Scientists criticized the "DNA portrait"
A person's face cannot be "calculated" from his DNA, geneticists said
The statements of biologist-entrepreneur Craig Venter about the possibility of reconstructing the face of any person by small mutations in his DNA do not correspond to reality, say scientists who found "gross errors" in his article in the journal PNAS (Identification of individuals by trait prediction using whole-genome sequencing data – VM) and published their findings in the electronic library bioRxiv.org
"The concerns that Venter and his co-authors write about are completely groundless and sucked out of the finger. In fact, I don't think this article demonstrated their realism, since they failed to prove that they can "calculate" a person's appearance from his DNA. What they found and what they write are completely different things," said anthropologist Mark Schriver, one of the reviewers who did not allow the publication of this article in the journal Science, whose words are quoted by the journal Nature.
Genomic Puzzle
In early September, an article appeared in the journal PNAS, in which Craig Venter and his colleagues told about the results of DNA analysis of about a thousand people living in different parts of the world and belonging to different ethnic groups. By comparing sets of small mutations in their genomes, California geneticists, according to Venter, were able to find DNA regions that control the shape of the face and other anatomical features.
These sites, according to the authors of that publication, make it possible to make fairly accurate "sketches" of their owners, correctly determining the shape of a person's face and body in 75% of cases, even if the geneticist has never seen him. According to Venter, this possibility is a huge problem from the point of view of personal data protection.
Despite the huge attention of the press to this publication, scientists met it with a great deal of skepticism. As Shriver notes, Venter initially sent the article to the journal Science for review, but reviewers, including the anthropologist himself, found errors in it and rejected it.
This, however, did not stop Venter, and he forwarded the text of the article to another prestigious journal, PNAS, whose editorial board allows authors to independently select some or even all reviewers. Thanks to this, the article was published and caused fierce controversy among scientists.
One of Venter's opponents, biomathematicist Yaniv Erlich from Columbia University in New York (USA), checked the conclusions and methods of California geneticists, and found "gross errors" in them.
The Face of Genetic Capitalism
According to Ehrlich, the conclusions that Venter and his colleagues obtained by studying small mutations in DNA could be obtained using three simple human characteristics available in their database – age, gender and ethnicity. As the mathematician notes, even if you throw out all the mutation data from the dataset, the probability of "determining" a person's identity will remain virtually the same – 75% – due to the small size of the dataset used by California scientists.
In addition, the significance of many genetic markers linking, for example, human growth and mutations in his DNA, was significantly overestimated and exceeded the generally accepted values by several times. The reason for this, according to Ehrlich, could be both relatively "honest" methodological errors, and quite conscious distortion.
In fact, the whole point of Venter and his team's article was reduced to a simple thought – a person's age, as well as his gender and ethnicity can be calculated from his DNA, and use this data to narrow down the circle of potential "suspects". This works in small groups of people, but will not work even at the level of the dwarf states of the world. Accordingly, there is no question of any "genetic portrait" of a person. According to Ehrlich, this cannot be done in principle, since many features of human anatomy are explained by variations in genes by only 15-20%.
Why did Venter and his colleagues take such steps? According to Shriver, the reason for this may be that Venter, as an entrepreneur and founder of Human Longevity, would not mind if access to genetic data for the "unenlightened public" was maximally restricted by the government. This would significantly reduce the level of competition in the industry and deprive small startups of the chance to start working in it.
"The most important thing is that such articles always attract the attention of politicians and officials. New rules and restrictions always appear and are based on such "discoveries". It is extremely important that we are aware of which risks to our personal data are real, and which remain the subject of fantasy of such people," concludes Ehrlich.
Portal "Eternal youth" http://vechnayamolodost.ru
11.09.2017