A crime is committed. All that's left at the scene is a single hair. Today's detectives can use that hair to identify the criminal to some degree, but what if DNA in the hair could completely reproduce the perpetrator's face? Science fiction you say? Think again.

A new study led by researchers at Erasmus University in Rotterdam suggests that this could be entirely possible. Faces are the easiest way to recognize relatives--a mother's chin, a grandfather's eyes, and a father's nose are some of the most inheritable traits we have. The study led by Manfred Kayser, professor of forensic molecular biology at Erasmus and published in the Sept. 14 Public Library of Science (PLOS), found genes responsible for these unique facial features. Five genes help shape our faces during development, Kayser's study found. By matching magnetic resonance images (MRI) of more than 5,000 people of European origin with their fully sequenced genomes, the researchers saw that these five genes helped shape features like the distance between the eyes, the length and width of the nose, and the position of the upper jawline. Variations in any of these five genes (called single-nucleotide polymorphisms, or SNPs) resulted in changes to these facial dimensions.
At least three of the genes were also associated with known diseases of facial development, including cleft palate, a facial disorder called Waardenburg syndrome, and stunted growth of the lower jaw. By connecting the variations of the genes responsible, it would be possible to develop treatments for these disorders.
Just as significant, however, was the implication that knowing the genes that develop the face could allow for the reproduction of a human face, even without the aid of a photograph or eyewitness.

"Perhaps sometime it will be possible to draw a phantom portrait of a person solely from his or her own DNA left behind," Kayser said. "We already can predict from DNA certain eye and hair colors with high accuracies," he added.

Significantly, the study could not make associations between MRIs and genomes of about 560 French-Canadian teens; the researchers suspect that the ongoing development of facial features during that age could be controlled by very different genes, or that the same genes may be behaving in entirely different ways.  Further research will include studying the whole face (this study only focused on faces above the nose), as well as studies involving much larger numbers of participants, in order to fully capture more genes that make our faces resemble our relatives, yet differ in so many ways.