I've got you under my skin

Which genes determine a resemblance between family members? New genetic findings reveal that a single hair can paint a full portrait. By Jim Giles

If Inspector Morse hadn't been killed off by his creator, Colin Dexter, he would no doubt have had a keen interest in the research being carried at University College London. There, in the University's Galton Laboratory, scientists are searching for the genes that shape our faces. One day this could lead to crude photofits created only from hair or skin samples found at the scene of a crime.

If Inspector Morse hadn't been killed off by his creator, Colin Dexter, he would no doubt have had a keen interest in the research being carried at University College London. There, in the University's Galton Laboratory, scientists are searching for the genes that shape our faces. One day this could lead to crude photofits created only from hair or skin samples found at the scene of a crime.

Working out why we look like we do is not an easy task. Our facial appearance is determined by a complex interaction between our genes and our environment, both before and after birth. Genes obviously play a role - which is why we tend to look like our parents - but scientists do not know which of the 100,000 or so human genes are involved. The team at the Galton Laboratory, which is being funded by the Forensic Science Service, says the immediate aim is simply to identify the genes that shape our faces. Improvements in treatments for abnormalities of facial development, as well as the sort of forensic technology Morse would have loved, could well follow.

Professor David Hopkinson, the head of the team, says the future implications could revolutionise crime detection. "If a criminal leaves a spot of blood or a hair it may be possible to test this material and predict his or her appearance. I don't imagine that we will ever be able to provide a full photographic description, but an outline 3D photo-fit of the major distinguishing features should be achievable."

Understanding which genes influence the shape of any part of our body is difficult, but Hopkinson's team has an advantage in the case of the face because evolution has carefully selected the genes controlling facial appearance. Recognising the face is something we are, quite literally, born with - as any parent of a newborn baby can testify.

These "face genes" are the reason we look like our parents and that our children look like us. Francis Galton, the eccentric 10th century scientist after who the laboratory is named, was fascinated with the face. Indeed, he even constructed a "beauty map" of Britain by scoring points against the faces of women he observed in different towns.

The members of the Galton Laboratory team are linking similarities in the faces of family members to genes that these relatives share, allowing them to identify which genes are involved in the development of the different parts of the face. Before they could begin, they needed to determine a way of quantifying the similarities between two faces. To help them do this they turned to face-scanning technology more often used in computer animation and very sophisticated special effects. The process begins by two family members having the surface of their faces scanned by a low-power laser. The results of this scan are used to build up a detailed computer model of the contours of their faces. Developed by the Medical Imaging and Graphics group at UCL, the scanner can record the position of 50,000 different coordinates on a persons face in a mere 10 seconds. These computer models can then be compared mathematically to put a figure on the degree of similarity between, say, two noses or mouths.

The two volunteers now give DNA samples - a simple mouth swab is enough - so that their DNA can be compared. The crucial step is to link similarities between facial features to similarities in the subjects' DNA. It may be, for example, that a certain region of DNA is often shared by family members that have similar noses. The researchers can then start looking within this region for the genes that control nose growth.

So will the Forensic Science Service see a return on its investment in the Galton Laboratory's efforts? As Hopkinson admits, the notion of creating a detailed photofit from a single hair is still the stuff of science fiction.

Even if all the genes involved in face development were understood, forensic scientists would still be left guessing as to whether the suspect had a beard, wore glasses or had any distinguishing marks such as scars or tattoos.

But basic information about a person's features should be available clearly enough to persuade the Forensic Science Service to support Hopkinson's work. It's an exciting development - though probably not enough to persuade Dexter to revive Morse.

If your family is interested in taking part in this project please email; face@galton.ucl.ac.uk or phone 0207-504 5035. Jim Giles works on the Wellcome Wing project at the Science Museum in London