What is currently happening with personal genomics



The cost of sequencing the human genome is falling. It could soon become affordable for many people - at least in industrialized countries - to send in a blood sample. Two weeks later he or she would find a CD with the complete sequence of his genetic material in his or her mailbox (and an invoice for around 1,000 euros).

But the money would be badly used. Because a meaningful interpretation of the genetic letter salad is not possible for the foreseeable future. Above all, this concerns the prospect of finding those “genetic defects” in one's genome that increase susceptibility to diabetes, hypertension, arthrosis, obesity, Alzheimer's disease or other “common ailments”. The previous genome-wide association studies have not found any gene variants that explain a large proportion of the familial risk.

The reason is simple: the genome is 3 billion base pairs long, and a multitude of variants are possible on the 30,000 protein-coding genes. The variants are the result of gene mutations (in the germ cells) that are distributed more or less randomly over the genome.

What this means is shown by two studies that have now been published in Science. In the first study, John Novembre's team from the University of California in Los Angeles sequenced a small number of 202 genes in a large group of 14,002 people (Science 2012; doi: 10.1126 / science.1217876).

Genes that can influence the effect of drugs were selected. Gene variants were common and widespread. On average, there was a change in every 17th base pair. It is difficult to assess biochemically whether this leads to a functional disorder in the protein. In order to examine the effects on health, a larger study would have to be carried out for each variant, which is simply impossible.

The Exome Sequencing Projec of the US National Heart, Lung, and Blood Institute led by Michael Bamshad from the University of Washington State in Seattle also offers no cause for optimism. The researchers sequenced 15,585 protein-coding genes in 2,440 people (Science 2012; doi: 10.1126 / science.1219240). They came across more than half a million gene variants, most of which were extremely rare: 86 percent were present in less than 0.5 percent of the participants. In the case of the variants that the researchers attributed a possible disease-causing effect, the proportion was as high as 95 percent.

In summary, this means: Yes, gene variants can presumably increase the susceptibility to certain diseases, which has been known for a long time due to the positive family history and the associations in twin studies with many diseases. No, the susceptibility is not due to a few gene variants. To put it bluntly, everyone has their own genetic risk. It would be included in the genome analysis, but indistinguishable from the many harmless variants.