A new genetic test can reveal the hair color of unseen criminal suspects or unidentifiable victims.
The new analysis used a collection of recently discovered mutations linked to hair color, and it can predict the hue of an unknown person’s hair with about 80 to 90 percent accuracy.
“This could be really useful in cases with no eyewitnesses, where police have no idea who to look for at all,” said human geneticist Manfred Kayser at Erasmus Medical Center in the Netherlands, who led the study to be published in an upcoming issue of Human Genetics.
Forensic investigators extract DNA from blood, bones, spit, semen and other bodily stuff left at crime scenes, then check for matches in genetic databases. Until the past decade or so, unmatched samples weren’t immediately helpful in crime solving. Recent genetic research, however, has buoyed DNA’s forensic value by identifying markers linked to age, eye color, skin color and hair color.
Two types of melanin pigments, one light and one dark, control hair's basic shades. Most people produce only the dark melanin, resulting in black hair. But people of European descent can produce the lighter melanin and have red hair colors, or produce some combination of both for blond hair colors.
More subtle shade differences are controlled by a mutation called a single nucleotide polymorphism, or SNP, which is single change to one of 3 billion base pairs in the human genome. While many have no detectable effect on the body, some cause disease or result in visible changes such as hair coloration.
Such mutations are common in DNA but very stable, Kayser said, because the chance one will be erased is roughly one in 100 million generations. SNPs linked to hair color are even more stable because they're visible, and people tend to gravitate toward mates with specific hair colors.
Tracking down SNPs that affect physical appearance is akin to looking for needles in a haystack, but new large-scale genomic analyses have identified dozens of hair color markers.
To see if hair color could be predicted using 45 SNPs from 13 genes, Kayser and his team sampled DNA from 385 Polish volunteers and had dermatologists record their hair color. Their testing singled out 13 SNPs on 11 genes that could predict red and black hair colors with about 90 percent accuracy, as well as blond and brown colors with better than 80 percent accuracy.
“I don’t see this as going-to-court kind of evidence, but as an investigative lead it’s strong and could work in some scenarioses really well,” said forensic scientist Bruce Budowle of the University of North Texas at Fort Worth, who was not involved in the study.
In addition to helping describe mystery suspects, said Budowle, a former FBI senior scientist, the new test should help artists create better renderings of unidentifiable victims.
“A physical resemblance that’s strikingly good is much better than a more abstract one. Hair color may be all it takes to push someone over the edge,” Budowle said.
The new model’s accuracy could be improved by testing it on bigger and different populations. Finding more SNPs linked to hair coloration with genome-wide analyses may also help, but such efforts are expensive.
“Given the complexity of hair coloration and our sample size, I’m pretty amazed at how accurate it is,” Kayser said.
Image: Flickr/Jari Schroderus
See Also:
