Journal of Forensic Investigation
Research Article
Comparative Evaluation of Salivary Proteomic Profiling and STR DNA Typing for Forensic Subject Discrimination
Doan H1, Hogan C1, Viray J2 and Giulivi C1,3*
1Department of Molecular Biosciences, School of Veterinary Medicine, Davis, CA 95616,
2Sacramento District Attorney’s Office, Biology Laboratory, Sacramento, CA 95814
3Medical Investigations of Neurodevelopmental Disorders (M.I.N.D.)
Institute, University of California Davis, CA 95817
2Sacramento District Attorney’s Office, Biology Laboratory, Sacramento, CA 95814
3Medical Investigations of Neurodevelopmental Disorders (M.I.N.D.)
Institute, University of California Davis, CA 95817
*Address for Correspondence:Cecilia Giulivi, Department of Molecular Biosciences, School of
Veterinary Medicine, 1089 Veterinary Medicine Drive, Davis, CA, USA, E-mail: cgiulivi@ucdavis.edu
Submission:18 March, 2026
Accepted:29 May, 2026
Published:03 June, 2026
Copyright: ©2026 Doan H, et al. This is an open access article
distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium,
provided the original work is properly cited.
Keywords:Forensic identification; short tandem repeat (STR) typing; DNA
analysis; salivary proteomics; liquid chromatography–tandem mass
spectrometry (LC–MS/MS); principal component analysis (PCA); forensic
biology; Random Match Probability (RMP)
Abstract
Identifying persons of interest from biological evidence is central
to establishing probative value in forensic investigations. Short tandem
repeat (STR) DNA typing remains the forensic gold standard due to
its statistical robustness and high inter-individual discriminatory power.
However, low-template DNA, degradation, and complex mixtures
may limit profile clarity. In contrast, forensic proteomics leverages
the relative abundance and chemical stability of proteins and may
provide complementary biological information.
This study directly compared subject discrimination using STR genotyping and salivary proteomic profiling. Saliva samples from 41 individuals (including three samples from the same individual collected across ~2 years) were analyzed by LC–MS/MS, and protein abundance profiles were evaluated using unsupervised principal component analysis (PCA) with Euclidean distance metrics. DNA genotyping was performed on 16 samples using real-time qPCR quantification, PowerPlex ® Fusion STR amplification, and capillary electrophoresis.
Full STR profiles were obtained for all DNA samples, and identical alleles across all loci confirmed perpetrator identity through populationbased Random Match Probability (RMP) calculations. Proteomic PCA of 269 shared proteins explained 20.1% (PC1) and 15.0% (PC2) of total variance; reduction to 10 discriminatory proteins increased explained variance to 27.6% and 20.2%, respectively. Perpetrator-derived proteomic samples formed significantly tighter clusters than unrelated individuals (Mann–Whitney U = 0, p = 1.88 × 10-4). Notably, samples collected approximately two years apart remained more similar to each other than to any other subject in multivariate space.
While STR typing provides deterministic identification through locusby- locus allele concordance, proteomic profiling captures continuous, biologically dynamic variation. These findings support the concept that salivary proteomics may serve as a complementary forensic tool, particularly when DNA quality or quantity is compromised, though further validation and the development of a statistical framework are required before evidentiary implementation.
This study directly compared subject discrimination using STR genotyping and salivary proteomic profiling. Saliva samples from 41 individuals (including three samples from the same individual collected across ~2 years) were analyzed by LC–MS/MS, and protein abundance profiles were evaluated using unsupervised principal component analysis (PCA) with Euclidean distance metrics. DNA genotyping was performed on 16 samples using real-time qPCR quantification, PowerPlex ® Fusion STR amplification, and capillary electrophoresis.
Full STR profiles were obtained for all DNA samples, and identical alleles across all loci confirmed perpetrator identity through populationbased Random Match Probability (RMP) calculations. Proteomic PCA of 269 shared proteins explained 20.1% (PC1) and 15.0% (PC2) of total variance; reduction to 10 discriminatory proteins increased explained variance to 27.6% and 20.2%, respectively. Perpetrator-derived proteomic samples formed significantly tighter clusters than unrelated individuals (Mann–Whitney U = 0, p = 1.88 × 10-4). Notably, samples collected approximately two years apart remained more similar to each other than to any other subject in multivariate space.
While STR typing provides deterministic identification through locusby- locus allele concordance, proteomic profiling captures continuous, biologically dynamic variation. These findings support the concept that salivary proteomics may serve as a complementary forensic tool, particularly when DNA quality or quantity is compromised, though further validation and the development of a statistical framework are required before evidentiary implementation.