Over the past twenty years, proteomics has not only improved in general, but also found applications in forensic science. Despite the fact that a golden standard has already been developed in DNA and RNA analyses, there is still a lot more information to be gained, write Shirin Alex, Marcel de Puit and their colleagues from the NFI in a review in Science & Justice. Alex is also the first author of two research papers that examine specific applications of proteomics in forensics.  

‘The idea behind the review was to see how proteomics in forensics has grown over the last two decades’, says Alex, a PhD student at the NFI. ‘Initially, it was mainly used for identifying body fluids, but since then, many other applications have emerged.’ However, proteomics has not become the go-to technique, primarily due to the well-established techniques already in place. ‘You can use proteomics to identify donors based on their biological sex. However, DNA analysis generally performs this task much more effectively. What we wanted to demonstrate is that you can obtain much more information in cases where there is not yet a golden standard.’ This allows you to go beyond the question of who it came from and start looking at what exactly took place and when.  

Hair and eye colour

Alex and her colleagues developed a proof of concept to assist with identification when dealing with mixtures of body fluids, for example in cases of sexual assault. ‘It is difficult to say who contributed to which fluid in a case of mixtures with just DNA analysis’, explains Alex. ‘But with proteomics, it’s possible. We looked at proteins, specifically single amino acid variants, which are highly specific to an individual.’ The team hypothesised that these SAAVs could be used to determine whether a fluid came from person A or person B. ‘It’s not perfect, but the proof of concept shows that SAAVs can be used for source attribution.’ 

The team were also curious to see if proteomics could be used to ascertain phenotypic features such as hair and eye colour and height from whole blood samples. ‘That turned out to be too difficult, so we opted for a simpler approach to determine the sex of an individual’, says Alex. The research generated a huge amount of data, so they used a machine learning model to analyse it. ‘For known or clean samples, the model works really well. However, when you have mock samples that are purposefully contaminated, the model struggles.’ She concludes that phenotypic traits can be obtained, but DNA is much more effective. Proteomics is more valuable when applied to source attribution or determining the timeline of events. 

Mismatch

The research sounds promising, yet the papers’ conclusions are very cautious. Alex explains: ‘When we wrote the grant proposal, we had high expectations because I genuinely thought it would be much easier to use proteomics in forensics. The technique itself is well established and the protocols are available, so it shouldn’t be that difficult to apply it to a different research goal.’  

However, in forensics, there are many unknowns, and it is difficult to design an experiment that considers everything. ‘There’s always a scenario you didn’t anticipate. The complexity of forensic science really took me by surprise’, Alex laments. 

There was also a lot of scepticism from people in the field who use the established techniques. ‘I think the mismatch in expectations was caused by the desire in forensics to use techniques immediately, whereas our research is still in the developmental, fundamental stage’, says Alex. ‘The question we wanted to answer was whether there is enough application potential for proteomics in forensic science.’ 

Alex answers in the affirmative, but much more research needs to be done before proteomics can be used in routine investigations. ‘I think it is more suitable for special cases and specific questions, such as how many donors were present and which bodily fluid came from which donor. But I’m not sure if people are willing to change their standardised routine protocols.’ This is because the standard procedure involves removing all proteins in order to obtain as much DNA and RNA as possible, so A new protocol that preserves proteins while enabling simultaneous extraction of DNA, RNA, and proteins would require considerable effort. ‘I’m convinced it would be worth it in the end, but I’m not sure who will take this research forward.’  

Successful application

An example of the power of proteomics in forensic science was reported in 2019 in the Journal of Proteomics. ‘In the papers and in my thesis, we are being very cautious. But it was successfully applied in a rape case.’ After a night out where a woman had drunk too much, she went home with a man in his car. The next day, she filed a report saying that she had been raped after losing consciousness. The man contradicted her, saying it had been consensual. She said that she had passed out after vomiting in the car. Although the man had thoroughly cleaned his car, investigators were able to find traces of the previous night. DNA was collected, but it did not provide enough information to corroborate either account. However, proteomics analysis revealed the presence of partially digested proteins, which provided further evidence that she had indeed vomited, corroborating her account. ‘Though it doesn’t provide conclusive evidence, proteomics did provide more context to the scenario.’  

Overall, it was a really challenging project. Alex: ‘I didn’t always get results when I followed established protocols, and the project was also spread out over multiple institutions. But in the end, it was really rewarding when everything came together. The field of forensics is already small, and forensic proteomics is an even smaller community within it. However, they are very welcoming and love to brainstorm and discuss the latest research. As a first-year PhD student, it was difficult to reach out and find my place, but it became much easier to connect after a while.’ 

Alex, S. et al. (2025) Journal of Proteome Research 24(10), DOI: 10.1021/acs.jproteome.5c00098

Alex, S. et al. (2025) Science & Justice 65(6), DOI: 10.1016/j.scijus.2025.101320

Shehata, T.P., Alex, S. et al. (2025) Forensic Science International: Genetics 81, DOI: 10.1016/j.fsigen.2025.103343

Pieri, M. et al. (2019) J. Proteomics 209, DOI: 10.1016/J. JPROT.2019.103524