San Diego’s DNA intelligence company Human Longevity has developed an algorithm that can predict what people look like using only their genome sequence.

The technology has applications in solving crimes by putting a face to a DNA sample, making Human Longevity (HLI) one of several local genomics companies dabbling in forensics.

photo

Examples of real (Left) and predicted (Right) faces from the Human Longevity study predicting facial and other physical traits from whole genome sequencing data. Images courtesy of Human Longevity Inc.

The company’s new research was published in the journal Proceedings from the National Academy of Science, authored by Christoph Lippert and HLI’s executive chairman J. Craig Venter.

For the study, Venter and his colleagues sequenced the genomes of more than 1,000 research volunteers in San Diego, all from ethnically diverse backgrounds. The participants allowed the company to capture their height, weight and eye color, along with 3-D images of their faces and recordings of their voices.

Using machine learning, HLI trained computer algorithms to generate predictions of a person’s face based on raw genomic data with about 80 percent accuracy.

In a press release announcing the study, HLI warned that although this technology would be useful in solving crimes, it also has serious implications for data privacy.

“We set out to do this study to prove that your genome codes for everything that makes you, you,” Venter said in a statement. “We are also concerned that the public and the research community at large are not adequately focused on the need for better safeguards and policies for individual privacy in the genomics era and are urging more analysis, better technical solutions, and continued discussion."