The key to shedding light on crime might soon come from a creature that’s more known for shedding on carpet and clothes.

Here’s the scenario: a woman is killed – when police begin to investigate, they learn that she had a lap dog that she often pet. When dog hair is found in the car of the prime suspect in the case, prosecutors need some way of proving that it came from the lap dog.

Enter professor Marc Allard.

Allard, an associate professor of biology at GW, has spent the past five years mapping canine DNA in an effort to compile a database of genetic code for dogs. His goal is to make that information available to the forensic science community at some point in the next few years.

Although the practice of extracting mitochondrial DNA, or mDNA, from dogs is new, man’s best friend has been leaving criminal investigators clues for years.

In most cases today, when investigators want to compare dog hair found at a crime scene with a hair from a particular dog, they need to rely on a trained expert to compare the hair samples under a microscope and determine whether they are the same.

Allard said the problem with microscopic analysis is that it leaves cases based on subjective – and potentially disputable – judgments. With mDNA testing, a lab technician can prove that two mDNA samples match.

“The beauty of DNA is that you don’t have to have a skilled investigator sorting this out,” he said. Using robotics, the testing process is more consistent, faster and less reliant on costly experts, he said.

Animal DNA has proven effective in prior cases. In 1994 a Canadian man was convicted of murdering his former wife after police determined that the genetic code removed from cat hair found on the man’s jacket was identical to the woman’s cat’s DNA.

Allard became involved with the dog DNA project in 2000, when he started working at the FBI Academy in Quantico, Va. After Sept. 11, the FBI dedicated the bulk of its funding to combating terrorism, and dog DNA research fell by the wayside.

Allard did not want to see the project end, so he applied for a grant from the National Institute of Justice, the research and development agency of the Department of Justice. For the past year, Allard has been working with a team of a few undergraduate students and one graduate student on the project. His grant will last until next year.

Junior Stephanie Carnation, who spent this past summer working in Allard’s lab with two graduate students, said the process of testing DNA samples is relatively easy.

“It is pretty simple once you get everything to work right,” she said.

But mDNA is not a sure-fire way of proving guilt because it is possible for two or more dogs – especially dogs from the same litter – to have the same mDNA. Allard is trying to calculate the likelihood that multiple dogs share a genetic code so that the information he provides investigators can be put in perspective.

James Starrs, professor of law and forensic sciences at GW, said that while multiple dogs might share the same mDNA, “reasonable” suspicion that dog hair matches a dog can make compelling evidence.

“It’s some evidence; it’s not certain evidence,” he said. “You don’t have to have certainty to testify in a court of law.”

Whether the database will have an impact on day-to-day criminal investigations depends on whether investigators become aware of the advances, said Walter Rowe, GW professor of forensic sciences.

“The tendency of investigators is to focus on the evidence they know they can use,” Rowe said.

Until scientists develop a way of consistently removing a more variable form of DNA from hair samples, Allard’s work could be quite useful, Rowe said.

“Depending upon the circumstances, it can be very helpful,” he said.

“The value of any kind of trace evidence is always dependent on the context.”

For example, if a live-in boyfriend suspected of murder has the victim’s dog’s mDNA on his jacket, it is of little value to investigators because the boyfriend would naturally spend time with the victim. If a stranger suspected of a crime has mDNA from the victim’s dog on his jacket, the evidence can be important.

A large part of Allard’s team’s task is to categorize dog families that share the same mDNA so that courtroom experts can determine how often a specific dog has the same genetic code as another dog.

“What we’re going to have to do is describe the variation we see,” Allard said.

Though forensic science is a relatively new area of interest for Allard, he has long been interested in dogs. For years Allard and his students have studied the evolution of dogs from a biological and anthropological perspective.

“I like dogs in general,” he said. “There’s a whole other story that isn’t forensic.”

Allard’s team is accepting more dog DNA samples, which are retrieved by gently brushing the inside of a dog’s cheek. Any dog owner willing to participate is asked to contact Kristen Webb: [email protected]