How researchers are using DNA to create images of people's faces
Researchers took an ethnically diverse sample of more than 1,000 people of different ages and sequenced their genomes to come up with an image of what they look like. (PNAS journal)
Published Tuesday, October 2, 2018 4:45PM EDT
Last Updated Tuesday, October 2, 2018 5:37PM EDT
Advancements in facial recognition and DNA sequencing technology have allowed scientists to create a portrait of a person based on their genetic information.
The leap could have applications in criminal investigations, by sketching out the face of a perpetrator based on genetic material left at the scene such as blood or other bodily fluids.
Researchers have already produced images of faces based on genetic material or genome.
A study published last year and co-authored by biologist Craig Venter, CEO of San Diego-based company Human Longevity, showed how the technology works.
The research team took an ethnically diverse sample of more than 1,000 people of different ages and sequenced their genomes.
They also took high-resolution, 3D images of their faces and measured their eye and skin color, age, height and weight.
This information was used to develop an algorithm capable of working out what people would look like on the basis of their genes.
Applying this algorithm to unknown genomes, the team was able to generate images that could be matched to real photos for eight out of ten people.
The success rate fell to five out of ten when the test was restricted to those of a single race, which narrows facial differences.
In the study, Venter wrote: “Individually, for a large fraction of the traits, their predictive accuracy beyond ancestry and demographic information is limited.
“However, we have developed a maximum entropy algorithm that integrates multiple predictions to determine which genomic samples and phenotype measurements originate from the same person.”
Venter said the algorithm predicted genetically simple traits such as eye color, skin color and sex at high accuracy.
It also built models to predict voice, age, height, weight and body mass index.
But critics of the research say the study seemed to predict average faces based on sex and ancestry, rather than specific faces of individuals.
The development could also be applied to identifying victims who have been burned or maimed. Unsolved criminal cases might be reopened if suitable genetic samples are still available.
The authors of the paper said the research has ‘significant ethical and legal implications on personal privacy, the adequacy of informed consent, the potential for police profiling and more’.
Earlier this year, investigators in Washington State unveiled an image of a suspect created from DNA in the 30-year-old murder case of young Victoria-area couple Tanya Van Cuylenborg, 18, and Jay Cook, 20.
The police used technology called snapshot DNA phenotyping that can determine a person's eye, skin and hair color, facial features and ancestry to create mugshots of what the suspect may have looked like at 25, 45 and 65 years old.
The profile did not account for weight, facial hair or other features like scars.
And in Calgary in February police released a high-tech image they said was a likeness of the mother of a baby girl found dead in a dumpster on Christmas Eve.
The image was produced by a company in Virginia that specializes in DNA phenotyping, which can predict physical appearance and ancestry from unidentified DNA.
Police found biological material at the scene that they sent in for the DNA phenotyping.
It was the first time Calgary police used the technology.
Similar technology was used in several other Canadian homicide investigations.