Understanding How Breed Relationships Facilitate Genetic Studies of Complex Traits and Diseases

Heidi Parker, Pill; National Human Genome Research Institute/National Institutes of Health

In 2004 we examined the phylogenetic relationships between populations of the domestic dog and produced the first genetic classification of dog breeds. Since its release, the classification system has been a prominent focus of studies in all areas of canine genetics. To increase the applicability of the breed population study to the ever growing field of dog genetics, we have extended the initial analysis from 85 to 132 breeds and varieties, with an emphasis on including breed types that were under represented in the first study. The addition of 47 breeds has revealed a fifth cluster as well as a number of closely related sub-clusters. These results are currently being applied to the investigation of the genetic basis of complex diseases occurring in several breeds. Breeds within a cluster have more genetic similarities to each other than to breeds in other clusters. These similarities are likely obtained from common ancestors, therefore, traits shared by these breeds are more likely to be caused by the same mutation than those found in unrelated breeds. We can compare patterns between clustered breeds that share a particular phenotype in order to find the mutations that cause that phenotype. An excellent example of this is shown in the fine mapping of Collie Eye Anomaly where multi-breed comparisons reduced a region of 3Mb to just over 100 kb and allowed for the identification of a causative mutation. This multi-breed approach is currently being applied to studies of Addison's disease and transitional cell carcinoma.

Biographical Profile

 

Dr. Heidi Parker is a Research Fellow in the Cancer Genetics Branch at the National Human Genome Research Institute of NIH. Dr. Parker received her Ph.D. from the University of Washington and Fred Hutchinson Cancer Research Center where she studied the population to structure of the domestic dog. She continues her work on canine genetics at the National Institutes of Health by applying information gained from population studies to the mapping and identification of genes involved in complex traits and diseases. Dr. Parker is currently leading projects aimed at identifying genetic variants associated with cancer susceptibility, auto-immune disorders, and limb morphology, as she continues to study breed structure, development, and history through genetics.

Select Publications:

Parker, H.G., Kukekova, A.V., Akey, D.T., Goldstein, O., Kirkness,E.F., Baysac, K.C., Mosher, D.S., Aguirre, G.D., Acland, G.M., Ostrander, E.A., Breed Relationships Facilitate Fine Mapping Studies: A 7.8Kb Deletion Cosegregates with Collie Eve Anomaly Across Multiple Dog Breeds. (2007) Genome Research. Nov; 17(11): in press.

Mosher, D.S., Quignon, P.1, Bustamante, C.D., Sutter, N.B., Mellersh, C.S., Parker, H.G., Ostrander, E.A., A Mutation in the Mvostatin Gene Increases Muscle Mass and Enhances Racing Performance in Heterozvgote Dogs. (2007) Public Library of Science - Genetics. May 25; 3(5):e79.

Sutter, N.B., Bustamante, C.D., Chase, K., Gray, M.M., Zhao, K., Lan Zhu, Padhukasahasram, B., Karlins, E., Davis, S., Jones, P .G., Quignon, P., Johnson, G.S., Parker, H.G., Fretwell, N.X., Mosher, D.S., Lawler, D.F., Satyaraj, E, Nordborg, M., Lark, K.G., Wayne, R.K., Ostrander, E.A., A Single Ancient IGFI Allele Causes Small Size in Dogs, (2007) Science. April 6; 316(5821):112-5.

Parker, H.G., and Ostrander E.A. (2005). Canine genomics and genetics: Running with the pack. PLOS Genetics. 1:507-53.

Parker, H.G., Kim, L.V., Sutter, N.B., et al.(2004). Genetic structure of the purebred domestic dog. Science 304:1160-4.

 

 

Heidi Parker – Breed relationships and the Study of Canine disease – conference notes

 

Modern domestic dog breeds – greatest diversity in size, shape and behavior of any mammalian species.  Some breeds are over 1000 years old.

 

This did not happen by chance… Phy-Do Study.  What can DNA tell us about the breeds?

            5 unrelated dogs, 132 breeds/varieties, 96 markers, that cover all autosomes

 

Every breed has a different genetic signature – graph showed how they were grouped by the computer after analyzing their genetic component – 19 breeds

 

When done for 132 breeds, similar breeds blocked together – bully breeds, Belgian breeds (sheepdog, terv, malinois), but other breeds showed individuality within the map. 

 

Took breeds and clustered them into five different groups. 

            Asian-Ancient breeds – wolf closest to them – akita, shar-pei, basenji, afghans, arctic breeds

            Mastiffs – Mastiffs, bulldogs, bull terriers, other terriers

            Herding dogs – herding dogs, sighthounds

            Hunting group – pointers, setters, scent hounds, retrievers, spaniels

            Mountain dogs – mountain dogs, GSDs, more spaniels, std poodle

            Small group – misc breeds

Using population structure to enhance mapping studies

1. Closely related breeds with identical traits may be combined to increase mapping power

2. Distantly related breeds with identical traits can greatly reduce region of linkage through shared haplotype (pattern of mutation) analysis

            3. Unrelated breeds with similar traits reveal gene pathways

            4. Closely related breeds lacking the trait rule out false positives.

 

Example: Collie Eye Anomaly – genetic mutation carried across similar breeds.  All affected dogs carried four genes in common in the same region of chromosome 37. 

 

Inherited diseases in dogs.

            More than 350 genetic diseases identified

 

During breed formation, some phenotypes are heavily selected for and driven to near fixation

           

Near by disease alleles rise to increased frequency

           

Popular sire syndrome

 

            Small number of foundation dogs created a foundation for genetic diseases

 

Malignant Hystiocytosis – another example

*found at least two common loci in study within the BMD (family study, genetic study and European family)

            Other breeds: Flat coats (possibly higher than BMD), goldens and rottweilers

            With the BMD dog in the US, they are more similar than the BMD in Europe, this will be helpful in comparing linked regions to find associated mutations. 

            Use the Greater Swiss MD as a control.

 

Added Benefit of mapping in a population rather than a family

 

Possible to map in one generation – important for late onset diseases

            All the samples are equally useful, both affected and controls

            Data set has multiple uses:

                        Family studies – 1 genome scan

                        More effective use of resources:

                        Many samples = many disorders can possibly be mapped