Just a year ago, science understood little about the genetic origins of common conditions such as diabetes and cancers of the lung, breast and prostate. These were known to be affected by inheritance, but with the exception of rare mutations with catastrophic effects, the genetic mechanisms responsible remained elusive.

The picture has changed decisively, as was highlighted this week by the discovery of six new genes that affect type 2 diabetes, and the first genetic variant linked to lung cancer. These are only the latest chapters in a story that is transforming medicine. In the past 12 months, close to 100 common genetic variants have been linked to major diseases, using a new gene-hunting technique called whole-genome association.

Such knowledge is providing unprecedented insights into the molecular pathways of disease, which should help with the design of new therapies.

Some of these could be aimed at patients with particular genetic profiles, in whom they are most likely to work. Another prospect is genetic screening, allowing people to assess their disease risk, with a view to lowering it through drugs, diet or exercise.

It is not hard to see why this “new genetics” has caused so much excitement, but it has to be tempered by realism. The newly identified variants do not, on their own, make much difference to disease, raising risk by as little as 10 per cent. Most conditions are affected by dozens of low-risk genes interacting with one another and environmental factors in a complicated web. While these discoveries are clearly important, it could take years to translate them into clinical advances that help patients.

This complexity is amply demonstrated by this week's findings. People with a particular variant of the JAZF1 gene, for example, are between 15 and 30 per cent more likely to develop type 2 diabetes, depending on whether they inherit one copy or two. The same gene, however, also affects men's risk of prostate cancer, but in the opposite direction.

There seems to be a genetic trade-off: those whose genotype protects against diabetes may be more likely to get cancer, and vice versa.

Scientists think that this gene, and two others with similar effects, probably affects cell division: variants that speed it up are good for diabetes and bad for cancer, while those that slow it down work the other way round. This should help research into both conditions, but it has pitfalls, too. There is a danger that diabetes drugs that target this pathway might inadvertently promote tumour growth. These results are going to be useful, but they will have to be handled with care.

The same is true of the lung cancer findings. Here a variant on chromosome 15 is associated with a 30 to 80 per cent raised risk, and the link was found by three independent teams.

These groups, however, differed in their explanations. One attributed the gene's impact to its effect on smoking behaviour, finding that people with the risky version become addicted more easily and smoke more. Another came to the opposite conclusion, finding that the raised risk was independent of tobacco consumption, and applied even to people who have never smoked.

Either way, this gene is important. But before it can contribute much to medical research, we are going to have to establish who is right.

Locating the gene is a big step forward, but it is only the first of many. It is understanding how it works that will open new approaches to treating lung cancer, or to helping smokers to give up.

As a commentary in the journal Nature said this week, our expectations of genetic research need to be managed in the manner of Winston Churchill. The wonderful fruits of whole-genome association studies are not the beginning of the end of the struggle to understand genetic diseases. It is better to think of them as the end of the beginning.

Mark Henderson is the Science Editor of The Times

Explore Health

• Expert Advice

• Health Features

• Mental Health

• Alternative Medicine

• Child Health

• Health Club