What does Italian data scientist Riccardo Sabatini look like? Using his technology, you don’t have to meet him or a see a picture to know. All you need is a few drops of his blood, and you can get scarily close.
Sabatini, who works now with J. Craig Venter’s company Human Longevity, Inc., has volunteers come to his lab to be photographed and have their DNA sampled. Using this data, he has taught his computer program to predict a human’s face with uncanny accuracy: height, weight, eye, and skin color, and even age can be known often within a range of 80% give or take. His facial predictions aren’t exact, but they are in the ballpark and will only continue to get better and we understand more of the human genome (other artists and researchers have also tried to do similar work). Today, scientists understand only about 2% of each individual’s 3 billion bits of DNA code.
"The predictions will get better, the code will get more precise—the more we will learn every time and we will be confronted with decisions that we never had to face before: about life, about death, about parenting," Sabatini told an audience at TED 2016 on Monday.
To demonstrate how much there still is to learn, Sabatini dragged out dozens of volumes of printed books of Craig Venter’s genetic code (Venter was one of the first to sequence the human genome, his own). Printed out, the code filled up 262,000 small-print pages that weighed 450 kilograms.
"[We are] trying everything we can try and learning everything we can learn from these books with one target: making real the dream of personalized medicine," Sabatini says.
Of course, this is an idealistic use case. A dystopian world where police and criminals can easily profile our identities based on a scrap of DNA is a scary one. So is one in which parents might be able to predict many more traits of a child before they are born. Sabatini says a conversation about these technologies must open up among all of society. "It’s a revolution that cannot be confined in the realm of science or technology," he says.