True Tales of Canine CSI
Investigators turn to dog DNA to convict criminals and acquit the innocent.
D. Caroline Coile, Ph.D.
Page 3 of 4
Halverson has since founded QuestGen Forensics for analyzing non-human DNA. Now, as QuestGens director and senior scientist, Halverson analyzes eight to 10 cases a year, mostly involving dogs. Sometimes the evidence is enough to convince the subject a plea bargain is a better idea, but about 20 percent of those cases end up with Halverson testifying in trials. QuestGen and the Veterinary Genetics Lab at the University of California, Davis (which created a separate forensics division in 2000), both located in Davis, Calif., are considered the forerunners in this growing field.
Both laboratories make use of two types of DNA analyses for canine forensics. The first uses Short Tandem Repeat (STR, or microsatellite) markers, which is the same technique adopted by the AKC and other registries for parentage verification. A complete STR is virtually unique for each dog, discriminating even between closely related dogs. Using the standard 10 STR markers, the probability of the sample matching a random dog in the population is greater than one in one million. Such probabilities can be computed because of a large database that has been collected and analyzed for the degree of genetic relatedness between individual dogs.
The problem with STR analysis is that it requires DNA from cell nuclei, most often from blood, saliva, and hair roots. With only one copy per cell, an adequate sample in good condition is needed and that's not always the case. In fact, because most shed dog hair does not contain the hair root, STR analysis may not be possible with even mounds of hair.
That's when the second type of DNA analysis comes into play. Mitochondrial DNA (mtDNA) is present in non-nuclear parts of the cell, and has hundreds of thousands of copies in each cell. Because of this, it can be extracted from the hair shaft, as well as from degraded blood and tissue samples. Unfortunately, mtDNA does not provide the unique identity that STR analysis does because mtDNA is inherited only from the mother. Many dogs share the same maternal line and so have the same mtDNA. A region of mtDNA that is particularly variable, called the hypervariable region, is used to categorize dogs into different haplotypes. The mtDNA database is much smaller than the STR database, but demonstrates that some haplotypes are far less common than others, with the most common haplotype occurring in 18 percent of dogs. Although mtDNA matches are not as statistically compelling as STR matches, they can be significant and can also definitely exclude a dog as a sample donor.
Some skeptics worry that dog DNA forensics cannot finger culprits with the precision that human DNA can, because dogs tend to be more inbred. But because of the large number of DNA markers used, that is seldom a problem. In addition, statistical methods factor in how common a certain haplotype is or how inbred or common the suspect dogs breed tends to be. Although animal DNA was sometimes ruled inadmissible in earlier court cases, it is now usually admitted as solid evidence.Page 1 | 2 | 3 | 4
Give us your opinion on True Tales of Canine CSI
Login to get points for commenting or write your comment below
Get New Captcha