What is the accuracy of online DNA testing?

The era of consumer genomics has arrived. Nowadays, you can send a bottle of your brooch by mail and pay to see how your unique genetic code relates to all kinds of human activities – from sports to diets to skin creams to skin creams. Fine wines, and even encounters. The popular and popular companies in this market analyze the origins, and the most important of them are 23andMe and AncestryDNA, both of which have more than five million users in their databases. These numbers divide the number of human genomes in the scientific databases. Genetic genealogy is a big deal and has become mainstream. But what is the accuracy of these tests, really?

First of all, a bit of genetics 101. The DNA is the code in your cells. This is the richest but also the most complex information treasure we have ever tried to understand. Three billion individual DNA letters, roughly, organized into 23 pairs of chromosomes – although one of those pairs is not a pair half the time (men are XY, women are XX) . The DNA is organized into about 20,000 genes (although it remains to debate the definition of a gene). And rather than genes, almost all your DNA – 97% – is a mixture of control regions, scaffolding and huge pieces of repeated sections. Some are just garbage left by billions of years of evolution.

Modern genetics has revealed an image of immense complexity, an image that we do not quite understand – although we are certainly far from Mendel's experience and his pea experiments, which first identified the heritage units we call genes. During the twentieth century, we have gained a thorough knowledge of the basics of biological inheritance: how are genes passed from one generation to the next and how do they code the proteins that are at the origin of all life? In the 1980s, we identified mutated genes, producing defective proteins that could cause terrible diseases such as cystic fibrosis or muscular dystrophy, for example.

In 2003, the human genome project delivered the human DNA sequence in its entirety. One of the most important by-products of this business was the advent of technology that allowed us to read DNA at unprecedented speed and at ever lower costs. We can now pump the genomes of hundreds of thousands of people to get peanuts. With these data, we are gaining insight into the deeper issues of inheritance, evolution, and disease. Variations in human genomes are in fact infinite and scanning our DNA helps us understand what makes us human beings as a species and as individuals.

The huge costs of gene sequencing have led to commercial interests. Suddenly, any business can install and, in exchange for a little money and a flask of saliva, can extract your DNA from your cells. mouth and sequence your genome. Along with the 23andMe and AncestryDNA giants, dozens of companies have done just that.

The question of the accuracy of the results raises two potential problems. The first is a bit trivial: was the sequencing done well? In criticizing this activity, it seems fair to assume that the data generated is accurate. But there have been some odd cases of failure, for example, the company that failed to identify the DNA of the sample as coming not from the company. a human, but a dog. A recent analysis found that 40% of the variants associated with specific diseases, "direct-to-consumer" genetic testing (DTC), were false positive when the raw data were reanalyzed.

Assuming accurate testing, some discrepancies may still result from differences in companies' DNA databases. Almost all DTP genetic tests do not sequence your entire genome, but rather look at the positions of your DNA of particular interest. During my test by 23andMe, they stated that I did not have a version of a gene strongly associated with red hair. Another ancestry company said that I had done it. This simply reflects the fact that one company was looking at different variants of the gene that codes for red hair.

If we assume that the data generated is accurate, the second question that arises is the interpretation. And that's where it gets dark. Most of the positions of interest in your DNA are determined by experiments known as wide association genome studies, or GWAS (pronounced gee-woz). Take a group of people, as much as possible, who share a common characteristic. This could be a disease, such as cystic fibrosis (CF) or a normal trait, for example, red hair. When you sequence all their genes, you look for individual places in their DNA that are more similar in the test group than in another population. In the case of cystic fibrosis, chromosome 7 would rise sharply because the majority of cases of cystic fibrosis are caused by a mutation in a gene. For redheads, you should see 16 or 17 ears very close to each other because there are several variants in the same gene that grant all strands of ginger. But for complex traits such as taste or those related to diet or physical exercise, dozens of variants will emerge and all of them offer only a probability of susceptibility to certain behavior resulting from your DNA, as measured in a population. This even applies to something as simple as the color of the eyes: a variant of the gene associated with blue eyes is only a probability of blue eyes, and it is perfectly possible to have two blue-eyed genes without having blue eyes. .

Genetics is a probabilistic science and there are no genes "for" anything in particular. I have serious reservations about the usefulness of genetic tests that indicate a person's propensity for certain conditions outside the clinical setting; if you do not have a doctorate in genetics, these results can be misleading or even troubling. Even if, like me, you wear a version of a gene that increases the likelihood of developing Alzheimer's disease, most people with this variant do not develop the disease, which is also deeply influenced by many choices of life and blind luck. A geneticist can not tell you this information that goes beyond the usual lifestyle tips: do not smoke, eat a balanced diet, exercise regularly, and wear sunscreen.

With regard to ancestry, DNA is very helpful in determining close family relationships such as siblings or parents, and dozens of stories are emerging that bring together or identify lost family members (or criminals). For deeper family roots, these tests do not really tell you where your ancestors come from. They say where DNA like yours can be found on Earth today. By inference, we must assume that significant proportions of our deep family originate from these places. But to say that you are 20% Irish, 4% Native American or 12% Scandinavian is fun, trivial and has very little scientific significance. We all have thousands of ancestors and our family trees get tangled up as we go back in time, which means that in a very short time, our ancestors will become the ancestors of all. Humanity is intimately related to DNA and the DNA will tell you little about your culture, your history and your identity.