The domestic dog genome

Why are dogs of interest to biologists? Over 400 breeds of dog exist worldwide today, each a closed breeding population with a unique pattern of size, morphology and behavior. But the aggressive breeding programs used to create purebred dogs have burdened them with over 350 inherited disorders, many associated with just one or a few breeds. Breeds expanded rapidly from popular sires to meet breeders’ demands suffer the most. It has been suggested that the unique structure of the purebred dog population may greatly simplify the mapping of genes associated with complex traits that have proven intractable by studying human families.The canine genome map. The most recently published canine map, a joint venture between investigators at the University of Rennes and the Fred Hutchinson Cancer Research Center (FHCRC) has an average of one marker every 900 kb across the dog's 38 autosomes and X and Y chromosomes. A minimal mapping set of 325 well distributed, highly polymorphic microsatellite markers is available for genome-wide scans of 8–10 cM density. Efforts are also underway, in collaboration with investigators at The Institute for Genomic Research (TIGR), to order 10,000 independent genes derived from the 1.5X sequence on the canine chromosomal map. This will provide a high density resource for comparative mapping between the human and dog genome. The canine expressed sequence tag (EST) project, an additional resource, is providing a foundation for much needed canine expression arrays. The effort has been led by investigators at Cold Spring Harbor Laboratory who have already deposited over 10,000 random sequences constructed from brain, testes and MDCK cell cDNA libraries into a publicly available database (http://nucleus.cshl.edu/genseq/dogweb/).The 1.5X canine genome sequence. Investigators at TIGR and the Center for Advancement of Genomics recently surveyed the dog genome sequence and provided data summarizing 1.5X coverage from a purebred male standard Poodle. The sequence is estimated to cover 77% of the genome and contains 6.22 million reads. More than 650 million base pairs of dog sequence align uniquely to the human genome, and is estimated to include fragments for 18,473 of 24,567 annotated human genes. Estimates of the euchromatic genome size for dog range from 2.31 to 2.47 billion base pairs, somewhat smaller than the human genome but similar to mouse. Comparison of human and dog genomes by radiation hybrid mapping have revealed ∼85 conserved blocks. Analysis of the 1.5X sequence confirms most of these blocks, but predicts many additional syntenic breaks and suggests the final number of orthologous blocks is ∼200. Among the most interesting features of the canine genome is that ∼31% is composed of repetitive sequences, smaller than the 46% and 38% found in the human and mouse genomes, respectively. The most abundant repeat, a SINE element (SINEC_CF), is thought to be derived from tRNA sequences. An estimated 7% — 16,000 of 230,000 copies in the genome — of the SINEC_CF elements are bimorphic (two alleles) in the sequenced Poodle. By comparison, the number of bimorphic Alu sequences in humans is estimated to be only 1200.Mapping simple mendelian traits. Existing resources have been used to tackle the mapping of common canine disorders. The top 10 diseases in purebred dogs include several of major concern to human medicine, such as cancer, epilepsy, immunodeficiencies, retinal disease, cataracts and heart disease. Disease loci have been localized for several canine diseases; some highlight genes already suspected from studies of human disorders such as kidney cancer, hemophilia and immune disorders. Others, such as the hypocretin 2 receptor associated with narcolepsy in Doberman Pinschers, were previously unknown. The latter has opened new avenues of study regarding the biology of sleep.View Large Image | View Hi-Res Image | Download PowerPoint SlideThe 6X canine genome sequence and SNP mapping. Last year, a white paper submitted to support the 6X sequencing of the dog genome received approval from NHGRI. Sequencing of a highly inbred female Boxer began at the Whitehead Genome Center in June 2003, with completion expected by Spring 2004. Plans are also underway to find single nucleotide polymorphisms (SNPs) in nine distinct breeds of dog and a wolf. While final selection is still underway, breeds being selected represent significant phenotypic variation, distinct lineages and high levels of genomic variation.Mapping complex traits. The genome sequence has increased interest in using the dog to map the genetics of complex traits. Most dog breeds have been created in a few generations, so a small number of key loci are likely responsible for their characteristic features. A recent analysis of canid morphology in Portuguese Water Dogs demonstrates the value of breeds with limited numbers of founders for studies in quantitative genetics. Investigators at University of Utah identified four significant quantitative trait loci (QTLs) for distinct aspects of canine skeletal morphology. Similar approaches are now applied to the study of complex diseases such as hip dysplasia.Linkage disequilibrium mapping. The unique history of dog breeds suggests that linkage disequilibrium (LD) mapping will be useful for identifying genes associated with both simple and complex traits. Additional studies will determine how truly polymorphic various dog breeds are, how much haplotype information breeds with anecdotal histories share, and how far regions of LD extend across the dog genome.Summary. The development of dog breeds by selection for rarified traits represents one of the greatest experiments in biological variation ever done by man. The dog genome map and DNA sequence offer great opportunities for understanding the genetic regulation that accounts for the greatest extremes of natural variation.