Human odor is a complex mixture of volatile organic compounds (VOCs) that are emitted from the skin and breath. The composition and intensity of human odor can vary depending on factors such as age, diet, health, genetics, and environmental conditions. One of the most intriguing questions in the field of human odor research is whether it is possible to identify the gender of a person from their odor VOC composition. This article reviews the current state of knowledge on this topic and discusses the challenges and opportunities for future research.
The main sources of human odor are the eccrine, apocrine, and sebaceous glands, which produce sweat and sebum that contain various VOCs. These VOCs can be influenced by the activity of skin microorganisms, which metabolize some of the compounds and produce new ones. Some of the VOCs that have been associated with human odor include aldehydes, ketones, alcohols, esters, acids, hydrocarbons, terpenes, and sulfur-containing compounds. However, the exact composition and concentration of these VOCs in human odor are not well characterized, as they depend on many factors that are difficult to control or measure.
One of the main challenges in identifying the gender of a person from their odor VOC composition is the high variability and overlap of odor profiles between individuals and within the same individual over time. Moreover, there are no specific VOCs that are uniquely associated with either male or female odor, as both genders share many common compounds.
However, some studies have suggested that there may be subtle differences in the relative abundance or ratios of certain VOCs between male and female odor, such as higher levels of aldehydes, ketones, and acids in female odor and higher levels of hydrocarbons and terpenes in male odor. These differences may reflect the influence of sex hormones on the production and metabolism of VOCs by the glands and microorganisms.
Another challenge in identifying the gender of a person from their odor VOC composition is the lack of standardized methods and criteria for collecting, analyzing, and interpreting human odor samples. Different studies have used different techniques and instruments for sampling and measuring human odor VOCs, such as gas chromatography-mass spectrometry (GC-MS), proton transfer reaction-mass spectrometry (PTR-MS), or electronic noses.
These methods have different advantages and limitations in terms of sensitivity, specificity, accuracy, reproducibility, and cost. Furthermore, different studies have used different criteria for defining and classifying male and female odor, such as self-reported gender, hormonal status, menstrual cycle phase, or genetic markers. These criteria may not always be reliable or consistent across different populations and contexts.
In line with Locard’s Exchange Principle, perpetrators leave behind trace evidence during these interactions, including biological and inorganic material [1–3]. While fingerprints and DNA are commonly used for identification, they may be found in insufficient quantities, rendering them unusable. In such cases, human scent evidence, specifically hand odor, may still provide valuable forensic information [4, 5].
Human Odor and its Composition
Human odor is a complex mixture of volatile organic compounds (VOCs) emitted from the body, influenced by host genetics, environmental factors, and physiological secretions [6]. VOCs are organic compounds that are released into the environment as gases. The persistence of an individual’s odor is attributed to the shedding of the outer layer of the skin, leaving behind epithelial cells, sweat, oils, and glandular secretions [7, 8]. Various compound classes, such as acids, alcohols, aldehydes, hydrocarbons, esters, and ketones, contribute to human emanations [3].
The Composition of Human Odor
An individual’s odor consists of primary, secondary, and tertiary odors [3, 8, 9]. The primary odor, which is stable over time and distinct to an individual, is influenced by the Major Histocompatibility Complex (MHC), a polymorphic gene family. The MHC may produce molecules found in sweat or bind to specific peptides that contribute to the volatile metabolites responsible for skin odor VOCs. Additionally, the MHC may have a direct influence on the microbial biota present on the individual’s skin, further contributing to primary odor [9].
Microbial diversity, influenced by genetics, also plays a role in primary odor. The breakdown of non-volatile chemicals by the skin’s microbiota produces volatile molecules that contribute to the distinctive human “scent” [10]. In addition to primary odor, the skin’s multi-layer composition and the physiological secretions from various glands contribute to the variable and endogenous secondary odor. Exogenous compounds, such as non-resident bacteria, cosmetic products, soaps, and perfumes, further contribute to the tertiary odor, which has the highest variability [11].
The Role of Canines and Analytical Techniques
Well-trained canines (Canis familiaris) have been used as specialized detectors to identify personal human odor and other relevant chemicals [12–14]. However, laboratory-based subject identification using analytical instruments has been challenging due to the lack of robust datasets and developed techniques.
The Efficacy of VOC Profiles for Gender Discrimination
This study analyzed hand VOC odor profiles from 60 self-identifying male/female participants using supervised multivariate regression models for gender classification. Partial least squares-discriminant analysis (PLS-DA) revealed the ability to cluster female and male subjects but displayed lesser discrimination of gender in the 2D model.
The addition of a thirdcomponent improved gender classification and discrimination in the 3D model compared to the 2D model. Orthogonal projections latent structures-discriminant analysis (OPLS-DA) and linear discriminant analysis (LDA) demonstrated the highest discrimination and classification of subject gender, with non-intersecting confidence level ellipses.
Stacked bar graphs showed individualization of VOC profiles but lacked additional information, such as gender, when donor characteristics were not known. The LDA model achieved a 96.67% accuracy rate for both male and female samples during cross-validation. The statistical workflow employed in this study provides a potential tool for standardized VOC comparisons and can be applied to various forensic datasets, even in the absence of other discriminatory evidence like DNA.
Conclusion
The analysis of hand VOC odor profiles using supervised multivariate regression models has shown promising results in gender classification and discrimination. The use of PLS-DA, OPLS-DA, and LDA models allowed for the identification of characteristic VOC profiles and accurate gender prediction based on hand odor samples. Further research is needed to explore the feature selection provided by VIP scores and loading plots, which can aid in compound discrimination and prediction of other donor characteristics, such as ethnicity/race and age.
The statistical workflow described in this study can be applied to various forensic scenarios and serves as a foundation for the development of standardized VOC comparisons. With ongoing advancements in analytical techniques and the accumulation of robust datasets, the forensic potential of human hand odor may be further harnessed for investigative purposes.
reference link : https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0286452#sec016
