When DNA Fails, Proteomics Untangles Crime Scene Hair Evidence
Protein biomarkers in hair enabled accurate species identification without relying on DNA evidence.
Imagine standing over a cold case file, staring at a crucial strand of crime-scene hair, only for the lab to tell you traditional DNA barcoding is impossible.
To fix this investigative problem, researchers at Edith Cowan University and the University of California, Davis, turned to forensic proteomics to read the unique protein “barcodes” that can be found inside the hair shaft instead.
Their new method successfully bypassed the need for DNA altogether, identifying the exact human or animal species for 100% of the trace samples tested.
The rise of forensic proteomics
Hair fibers are among the most common forms of biological trace evidence recovered from crime scenes; they are chemically stable, easily transferred, and highly persistent in the environment.
Since humans often live closely with domestic pets and livestock, hair found at a scene can easily be human or nonhuman. Identifying these fibers helps investigators prioritize evidence and reconstruct events.
Forensic scientists currently rely on DNA barcoding for molecular confirmation, but accessible DNA is frequently missing from rootless, aged, or degraded hair shafts, leading investigators to often fall back on microscopic hair comparison.
“Traditional methods of classification, based on hair morphology, require expert interpretation, offer limited taxonomic resolution, and are not easily scalable,” said the authors. “Mass spectrometry (MS)-based proteomics provides a biomolecular alternative.”
While MS has successfully identified species in archaeology and paleontology, these methodologies rely on large sample sizes. In comparison, modern casework requires a highly sensitive approach optimized for trace samples.
The new study aimed to develop a robust workflow to discover and characterize taxonomically diagnostic hair peptides across 15 forensically relevant mammalian species. The authors aimed to integrate these markers into a streamlined pipeline capable of reliably identifying species from trace-level, single-hair shafts without requiring any prior source information.
New forensic proteomics workflow delivers rapid species resolution
The paper introduced a two-phase framework: an initial biomarker discovery phase using full reference proteomes, followed by a sample processing phase integrated into a usable workflow. Reference fur/hair samples were collected from humans and 14 animal species common to human environments, including dogs, cats, mice, rats, foxes, and livestock.
The team optimized a method to grind down small hair fragments—measuring 10–30 mm long—using a specialized blending mill. They used standard enzymes to break down the hair proteins, allowing a nano-liquid chromatography tandem mass spectrometer to read the molecular sequences like a barcode.
Instead of searching entire, massive proteomes, the authors built a custom, curated hair protein database containing 1,578 keratin and keratin-associated protein entries from the target species.
The workflow identified a panel of 226 taxonomically informative peptide biomarkers, 207 of which are entirely novel to this study. Within this panel, 59 peptides offered single-species resolution, while 7 human-specific peptides were confirmed to be entirely exclusive to the genus Homo.
The researchers also identified three taxonomically invariant mammalian peptides that appear across all studied species, which were added to the peptide panel to be used as internal controls.
“The method was able to classify all hairs analyzed with genus- or species-level resolution,” said the authors.
When validated against 135 hair extracts from multiple individuals and species, the assay achieved a 97% mean peptide recovery rate and successfully classified 100% of the samples. Using a broad panel of markers ensured the test remained accurate even if an animal's unique marker was not detected.
The validation trials also yielded a true-positive rate of over 90%, meeting the rigid accuracy standards needed for legal casework.
Importantly, the custom database also cut bioinformatic data processing time by 90%.
The future and technical limits of forensic proteomics
The proposed workflow effectively unlocks investigative data from hair evidence that would otherwise have failed traditional DNA profiling.
Accelerate your PROTACs drug discovery and development
This eBook shows how an integrated, end-to-end analytical approach helps labs generate reliable, regulatory-ready PROTAC data at every stage of development.View eBook
“The developed assay offers an alternative approach to the source identification of trace hair samples in a forensic setting, which is objective, reproducible, and robust,” said the authors. “Employing technology that is widely accepted for providing molecular based identification promotes ease of integration into current forensic science laboratories.”
However, the authors noted a few limitations. As public protein databases are incomplete, a peptide that appears unique now might undergo signal erosion and lose its single-species specificity as more non-model-organism proteomes are published; panels will require periodic reevaluation.
Signal erosion
Signal erosion occurs when a protein marker that initially seems unique to one specific animal turns out to be shared by others. As scientists map the genetics and proteomes of more wild or rare animals over time, the exclusivity of a marker's signal may be reduced, meaning it can no longer uniquely identify just one species.
The high sensitivity of MS instruments also makes them vulnerable to environmental contamination, which demands strict quality assurance protocols. Extremely close evolutionary relatives, such as red foxes and arctic foxes, also share identical peptide markers, meaning the method must occasionally be paired with local ecological context for definitive answers.
Future research directions should include testing the performance of the biomarker panel under harsh environmental conditions, such as on degraded, aged, or chemically treated hair. The team also plans to expand interindividual datasets to ensure the chosen biomarkers stand up against widespread natural genetic variation within species.
“A proteomic approach to source identification removes the subjectivity that plagues traditional methods of hair analysis,” the authors concluded.
Migrating the discovered biomarkers into a targeted forensic proteomics workflow will help build high-throughput screening assays for operational casework, leaving fewer detectives standing over a cold case file, staring at a crucial yet unusable strand of crime-scene hair.
Reference: Keane RE,
Gummer JPA, Tidy RJ, Parker GJ. Taxonomic source identification of hair. J
Proteome Res. 2026;25(6):2626-2637. doi: 10.1021/acs.jproteome.5c00638
Add Technology Networks as a preferred Google source to see more of our trusted coverage.
