14 June 2012

Genomics: a time of great expectations

The Era of Genomics
The ability to predict the risk of developing diseases is just around the cornerDmitry Krylov, Voice of America

A new era is coming before our eyes in medicine. The price of reading individual genomes has dropped to thousands of dollars and below, DNA–reading machines have become compact and easy to use so much that some of them can be used without any special training - all these are harbingers of the era of individual genomics. Today's Technological environment is about how information in human DNA can be used and what tools are being created for this.

Knome Company (knome.com ) is located in Cambridge, Massachusetts, that is, next door to Harvard University and the Massachusetts Institute of Technology. This company released a software package for genome analysis about five years ago. Then it cost about 350 thousand dollars. The program recently released by the same company will cost hospitals only a few thousand.

Why hospitals? The fact is that between a read individual genome and useful knowledge about health lies an area, if not an ocean, of the unknown. Human genomes are different from each other. If you compare any two, it is unlikely that there will be the same. But most of these differences say nothing about the health risk or sensitivity to medications – which is exactly what doctors (and, of course, their patients) would like to know. But in order to link certain genome variants and disease or drug sensitivity, huge amounts of data are needed. The task of identifying such a connection is a kind of search for a needle in a giant haystack.

To determine exactly where the needle lies, thousands and thousands of genome variants are needed... and diseases. Therefore, Knome chose doctors working in hospitals as the target audience for its new product. The bet is that doctors have the ability to accumulate information about diseases and – now – compare it with the data of individual genomes. As soon as sufficient statistics accumulate, it will be possible to talk about patterns and, more interestingly for doctors and patients, about the risks of a particular disease.

Currently, only a few probable diseases can be identified by individual genomes. They are also rare, and their prediction in itself is not the most valuable thing that genomics can do in principle. However, here you need to make a reservation.

Enthusiasts of individual genomics promise a lot of opportunities to predict the risks of developing diseases and treatment features. But they are silent about the fact that this connection itself is not proven at all for many diseases. It may turn out that even with large statistical samples, doctors and geneticists will not find sufficiently strong links between genome variants and diseases. Many factors affect health, and genetics is just one of them. In this sense, it would be more correct to talk not about the risks of diseases, the prediction of which is only a matter of time. We are actually talking about a giant study, the result of which may be the prediction of such risks. But it may also turn out that it will not be possible to predict these risks for sure.

The question of accuracy is crucial here. It is enough to imagine such a situation: a person who is actually healthy and, it seems, will never get sick, receives news of the likelihood of a serious illness. His life is changing. If the prediction was erroneous, the consequences for the patient, the doctor and the medical institution that made such a prediction can be very serious. At the other extreme, a person is given a guarantee of a long and healthy life, but soon he faces serious health problems. Who will be sued in the first place, in this situation it is clear without explanation.

All this is an extra reason for healthy skepticism. One should not believe in individual genomics. We need to wait for it to give concrete testable results. The same, however, applies to any other technology.

Portal "Eternal youth" http://vechnayamolodost.ru14.06.2012

Found a typo? Select it and press ctrl + enter Print version