Document Type



Doctor of Philosophy (PhD)



First Advisor's Name

Kenneth Furton

First Advisor's Committee Title

Committee chair

Second Advisor's Name

DeEtta Mills

Second Advisor's Committee Title

Committee Member

Third Advisor's Name

Yong Cai

Third Advisor's Committee Title

Committee Member

Fourth Advisor's Name

Francisco Fernandez-Lima

Fourth Advisor's Committee Title

Committee Member

Fifth Advisor's Name

Konstantinos Kavallieratos


human scent, Human Leukocyte Antigen (HLA), forensic chemistry

Date of Defense



Human scent has been previously defined as a complex mixture of volatile organic compounds (VOCs) detected in the headspace above a scent sample. Humans generate odor from several areas of the body including hair, mouth, hand, axillae, and foot. Due to the novelty of human scent research, human scent evidence has been undervalued in the court of law. However, this type of evidence has significant value when physical evidence is not available at crime scenes. To increase the individualization and differentiation power of human scent evidence, this study aimed to further investigate the identification of chemical signatures within axillae of specific ethnicities (Caucasian, Hispanic, and African American) and determine if ethnic-specific genetic signatures are present among Human Leukocyte Antigen (HLA) genes.

During the study, the axillae of 68 participants were investigated. Upon collection, samples were extracted using Headspace Solid Phase Micro extraction (HS-SPME) and solvent extraction. The samples were analyzed using Gas Chromatography- Mass Spectrometry (GC-MS). The utilization of SPME immediately followed by solvent extraction complements the extraction of both semi-volatile and non-volatile compounds, thus filling in the gaps of the compounds that could not be recovered using HS-SPME alone.

The samples were evaluated statistically via logistic regression and Receiving Operating Characteristic (ROC) curves to evaluate the performance and prediction power of VOCs for ethnicity inferences. The study concluded that logistic regression served as an efficient model predicting the VOCs capable of class characteristic determination when comparing ethnicities.

The HLA gene complex was evaluated to determine its contribution to human scent and the ability to differentiate between ethnicities. Using buccal swabs extracted from 31 subjects, five genes were successfully amplified using Multiplex Polymerase Chain Reaction (Multiplex PCR). The Multiplex PCR products were analyzed using capillary electrophoresis. The genotype frequencies were observed, and linear discriminant analysis (LDA) was performed to assess the ability of predicting ethnicity using genotype frequencies of individuals. Four of the five genes predicted ethnicity at 80% or greater accuracy, which validates that the HLA genes (D6S2925, D6S2937, D6S2917, and D6S2787), coupled to the VOCs, can be used as a biomarker for class characteristic determination of an individual.





Rights Statement

Rights Statement

In Copyright. URI:
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).