Can DNA convict a 16-year-old corpse?

Doubts surround the methods being used to solve the 'Bible John' murders, reports Mary Braid

The remains of John Irvine McInnes were dug up this week - 16 years after his death and 26 years after a series of gruesome murders which terrorised Scotland and still haunt it.

As his grave was defiled - in the full glare of the media - detectives revealed he might be "Bible John", the killer of three women and a dancehall regular with a fondness for the Scriptures.Strathclyde police, like many other forces, are placing their faith in DNA testing to solve an intractable case. But as the coffin was lifted out by six officers, we may have been witnessing something altogether more grotesque. For it is entirely possible that the whole macabre exercise - painful for relatives of the suspect and victims alike - is a waste of time, and that, not for the first time, the belief in DNA profiling, forensic science's greatest breakthrough since fingerprinting was perfected in 1901, is entirely misplaced. While no one disputes that genetic profiling is a wonderfully powerful tool, it is becoming evident that it is a far more limited technology than was originally suggested. Its capabilities have been overhyped to such an extent that a major miscarriage of justice may now be waiting to happen.

Scottish forensic scientists hope to match DNA from McInnes's remains with a semen stain left on a pair of nylon tights belonging to Helen Puttock, the last victim. In leaning so heavily on the new forensic science, they are following the lead of the Government. Last year, the Home Office, despite concerns over the readiness of the technology and civil liberties implications, opened a national DNA database.

Recently, more than 2,000 men were DNA-tested following the killing of the Cardiff teenager Claire Wood, and police seeking the killer of the French hitchhiker Celine Figard are currently working their way through the DNAsamples of 1,200 British lorry drivers. More bizarrely, DNA testing has been threatened in a Scottish court battle over the ownership of a Burmese cat and deployed by Brazilian scientists trying to prove that 70-year-old bones are those of the British explorer Colonel Percy Fawcett.

Yesterday, Dr Wilson Wall, a DNA specialist geneticist, questioned the exhuming of John McInnes's remains. After 16 years, he said, it was unlikely that enough quality DNA had survived to allow a viable match to be made. Any new evidence was unlikely to constitute anything near proof. Such assessments reinforce doubts about Thursday's grave visit. If, in fact, any alleged evidence did emerge, where would it be challenged?

The fact is that despite its common label of "genetic fingerprinting", DNA profiling or testing enjoys none of fingerprinting's certainties. Its scientific base is far more complex, making its findings harder to explain in court, where juries can be falsely convinced of its strength.

The double helix structure of Deoxyribonucleic acid (DNA) was discovered in Cambridge in 1952. It is a chemical found in chromosomes in all body cells and carries the coded instructions, or genes, that lay down an individual's genetic blueprint. With the exception of identical twins, no two people have the same DNA or genetic code in their cells. In 1984, Professor Alec Jeffreys, a Leicester University geneticist, released its crime solving potential when he discovered among the billions of bits which make up the DNA chain those that distinguished between individuals.

Because of the unique nature of our DNA, the term genetic fingerprinting was quickly adopted for the profiling technique. But while the new crime detection device is far more versatile than fingerprinting - a sample can be obtained from hair, semen, blood, bones or any other part of the body - it is far less clear cut. If two sets of fingerprints match, they were made by the same person. You may argue about how your prints came to be at the scene of a crime, but you cannot argue that they are yours. However, profiling rests on the probability of an exact match. It is simply not possible to analyse someone's entire DNA double helix, which would yield exact identification. Instead, forensic scientists have to sample segments which they hope would be characteristic of an individual. Analysing a part rather than the whole, there is always a chance that two samples will be wrongly linked. No wonder civil liberties groups have warned against blind reliance on testing.

While the last decade has brought refinements to early profiling techniques, with that has emerged more evidence to undermine it. Laboratory procedures used to prepare profiles have been found lacking and the statistical techniques used to determine the probability that two samples are a perfect match have been challenged. Concerns about laboratory preparation emerged as early as 1987 when Californian police set out to test the system by sending samples to various testing companies to identify matches. Cellmark ( a division of ICI) came up with with seven false positives and even when the tests were repeated under the supervision of a second technician a false positive was still recorded. This time, the company blamed cross- contamination between samples rather than mishandling in the lab, which was the earlier excuse.

The irony is that in practice, DNA testing is more decisive in ruling out suspects rather than pointing the finger at them. It is often forgotten that the world's first mass DNA screening by Leicestershire police into the murders of Lynda Mann and Dawn Ashworth in 1987 first ruled out a man who had confessed to the killing. The police - aided by DNA testing - then went on to find the real killer.

So as the police examine what is left of John McInnes, they would be advised to bear in mind the limitations of the technique they are relying on. Even if they believe they have found Bible John, what independent court of law can assess their judgement? Can justice be done to a dead man's DNA?