Identifying faces based on genomic data is not reliable

Release date: 2017-09-15

The ability to predict human appearance based on the genome has sparked heated debate among industry experts. Image source: nature website

Clegg Venter is an influential biologist in the United States. Although he is not a graduate of a so-called prestigious university, his every research move will always have a hurricane impact in the industry. For example, he set up a company to openly challenge the $3 billion human genome project, claiming to use the $330 million to pre-empt the human genome map, forcing the human genome project to be completed two years ahead of schedule; his research team created the world's first human life in the lab. ......

Recently, a company founded by Venter announced that their new research believes that DNA data will reveal the true identity of anonymous people through appearance and physical characteristics, which once again attracts other scientists' doubts and criticisms.

According to a recent report by Nature magazine, the research team of Human Longevity Company (HLI), which he founded, published a paper in the Proceedings of the National Academy of Sciences, suggesting that law-making institutions, scientists and other relevant personnel should strengthen the supervision of human genome data. Prevent personal privacy by exposing DNA data.

But well-known scientists from institutions such as the University of Pennsylvania and Columbia University have argued that the Venter team's research exaggerates the ability of genes to recognize people's identities, causing unnecessary panic about genetic privacy. Some insiders who participated in the study said that Venter’s move had a private purpose. He might use his influence to change the current status of public access to genetic data and use his own genetic database for profit.

Research conclusions or excessive exaggeration

The Venter team published a paper stating that they had genome-wide sequencing of 1061 volunteers of different ages and ethnic backgrounds, combined with high-resolution 3D photographs of these participants' faces, and they developed artificial intelligence software based on DNA. SNPs (single nucleotide polymorphisms) of the sequence recognize facial features such as height of the cheekbones, and it is found that a human SNPs imply individualized features such as height, weight, age, accent and skin color. It turned out that using their method and HLI's DNA database, the accuracy of identifying a person from the randomly selected 10 people was as high as 74%.

After carefully studying the papers of the Venter team, many geneticists questioned that the conclusions of the study were exaggerated. Mark Shriver, an anthropologist at the University of Pennsylvania, believes that this paper does not prove that DNA data can accurately identify a person's privacy. Therefore, the risks raised in the paper do not exist, and 10 people are free to choose. Samples, based on age, gender, and ethnicity, can completely distinguish these people without the need to use DNA data.

In order to make a strong rebuttal, Columbia University computer biologist Yanif Ehrlich re-analyzed the age, gender and ethnicity data mentioned in the paper and published the results on the pre-printed website BioRxiv. Any genomic information of a person, relying on these three characteristics alone is enough to distinguish a person from 10 people with an accuracy rate of 75%. He also concluded that the facial features reconstructed by HLI using SNPs data do not accurately point to a certain person, but are close to a certain type of person with characteristics such as gender and race.

Science magazine refused to send papers

Shriver said that before the paper was published in the Proceedings of the National Academy of Sciences, the Venter team submitted the paper to Science, and he was involved in the peer review of the paper. At the time, the jury thought that although the HLI research data was true and reliable, the new method of inferring the tester's age by sequencing the two ends of the chromosome was impressive, but the paper did not accurately identify the individual through DNA data as claimed. Information, research conclusions have a deliberate distortion tendency.

Although the authors of the paper include heavyweights such as Venter, Science finally refused to publish the paper. According to the regulations, as a member of the American Academy of Sciences, Venter has the right to choose three scientists to conduct peer review of his thesis, and then publish it in the Proceedings of the National Academy of Sciences. The Proceedings of the National Academy of Sciences acknowledged that Venter himself chose two experts in information privacy and a bioethicist, and the review passed the paper.

HLI refused to comment on the paper review process, but said in a statement that although the scale of the research sample set was too small, the methodology in the study was completely free of problems. Based on the genome database of about a thousand participants, one of them was accurately identified. It is entirely possible for people to do so, and this recognition will become more and more accurate.

Emphasize genomic privacy or for economic purposes

Jason Piper, one of the authors of the paper and now working as a computer biologist at Apple, admitted that he had signed a consent form with HLI and gave up the right to sign the consent before the publication of the paper, so that HLI stated the experimental data according to the company's wishes. The paper distorted the findings of his and other colleagues.

Piper severely criticized the paper through Twitter, and believes that HLI has interests in doing so. Due to the current system, all genetic data for research must be publicly available, but as a profit-oriented company, HLI hopes to build the world's largest database of human genetic information and lobby for policy development under the pretext of protecting personal privacy. The department sets various restrictions on the DNA database and uses this data as a tool for companies to make money. “Protecting genetic privacy is very important, but in order to learn more about genomic information, people must share all relevant data, and the way to set up barriers to this sharing is the opposite.” Piper said.

Although HLI has repeatedly stated that the purpose of protecting genomic information is to promote the security of data exchange, Erich believes that with the status of Venter, this paper will certainly attract more attention from policy makers. Such research papers have developed a new management system that prevents people from sharing genomic data in research.

Source: Technology Daily

Pain Relief Patch(Pain Areas)

Pain Relief Patch
[Name] Medical Cold Patch
[Package Dimension] 6cm×8cm 4pieces/box
The pain relief patch is composed of three layers, namely, backing lining, middle gel and protective film. It is free from pharmacological, immunological or metabolic ingredients.
[Scope of Application]
For cold physiotherapy, closed soft tissue only.
[Indications]
The patches give fast acting pain relief for strains, sprains, cramp, bruises, swollen areas or joint stiffness.
[How To Use a Patch]
Please follow the Schematic Diagram. One piece, one time.
The curing effect of each piece can last for 6-8 hours.
[Attention]
Do not apply the patch on the problematic skin, such as wounds, eczema, dermatitis,or in the eyes. People allergic to herbs and the pregnant are advised not to use the medication. If swelling or irritation occurs, please stop using and if any of these effects persist or worsen.notify your doctor or pharmacist promptly. Children using the patch must be supervised by adults.
[Storage Conditions]
Store below 30c in a dry place away from heat and direct sunlight.

Pain Relief Patch(Pain Areas),Shouler Pain Relief Patch,Joints Pain Relief Patch, Muscle Pain Relief Patch

Shandong XiJieYiTong International Trade Co.,Ltd. , https://www.sdxjmedical.com