Forensics look to Pinpoint Gender and Ethnicity from Hair

Imagine this scenario: Early in the morning employees at a small pharmacy call the police to report a robbery that took place overnight. It must have been an inside job, owners determined, since there was no sign of a break-in or struggle, and the suspect was able to access the register. No fingerprints could be detected at the scene.

However, despite great care and planning, this person was not in fact “gone without a trace.” Police are able to collect a tiny spec of hair near the register. With just this one piece of hair they can deduce the perpetrator was female and Caucasian—narrowing down the search considerably.

We’re almost there. Lily Huang, a four-year Master’s student at Queen’s University in Kingston, Ontario, is on the job. For her thesis, Huang wanted to do something in the field of forensic analysis and she had the idea to analyze hair for gender and for ethnicity.

So how can so much detailed information be extracted from such a tiny human sample? “The idea came to me…when it was shown that linear discriminant analysis (LDA) could be done on car paints to [find matches]. As I looked more into the literature, many forensic journal articles included using LDA for matching other items, such as recreational drugs. And with the support of Chemistry Professor Dr. Diane Beauchemin, we thought it would be great to perform tests to see whether you can match people based on major, minor, or trace elemental composition in head hair,” says Huang.

Hair has always been valuable to investigators. It is one of the stable “organs” of the human body and can be used to test for things like diet and environment. Huang points out human hair can absorb elements by internal means through sweat and protein growth, which can be affected by a person’s genetics. These trace amounts, then, can be detected by highly sensitive instruments and provide valuable information about a person.

Positive ID in seconds

In the lab, researches have a small graphite furnace into which they put a little bit—about two milligrams—of hair and then burn it. The sample then goes into an instrument called an inductively coupled plasma (ICP). “Basically we do multi element analysis of the hair and we don’t have to quantify the quantities of the elements, we just have to look at the relative distribution of them and that serves as a ‘fingerprint’,” says Beauchemin.

The method takes very little time—about 85 seconds, but Beauchemin and Huang aim to cut the time down even further and get results even faster.

Throughout her research Huang collected samples from people in the department and began testing her hypothesis. She soon learned she could distinguish between Caucasian, East Asian and Central Asian races. Beauchemin indicates this is just the beginning. She says they are looking to be able to identify more cultures, in addition to gender. “For gender, [we look for] four elements: magnesium, strontium, sulfur and zinc. Those are known to be accumulated to different extents between males and females.

“[This process] requires more elements to distinguish ethnicities, about nine—molybdenum, sulfur, strontium, chromium, potassium, nickel and zinc. We’re trying to refine that, so I suspect the list of elements will have to be longer to be able to say, ‘this person was Korean.’”

Already surprised by accuracy, Beauchemin says the results have been amazing. “We had something like 13 samples from people at Queens and we got two blind samples from Montreal and Sudbury, 300 km away, and they were correctly assigned by this method.” Beauchemin gave a talk March 9 to the Forensic Science Center in Toronto.

Unlike rapid DNA programs that match suspects and persons to DNA housed in a database, this method could help an investigator when he or she just doesn’t have much to go off of. In this case the tiny particle of hair is creating a profile of a complete unknown. This new information is sketching a picture that didn’t exist before.

“Although DNA analysis is still the best way for identification and is arguably 100 percent effective, it is not always readily available,” says Huang. “This work aims to provide a tool that can help investigators narrow down the number of suspects where DNA analysis is possible, or when there is not a DNA match”.

The study has captured much attention for a Master’s thesis project. Indeed, Beauchemin says people are sending her hair and beard samples—from as far away as Afghanistan—in the mail for testing…people who want to contribute to the study and play a game of “tell me who I am.”

Huang says she would like to continue in the forensic vein with her research. “I’m hoping that after I graduate, I can work for law enforcement and help solve crimes, whether through identification of a person, automobile or drugs, using my experience on developing methods based on ICP instruments.”

Editor’s Note: The paper can be purchased through pubs.rsc.org.