This article was taken from the January 2014 issue of Wired magazine. Be the first to read Wired's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.
Your whole genome sequenced and stored on your smartphone. A simple app that gives you access to your genetic data, computing your risk of various diseases and telling you what to do to prevent them. Your DNA is the answer to everything. It tells you which medications work best for you and which don't at all, what diet is optimal for you and even whether your partner is a good fit. It knows whether you have good sales skills and what products you are most likely to buy.
Really? Just how predictive is your DNA? Prediction is about uncertainty, probability, the odds that something will happen or not. DNA is far less predictive than many people think. Yet, at the same time, DNA is indeed very informative.
Ten years after the Human Genome Project ended, the most consistent research finding is that common DNA variants aren't very predictive. In particular, most common diseases, such as Type 2 diabetes, cardiovascular disease and asthma, are difficult to predict because they have complex causes. Many genes have been discovered, but their effects are small and contribute to the cause of disease by interacting with other genes and risk factors such as smoking or diet. The number of possible "causal" interactions is so enormous that every patient's disease likely has its own complex and unique cause.
This uniqueness limits predictive ability: it is impossible to accurately predict something that has never occurred before. The same reasoning holds for traits and skills that are known to have a genetic basis. Genetic factors may well play a role in intelligence, sexual orientation and sports ability, but their role is not large enough to be predictive. Not even close. DNA does better at predicting risk for some diseases, such as hereditary cancers or autoimmune diseases, where just one or a few major genes have large effects, but even here the clinical benefits of testing are often not evident.
For diseases caused by mutations in a single gene - such as cystic fibrosis and Huntington's - DNA testing is a very powerful diagnostic tool. It can be used to test carrier status of the mutation at any age, before the onset of symptoms, even before a fertilised egg is implanted during in vitro fertilisation. Thanks to recent advances in whole-genome sequencing, DNA testing is also increasingly informative for diagnosing rare syndromes present from birth, including many cases of severe mental retardation.
DNA testing is useful for ID. Paternity and forensic testing are two examples: one tests relation of father to child, and the other matches the DNA of a suspect with that found at a crime scene. Even bacterial DNA is a useful identifier when trying to pinpoint the source of an infectious disease outbreak.
The predictive ability of DNA is limited not by technology, but by the complexity of the causes and the heritability of the disease and traits. DNA testing might not predict diseases well, but it can diagnose them. It will not be able to predict who is going to commit a crime, but it might help find out who did it; and it is not going to predict who will be the best father for a child, but it can help find out who the father is. DNA might teach us more about the past and the present than about the future.
Cecile Janssens is professor of translational epidemiology at the Emory University Rollins School of Public Health in Atlanta, Georgia
This article was originally published by WIRED UK
