|
We all know that dogs can distinguish between humans from their natural scents, because they can track people down after smelling an object, such as clothing, which carries their scent. The characteristic scent varies from person to person and is stable over long periods of time, allowing the dogs to do their work. Their sense of smell is better than ours and we cannot match their performance. However, there are other ways of human scent discrimination and Ken Furton from the International Forensic Research Institute at Florida International University is at the forefront of developing a lab.-based method. The Furton group has studied various aspects of human scent, including identifying the compounds responsible and their variation between individuals. Recently, the group used SPME with GC/MS to analyse volatile organic compounds in the headspace above swabs of human hands. They showed that there is a high degree of variability between individual scents and that human odour should be determined on an individual basis for use in a forensic context. So, where does this unique scent originate from? It is generally believed that the set of genes known as the major histocompatibility complex (MHC) is at least partly responsible. The MHC varies by as much as 8.6% between individuals and changes in a single gene within the MHC have been shown to alter the scents of mice. One of the mechanisms responsible may be the action of microbial flora, the compositions of which are fashioned by the MHC. The bacteria attack dead skin cells and glandular secretions close to the human body surface, where the bacterial population is about five-fold greater than in the air further away from the person. Eccrine glands, which are found all over the body, especially on the soles of the feet and the palms of the hands, and sebaceous glands, found where hair is present, are the main glandular contributors. The odorous compounds which make up the scent of an individual have been classified by Furton into three groups. Primary odour compounds derive from within the body and are not influenced by diet or environmental factors. Secondary compounds are present due to dietary or environmental factors and tertiary odour compounds are due to the external application of lotions, creams, soaps and the like. Following on from their previous work, Furton, with co-reserachers Allison Curran and Paola Prada, has examined the ability of the primary odour compounds to provide a biometric measure of an individual, which he has referred to as a barcode. The hands from five male and five female volunteers were swabbed with gauze pads under controlled conditions and the VOCs present were analysed by SPME-GC/MS with electron ionisation and library matching. Each person was sampled three times. A total of 37 reported human VOCs were found, including acids, aldehydes, alcohols, alkanes, ketones and esters. By comparison with their previous work, the team identified six of these as high-frequency occurring VOCs and ten as medium frequency. The compounds were ranked into arrays according to their peak areas, equivalent to their abundances, and compared with the Spearman correlation. This approach had been demonstrated previously to be a valid method of data handling. Setting a correlation threshold of 0.9 allowed 88.05% of the cases to be correctly identified while a threshold of 0.8 identified 89.66%. These analyses produced a high number of errors, so the researchers narrowed down the comparison to the primary odour compounds alone, of which 24 were identified across the 10 subjects. This led to correct discrimination of individuals in 99.54% of the cases, at a correlation threshold of 0.9 or 0.8, and 100% for a correlation threshold of 0.7. An alternative discriminatory mechanism involved a comparison of the 24 primary odour compounds against a library of the same 24 VOCs compiled from 52 subjects. Again using the Spearman rank correlation, 99.34% of the individuals were identified. The relative ratios of the primary odour VOCs must vary sufficiently from person to person to produce individual and distinguishable scent profiles, although the results should be validated using a larger number of people. These scent "barcodes" can be used as a biometric measure for the exclusive identification of people, in the same way that fingerprints and retinal prints are unique.
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.
|
|
