This web page was produced as an assignment for an undergraduate course at Davidson College.
Study into the origins of dog domestication are complicated by several factors such as the wide phenotypic variation between breeds as well as trade and breeding of dogs across a wide geographic range. Despite a difficulty in the study, prior proposals on dog origins have situated the origins of dogs in the Middle East and East Asia.
New research presented by the paper suggests dog domestication occurred between 18,800 to 32,100 years ago in Europe and Siberia. Researchers came to the conclusion by sequencing and comparing 148 mitochondrial genomes consisting of 18 ancient canids, 49 wolves, 77 modern dogs, and 4 coyotes. Using the mitochondrial comparisons, the authors create a phylogeny that establishes modern dog ancestry from ancient European canids, and suggests divergence from wolves greater than 15,000 years ago.
Overall the paper’s methods and conclusions are easy to understand. The introduction provides a strong background on the prevailing thoughts on the origins of dog domestication that contrasts well with the author’s research. A strength of the paper is several tangents discussed, such as the failed domestication of the Goyet dog. The tangents make sense of the data and also present interesting relevant stories.
A weakness of the paper is the use of Bayesian analysis without prior explanation of the field. According to Wolfram Math World, Bayesian analysis is a type of statistical technique that utilizes parameters to determine relative probabilities. Given that the paper mentions Bayesian analysis in critical parts of the paper and it is used to describe both figures and tables, a cursory background would have benefitted readers.
I appreciated the discussion on the potential shortcomings of the paper, such as the insufficient coverage in ancient canids. I would have preferred more of a discussion on two-phase bottleneck as the discussion was interesting yet short. The mention of a decline in the population size between 5000 and 2500 years ago was interesting but was discussed shallowly and almost as an afterthought.
Figure 1. Figure 1 depicts the relations between modern and ancient dogs, indicated by a blue color, with wolves, indicated by the color orange. Specimens that were unable to classified without ambiguity were colored gray. The authors note that all dogs fall within four clades designated A through D, with most dog sequences belonging to clade A. Furthermore, all clades excluding D are related to ancient canids of European origin and the researchers assume that dog domestication thus originated in Europe.
Table 1. Table 1 consists of the 18 prehistoric canids from which mitochondrial genomes were sequenced by PCR. Each of the ancient specimens is categorized by morphology, either wolflike or doglike, age of specimen, mitochondrial genome coverage, and nucleotides sequenced. An average of 15,014 nucleotides were determined with a minimum two fold coverage.
Figure 2. Figure 2 is a Bayesian Skygrid plot that depicts the effective population size of dog clad A and pre-Columbian dogs. The plot notes areas of continuous population growth that the authors suspect are initial domestication attempts. Furthermore, the authors note the effective population size of clade A and pre-Columbian dogs, indicated by the solid line, exhibits growth at the same time of growth on the human population size trajectory.
Thalmann O., et al. 2013. Complete Mitochondrial Genomes of Ancient Canids Suggests a European Origin of Domestic Dogs. Science. 342: 871 - 874.
Andrew Liu's Homepage
Genomics Page
Biology Home Page
© Copyright 2014 Department of Biology, Davidson College, Davidson, NC 28035