COLD SPRING HARBOR, NEW YORK— Compare the sprinting Shetland sheepdog with the sluggish St. Bernard, and it’s clear a dog’s genes play a large role in how athletic it is. Now, at the Biology of Genomes meeting here, scientists report identifying 59 genes linked to canine athletics, which apparently affect everything from heart rate to muscle strength. Early results suggest some may eventually help us understand human superstars.

“Across dogs, all sorts of traits have been selected for in an extreme way,” says Alexander Godfrey, a genomicist at the Massachusetts Institute of Technology Whitehead Institute in Cambridge, who was not involved in the work. As such, dog genomics represents “a pretty unique and powerful system” for studying how genotypes, or sets of genes, result in phenotypes, or sets of observable characteristics in all types of animals, he says.

Past work on dogs has yielded genes for friendliness , hair type , and other relatively simple traits. But this new study looked at more complex ones, thanks to a new resource: a soon-to-be-released global database of the whole-genome sequences of 722 dogs across about 450 breeds, along with sequences for canine relatives, including wolves, foxes, and jackals.

Jaemin Kim, a postdoc working with canine genomicist Elaine Ostrander at the National Human Genome Research Institute in Bethesda, Maryland, focused on athleticism, in part because he wondered why he wasn’t any better at his favorite sport: basketball. He decided to start with the genes that turn sport dogs such as pointers, setters, and retrievers into the Michael Jordans of the canine world. He and colleagues compared the genomes of 21 individuals from 10 sport hunting breeds with 27 individuals from nine terrier breeds.

Fifty-nine genes, or the regions that control them, stood out, with certain versions of the DNA much more common in the sport dogs, Kim reported at the meeting. He and his colleagues could not easily verify their effects on athletic performance, but most are linked to traits including blood flow, heart rate, muscle strength, and even pain perception. One seems to help dogs remain calm after they hear a gunshot, he added, which may make them stable hunting companions; a different version in terriers may account for their well-known neuroticism.

To examine the role of these genes in other breeds, Kim needed a standard way of assessing athleticism. He decided to use agility trials, competitions in which dogs, guided by their owners, maneuver through an obstacle course in the shortest time possible. Data from the United States Dog Agility Association allowed him to calculate the best performing breeds: border collies and Shetland sheepdogs. The worst were Newfoundlands, bulldogs, and mastiffs.

Then, he compared whole genomes from the best and the worst, looking for differences in the 59 genes. Only one proved to be significant, a gene called ROBO1 that affects learning ability. So when it comes to agility, Kim said, it seems that a mental attribute may matter more than physical ones do. “It looks like it’s more of a training thing,” says Sarah Tishkoff, an evolutionary geneticist at the University of Pennsylvania who was not involved with the work.

“It’s interesting to think about what genes are associated with what traits,” Godfrey says. “That it would be a gene that’s not involved with muscles is not obvious.” Even though agility trials are a good measure, Godfrey cautions that in general humans are notoriously bad at objectively evaluating their own and other people’s dogs. And he wonders whether, even in agility, judges wind up “scoring aspects of human[like] behavior that they like” and not agility per se, he points out. Another issue is that there are other types of athleticism. Herding dogs, for example, are great athletes that race around keeping livestock together and headed in the right direction, even though they are not that muscular looking. Kim is starting to look at the genetic basis of that behavior.

Ostrander says the new results might one day help us better understand the genetic basis of athleticism in humans. Already, other researchers have implicated one of the 59 genes in improving human performance by improving blood flow, and it’s likely, she says, that others will also prove important. Dogs suffer many of the same health problems that people do , and canine versions of the relevant genes will be easier to track down. Because breeders work hard to bring out specific traits in their dogs, “you get mutations in pathways that have dramatic effects,” Tishkoff explains.