EUGENE, Ore. — Forensic DNA analysis is one of law enforcement’s strongest tools in identifying unknown crime suspects, but it has its limitations.
Researchers have found certain groups of people are at a higher risk of receiving a false positive result when it comes to identifying DNA in a multi-person mixed sample. For example, a door handle or a window might have the skin cells of many different people.
Researchers say that DNA analysis can involve a lot of uncertainty when it comes to certain groups of people from less diverse backgrounds.
Dr. Rori Rohlfs, a University of Oregon associate professor and data researcher who led San Francisco State University students on the study, said DNA analysis exists with variables and assumptions.
“People think of DNA like an oracle,” Dr. Rohlfs said. “Like it will have the truthful answer — but it’s just a piece of evidence like any other piece of evidence that might be misinterpreted or might have low accuracy in certain circumstances.”
Researchers offered an example of skin cells swabbed from a door handle. Investigators would most likely have to identify a lot of people to link a suspect to the handle.
“These mixtures of DNA are difficult to analyze,” she said. “There’s the opportunity for error or misunderstanding or misinterpreting the evidence.”
Researchers said some Indigenous populations, Latino or Pacific Islander groups may have lower levels of genetic diversity due to historic human demographic events. Based on the study, those with less diverse ancestry might be more likely to be incorrectly linked to a crime due to false positive results from mixed samples of DNA.
“The accuracy of these technologies really impacts people’s lives,” Dr. Rohlfs said. “Those people might be convicted based on these technologies, so it’s crucially important in our justice system to only use technologies that we feel very confident about.”
The study says forensic methods should be adjusted to consider the full range of genetic diversities. The more people in a DNA sample, the less accurate the results will be for investigators.
Dr. Rohlfs noted that both undergraduate and graduate students from San Francisco State University worked on the study.
Correction: In a previous version of this article, the last name of Dr. Rori Rohlfs was misspelled.