A Newcastle academic claims he has beaten the mainstream of schizophrenia researchers to discover that schizophrenia is caused by a single gene plus three environmental factors.
Schizophrenia, like coronary artery disease or asthma, is thought, by most researchers, to be caused by several genes, rather than one, as well as a host of environmental factors.
Single-gene diseases tend to be rarer than schizophrenia, with well-defined symptoms - for example, those of cystic fibrosis.
But Dorian Pritchard, of Newcastle University's department of human genetics, says he has devised a mathematical technique which can count the number of genes involved in complex diseases.
He does this by studying factors such as the rate of schizophrenia in identical and non-identical twin siblings of schizophrenia sufferers.
He says his work shows that most people with hereditary schizophrenia will have just one gene responsible, although this gene may differ between families because, for example, different mutations may have arisen on different sides of the world. The paper is published in Annals of Human Genetics.
Dr Pritchard says he can now predict the likelihood of someone developing schizophrenia, based on their family history and figures he has allotted to certain environmental effects, such as being born at a certain time of the year - thought to be because the mother is more likely to have had flu when she was pregnant - and being brought up with a schizophrenic in the family.
He is already being contacted by adoption agencies to assess a child's risk of developing the disease.
But Pak Sham, a psychiatrist and statistician at the Institute of Psychiatry in London, said: "This idea is quite counter to the consensus view. These days people are saying that probably there are three or four of half a dozen genes that can be identified."
He said that Dr Pritchard's work contained many assumptions and approximations. But, he said: "It does challenge the commonly accepted position and, in our present state of knowledge, we don't know for certain which kind of model is closer to the truth."