Here's a bit from an article in New Scientist giving an excellent example:
Our tale begins two decades ago, when a group of Dutch women set out to find an explanation for the antisocial behaviour of the males in their family. As well as having learning difficulties, these men and boys were prone to outbursts of aggression and were racking up a list of serious offences, including arson, attempted rape and murder. Suspecting that the behaviour might be hereditary, the women approached geneticist Hans Brunner at the University Hospital in Nijmegen, the Netherlands. Eventually, in 1993, he tracked down the culprit: a defunct variant of a gene called monoamine oxidase A, or MAOA, located on the X chromosome.This complexity of indirect effect gets even more complex with epigenetics where an effect of an environmental condition on a gene can show up as a trait a generation or two later:
Understandably, the announcement created a sensation. It was the first time a gene had been linked to human aggression - and MAOA seemed to be responsible for a history of violence stretching back five generations.
In the following years, evidence poured in to bolster the connection between MAOA and aggression. Then, in 2004, journalist Ann Gibbons sealed the link by giving MAOA the headline-friendly label "warrior gene". The moniker stuck, raising the profile of MAOA, and at the same time fuelling misconceptions about how our behaviour is affected by our genetic make-up.
Genes simply make proteins. So for behavioural geneticists the question is how, when and why they influence our behaviour. Clearly, this involves a complex tango between genes and environments. Working out the exact steps is devilishly difficult, and conveying the findings to those without a scientific background is just as problematic. The mis-selling of MAOA is a salutary lesson in what can go wrong.
The so-called "warrior gene" is actually just a molecular garbage collector. It encodes a protein that breaks down some of the brain's signalling molecules when they have outlived their usefulness - including serotonin, noradrenalin and dopamine. If it slacks on the job, the build-up of these neurotransmitters leads to abnormal moods and behaviour. The gene comes in several varieties, distinguished primarily by their levels of activity. Because it is found on the X chromosome, females may have two different forms, while males have just one. Brunner's aggressive Dutchmen had a rare and completely inactive variant. The so-called low-activity variant, or MAOA-L, has a slightly shorter than usual promoter - the region that controls the gene's activity - and so produces less protein. Another common variant, MAOA-H, is more active.
Like the defunct version of MAOA found in the Dutch family, MAOA-L, is associated with violence and aggression. Last year, Kevin Beaver from Florida State University, Tallahassee, found that boys who carry MAOA-L were more likely to join gangs, and those who did were four times more likely to use weapons in a fight (Comprehensive Psychiatry, vol 51, p 130). Headlines proclaimed that "gang culture may be due to warrior gene" and that "boys who carry the gene are likely to be dangerous, violent and carry weapons". But this kind of deterministic thinking is wrong-headed. MAOA-L is actually very common: a third of white people have this version and most of them have nothing to do with gangs.
The discovery that MAOA-L-related aggression appears mostly as a reaction to certain circumstances is perhaps not surprising, since this is how most aggression manifests itself. However, the clearest sign yet that the gene is no ruthless determinant of behaviour came in 2002 when Avshalom Caspi and Terrie Moffitt of Duke University in Durham, North Carolina, published their findings about a sample of 442 men from New Zealand who they had followed from birth. A third of these men carried the MAOA-L variant. Now, aged 26, this group was indeed more likely than the others to have developed antisocial disorders and violent behaviour - but only if they had been poorly treated or abused as children. Moffitt and Caspi concluded that the so-called "warrior gene" affects a child's sensitivity to stress and trauma at an early age. Childhood trauma "activates" bad behaviour, but in a caring environment its effect is quashed.
Marcus Pembrey and colleagues also observed in the Överkalix study that the paternal (but not maternal) grandsons of Swedish boys who were exposed during preadolescence to famine in the 19th century were less likely to die of cardiovascular disease; if food was plentiful then diabetes mortality in the grandchildren increased, suggesting that this was a transgenerational epigenetic inheritance. The opposite effect was observed for females -- the paternal (but not maternal) granddaughters of women who experienced famine while in the womb (and their eggs were being formed) lived shorter lives on average.Go read the Wikipedia article for more details on epigenetics.