Dog breed identification and what it means (Proceedings)

What breed a dog is, matters. In recent years, numerous public and private restrictions have been applied to specific breeds of dogs and to dogs believed to be partially those breeds. What breed is assigned to a dog can influence how people view a dog, interpret its behavior,  and  whether the dog is adopted.

Until the advent of breed clubs in the mid 1800's, registered breeds didn't exist. The majority of breeds today appear to have been derived from shared European stock. Many breeds have been developed from other breeds, either crossing breeds or selecting for a variation in an existing breeds. Such common ancestry results in shared morphological traits among many breeds of dogs. As Scott and Fuller (1965) demonstrated with their photographs of F1 and F2 crosses of purebred dogs, there is no telling what the offspring will look like. Basenji –Cocker Spaniels crosses  more closely resembled Labrador retriever and beagle mixes than their actual parents and grandparents.

Visual identification of mixed breed dogs is very unreliable. We recently looked at the results of matches of visual identification of dogs to DNA identification of the dogs. Over 900 people, in animal related professions and activities, viewed 20 one minute video-clips of dogs of unknown parentage.  DNA analysis indicated that none of the dogs were purebreds (Wisdom Panel TM). The vast majority of participants in the study identified the dogs as mixed breeds. However, for only 6 dogs did more than 50% of the respondents' visual identification of a mixed breed match any DNA breed identification of those dogs.  This doesn't mean that these professionals  don't know what breeds look like – it is that mixed breeds may not look like the dog breeds in their  ancestry and may have morphological characteristics that resemble other breeds.

Accuracy of DNA identification of mixed breeds is not perfect either. Companies that offer DNA breed analysis notify customers  that the validity of detection of the breeds in F1 crosses may be 90%, 84.5 %, etc. No companies claim 100% accuracy. Different companies may report different results depending on the dogs that were used to establish the reference-breed standards and the statistical algorithms and computer programs used to analyze the samples.

Regardless whether a portion of a breed is identified visually or by DNA, there is no way to know what behavioral or other characteristics of those breeds have been inherited. Who knows, the outside of the dog may look like something like a Rottweiler, but the brain is that of a poodle.  Dogs are individuals. Siblings, be they dogs or humans, have distinct personalities. They deserve to be evaluated as individuals.

References

Berkey J R. (2009) Dog Breed Specific Legislation:  The Cost to People, Pets and Veterinarians, and the Damage to the Human-Animal Bond, Proceedings of 146th Annual AVMA Meeting, July 11-13, 2009, Seattle.

Jones P et al. (2008) Single-nucleotide-polymorphism-based association mapping of dog stereotypes. Genetics 179, 1033-1044

Kochan M ( 2008). Can I see some I.D.? October, Dogfancy,38-41.

Markwell P.  (2007) The science behind the WISDOM Panel TM  MX: MARS Incorporated, Lincoln NE.

Parker HG et al.  (2004) Genetic Structure of the Purebred Domestic Dog , Science, 304,1160-1164.

Scott JP and Fuller JL,  (1965).  Genetics and the Social Behavior of the Dog. The University of Chicago Press.

Voith V.L. et al. (2009)  Comparison of Adoption Agency Breed Identification and DNA Identification of Dogs. Journal of Applied Animal Welfare Science, 12 (3), 253-262, 2009.

Voith V.L.(2010) Improving Communications: Inter-rater Reliability of Dog Breed Identification and Comparison to DNA Analysis 2010 AVMA Convention Notes, Atlanta,

Wisdom Panel Professional Mixed-Breed Genetic Analysis: Analysis Summary (2010), MARS veterinary, Lincoln NE.