In 1860, as monk Gregor Mendel experimented with pea plants in the Augustinian convent of Brno, Emile Zola wrote a letter to his publisher. He proposed a grand cycle of novels in which he would “study in a family the questions of blood and environments”, to explore the fate and fortune of several generations. His scheme was to unfold as the 20 books of the Rougon-Macquart series, including Germinal and Nana.

Zola, of course, knew nothing of genetics. Mendel’s work did not become well known until his death, which came 50 years before the double helix and a full century before the sequencing of the human genome. His Rougon-Macquart cycle, however, helped to create a powerful prism through which much of this science is still seen. The French novelist was among the great popularisers of what has come to be known as genetic determinism.

The founding principle of the Rougon-Macquart series was that heredity matters. While members of the clan spanned the social classes, all shared a deep inherited flaw, described by Zola as a “ravenous appetite”. It could manifest as alcoholism, rage, avarice or destructive ambition, according to circumstance. But it was something those who had it could not ultimately escape.

In the age of the human genome, this idea that our unique DNA profiles explain us has acquired new currency. Science can now read the genetic code and use it to shed light on all manner of human affairs.

DNA can elucidate human history and ancestry, charting population migrations and family relationships. It can reveal clues to physiology, highlighting health risks to which we are particularly prone. Though the genetic variants responsible are not yet known, it is becoming clear that heredity also contributes to personality. And this information is becoming increasingly accessible, through DNA tests that fall in price almost by the day.

As new advances allow the genome to speak ever more clearly, however, it is tempting to give its voice greater significance than it deserves. The success of DNA fingerprinting as a forensic tool, genealogy tests that declare us to be Vikings or Zulus, and embryo screening for dread diseases such as Huntington’s and cystic fibrosis, all encourage us to think of genetics as a conclusive science. Yet it is rarely as simple as in the tales of the Rougon-Macquart.

Genes, for example, almost certainly affect athletic ability. One of these has even been identified — a particular variant of ACTN3 has been associated with fast-twitch muscle fibres that produce explosive speed. This has spawned commercial tests, offered in China, the United States and Australia, which purport to tell parents whether their children have potential as athletes. They are probably worthless.

The qualities required for sporting success — power, dexterity, balance, temperament — will be affected by hundreds of genes. Testing for only one of them will offer few clues and may well mislead: the potential of a child with the wrong ACTN3 profile, but many other favourable genes, could easily be overlooked. What is more, genes are only one of the factors that produces world-beaters such as Usain Bolt and Jessica Ennis. Environmental influences such as training, diet and family are at least as important, if not more so.

Measuring performance, which takes nature and nurture into account, is a far better guide to potential than any genetic test, and is likely to remain so for the foreseeable future. ACTN3 is no predictor of destiny.

Another recent example of genetic wishful thinking is the UK Border Agency’s idea of using DNA to investigate the true nationality of asylum seekers claiming to be Somalis. It is seductive to imagine that the vexed issue of asylum might be decided definitively by the genome. The evidence that it might be, however, is not there. And perhaps this is why the agency is now saying that it won’t use the programme to decide the fate of asylum seekers.

Genetics can reveal something of a person’s geographical provenance: studies of Europeans have shown that it is possible to plot a person’s broad region of ancestry from their DNA. But that doesn’t equate to a viable nationality test. The sort of detailed genetic data that would be needed doesn’t exist for Africa, and even if available it could establish only a probable geographical ancestry, not citizenship. Ethnicity rarely follows national borders, and in any case, people move. Genetic genealogy tests often identify only a single distant ancestor and are difficult to interpret for people of mixed race. There would be a grave risk of injustice were these to inform asylum decisions.

Pitfalls such as these need to be properly considered as genetic technologies become more and more mainstream. Insights from DNA will often have great value, particularly in healthcare: I expect genetic tests to play a major part before long in disease screening and the prescription of drugs. But we need to be wary of falling for their novelty, and the perceived authority of genomic information, before their utility has been properly assessed. For all its power, genetics is not without its limits.

If we are to draw a lesson from Zola for the genomic age, it should not be the faith he placed in the power of heredity — important as DNA might be. Besides his novels, he is best known for his passionate defence of Alfred Dreyfus, a French captain wrongly convicted of treason. His respect for evidence is just what we need to make full use of genomics without being misled.

Times Recommends

• Fern Britton: 'I still can't see that I was wrong'
• Extreme cooking
• Easy French: Tarte Tatin