NEW YORK (GenomeWeb) – A research team has identified proteins in hair that can be used instead of DNA to identify individual humans.
Because DNA is subject to environmental and chemical degradation that limits its usefulness over time, archeologists, forensic scientists, and others have looked for other ways to identify individual human beings. Proteins are more stable and also demonstrate variations in single amino acid polymorphisms which translate to non-synonymous single nucleotide polymorphisms (nsSNPs) present in the genome.
"Based on exome analysis, there are over 35,000 nsSNPs with genotype frequencies greater than 0.8 [percent] in the European-American (EA) populations," the researchers wrote in their paper, published today in PLOS One. Using this information, they theorized that proteins would be a more effective way to identify humans.
To test the theory, the team collected cranial hair shafts from 60 self-identifying unrelated European Americans, five African Americans, five Kenyans, and six individuals from two separate archeological sites. Then they processed the samples to create a peptide mixture, analyzed it with tandem liquid chromatography mass spectrometry, and identified proteins using Matrix Science's Mascot software, X!Tandem using the GPM manager software, and X!Tandem using the Petunia graphic user interface. They further validated the findings with Sanger sequencing.
The researchers were able to demonstrate the robustness of hair proteins in all the samples, including the six 250-year-old archaeological hair samples. To date, they have identified 185 hair protein markers which they believe could be sufficient to provide a pattern that could distinguish an individual among a population of 1 million people.
"Across all samples, the total number of peptides detected ranged from 376 to 18,563 and yields of unique peptide spectral matches ranged from 156 to 2,011," the researchers wrote. "A total of 608 imputed genotype determinations were made, of which 596 were true positives that were confirmed with DNA sequences and 12 were false positives." They also noted that only five of the tested peptides had positive predictive values that were under 100 percent, indicating a high specificity for protein-based identification methods, though sensitivity did range widely.
Additionally, the investigators found that profile probabilities were lower in the subset of individuals of African ancestry compared with the samples from the EA population. They also noted similar trends in protein distribution in the archeological samples, further indicating that positive profile probabilities were lower in African populations.
In their paper, the researchers noted that there is still a need to refine the method, test it in a more diverse sample population, and better refine the software that was used in this initial experiment. Such changes could lead to the definitive identification of a core set of approximately 100 protein markers that could distinguish an individual among the world's population using a single hair.
"We are in a very similar place with protein-based identification to where DNA profiling was during the early days of its development," Brad Hart, director of the Forensic Science Center at Lawrence Livermore National Labs and co-author of the paper, said in a statement. "This method will be a game-changer for forensics, and while we've made a lot of progress toward proving it, there are steps to go before this new technique will be able to reach its full potential."
The custom protein reference database created by the team has been made publicly available through the Global Proteome Machine.