sequences to wolf, dog, divergent breeds of dog (Basenji and Dingo), coyote and Chinese indigenous dog sequences.  

Dogs could be categorized into four clades (A, B, C and D).  Almost 80% of the dogs fell into the A, C or D categories.  Clade A was the most diverse, while clade D was the least diverse.   Clade B was determined to be the closest relative to modern wolves. While looking at the data it is important to keep in mind that the ancient DNA was obtained from one locus.

From their data the group concluded that dogs most likely arose from ancient European dogs that were domesticated by European hunter-gatherers.  Sequences from Europe are the only ones closely related to modern dogs. The group proposed that there was a two-stage bottleneck.  The first stage was the actual domestication of dogs and the second was the recent increase in breeding for specific traits in dogs.  The group also noted that there were population fluctuations in dogs that mirrored human population changes, including a large increase 5,000 to 2,500 years ago.  This lead them to the conclusion that domestication of dogs occurred in Europe, but there were also failed domestication trajectories such as the Goyet of Belgium and the Altai Mountain specimen of Russia.

My Opinions:

I thought that the paper was well written. It was interesting to see how the researchers put together a phylogenetic tree to trace dog ancestry. My biggest problem with the paper was that I did not see why they spent so much time and money trying to figure out where dogs came from. I do not see how this really benefits the field of genomics other than showing that ancient DNA can be used to trace lineage. I expected that the conclusions would suggest a purpose to this study, but it only told us that successful domestication of dogs could be traced to Europe. I just want to know why they did this study and why science picked it up. My only explanation is that people like dogs. Some of the sections of this paper were a bit wordy and would have been better summarized by a figure. It is also important to note that any ancient DNA findings should be approached with caution. Amplification techniques used in ancient DNA testing use tiny pieces of DNA, making it very easy for foreign DNA to be amplified by mistake. Thalmann et al. had a sentence that was alarming to me regarding their amplification techinques, “mitochondrial and nuclear genomes suffer from incomplete lineage sorting, which, given the recent divergence of dogs and wolves, can potentially confound evolutionary inference,” (Thalmann et al. 2013). Apparently, this problem can be remedied by amplifying multiple loci, but the group said that they were unable to obtain enough coverage to make confident calls in any ancient sample. This statement makes me question how useful the information generated from this study is, as well as how accurate it is, leading me to question, once again, why science published this study.

Figure Summaries:

Fig 1.

Question: How are modern dogs related to wolves and ancient dogs (canids) from different parts of the world?

Method: They used Bayesian analysis that aimed for bootstrap values greater than 90%. Blue indicates modern or ancient dog sequences, orange are wolf sequences, and grey lines are ambiguous taxonomic classifications.

Take Home Message: There is one ancestor around 80,000 years ago that gave rise to the diverse canine population today. It also shows that the modern dog clades' most recent common ancestors are from Europe (Switzerland, Germany, and Sweden). The blue triangles indicate the diversity of the clades. Clade A has the most diversity and clade D has the least. It also shows that clade A is the diversified least recently and D has gained its diversity most recently. Also, almost every branch contains a star, indicating that they are confident in their calls. This image summarizes how the group put together all the information they gleaned from their work with the mitochondrial sequences.

Fig 2.

Question: What is the demographic trajectory of clade A?

Method: This is a Baysian Skygrid plot. The group created the solid line by comparing population size (median logNe) change to the years before present. They also used 95% Highest Posterior Density (HPD) interval to create the blue cloud around the line. HPD is a form of a confidence calculation, so they are 95% sure that the solid line would lie in the blue cloud.

Take Home Message: This figure shows the effective population size of dog clade A from its origin, around 20,000 years ago. It shows that as the human population has increased in the past 2,500 years dog clade A has also had a population spike.

Table 1.

Question: Which ancient specimens are the oldest and most dog-like, while having a reasonable amount of coverage?

Method: They used capture arrays to obtain the specimens then used long range PCR and biotinylated adapters to amplify the mtDNA.  They denoted questionable morphological classification with italics.  They compared the morphology to the coverage, origin, date, and coverage to achieve a big picture view of the origin of domestication.

Take Home Message: The table shows that the oldest doglike remains are from europe and that these are good reads with sufficient coverage.


O. Thalmann, B. Shapiro, P. Cui, V. J. Schuenemann, S. K. Sawyer, D. L. Greenfield, M. B. Germonpré, M. V. Sablin, F. López-Giráldez, X. Domingo-Roura, H. Napierala,H-P. Uerpmann, D. M. Loponte, A. A. Acosta, L. Giemsch, R. W. Schmitz, B. Worthington, J. E. Buikstra, A. Druzhkova, A. S. Graphodatsky, N. D. Ovodov, N. Wahlberg, A. H. Freedman, R. M. Schweizer, K.-P. Koepfli, J. A. Leonard, M. Meyer, J. Krause, S. Pääbo, R. E. Green, R. K. Wayne. (2013) Complete mitochondrial genomes of ancient canids suggest a european origin of domestic dogs. Science 342:871-874.


Genomics Page
Biology Home Page

Email Questions or Comments:

© Copyright 2014 Department of Biology, Davidson College, Davidson, NC 28035