Defense Date
2026
Document Type
Directed Research Project
First Advisor
Dr. Christopher Ehrhardt
Second Advisor
Dr. Sarah Seashols-Williams
Third Advisor
Dr. Susan Greenspoon
Abstract
While DNA profiling of sexual assault evidence can identify biological contributors, determining the tissue source of recovered profiles remains a significant challenge, particularly in cases of suspected oral assault involving mixtures of epithelial cells. The initial goal of this study was to identify non-genetic signatures for differentiating two types of epithelial cells relevant to oral assaults, saliva epithelial cells and epidermal skin cells. To develop tissue-specific signatures, reference cell populations from saliva and epidermal skin surfaces were collected from 20 donors and analyzed using full spectrum flow cytometry.
Results from single-source reference samples showed that saliva-derived particle populations exhibited higher forward scatter and side scatter values (above 106), compared to epidermal populations, demonstrating systematic differences in particle size between these two tissue sources. Additionally, saliva and epidermal populations showed distinct differences in the intensity of autofluorescence in three emission channels: B1 (498-518 nm), UV12 (678-706 nm), and V1 (420-435 nm). From this, cell-source specific fluorescence criteria were created to identify and quantify particles unique to saliva in both single-source and mixed-tissue samples.
In the second part of the study, cell type-specific autofluorescence signatures were used to isolate saliva and epidermal cell fractions from mixture samples using fluorescence-activated cell sorting (FACS). A tiered system of signatures was developed to classify and sort particles, which included an initial gate whereby particles were sorted according to their intensities in FSC and UV12 channels followed by a second gate that sorted particles by their intensities within the V1 and B1 channels. The unsorted mixtures and sorted fractions were then subjected to DNA extraction and quantitation to assess the degree of enrichment in the sorted saliva and skin fractions. One mixture sample showed evidence that contributor cell populations were enriched, but DNA yields were lower than expected based on the number of particles sorted (e.g., < 50pg). Subsequent modifications to the sorting criteria increased the number of saliva cells isolated from the original mixture (2-450 fold higher saliva counts) but did not lead to proportional increases in DNA yields. This indicates that DNA recovery from these types of sample mixtures may continue to be a challenge and require further optimization. However, even without sorting, the autofluorescence signatures themselves could still provide probative information regarding source tissue in unknown mixture samples comprising saliva and epidermal skin cells.
Rights
© The Author(s)
Is Part Of
VCU Master of Science in Forensic Science Directed Research Projects
Date of Submission
5-11-2026