We’re starting to go through some of the interesting vignettes in Science’s 10th anniversary celebration of the human genome project. One of these papers takes a realistic view of how genomic research has benefited human health over the past 10 years. A few areas that the authors touch on:

1. Identifying risk:  The predictive power of most genetic variants associated with diseases is not very high. This means that the potential benefits of separating patients even into gross categories such as “high” and “low” risk based on the presence/absence of disease-risk genes are in many cases outweighed by the cost of potentially misclassifying (and thus mistreating) them.

2. The difficulty of changing behavior: When someone is told they are at a genetically higher risk of developing a particular disease, there is really no evidence to indicate that they change their dietary or exercise habits (see also this post on the blog). Altering an individual’s environment (regardless of the presence/absence of disease-risk genes) is probably a better, and more lasting, way of convincing them to be less lazy, or to eat better and not smoke.    

3. False hope: Scientists and the press are both responsible for creating false hopes for genomic research in human health.

The authors do suggest that the following are realistic expectations:

1. The genes responsible for most Mendelian disorders will be identified. This will permit quick diagnoses, particularly for diseases that are caused by a single gene. 

2. Pharmacogenomics (the study of the influence of genetic variation on drug response) will enhance the safety and efficacy of treatments. However, because a lot of variability in drug response is tied to non-genetic factors, we can’t expect genomics to completely solve this issue.    

They make the interesting suggestion that because most mortality in high-income countries results from things like smoking, sedentary behavior, and excessive food and alcohol consumption, the diseases associated with these factors are best (or at least as effectively) researched via the social and behavioral sciences (i.e., how do we change these behaviors?) rather than through genomics (i.e., how do we identify individuals at genetic risk for these diseases?).  


Evans, J.P., Meslin, E.M., Marteau, T.M., Caufield, T. (2011). Deflating the genomic bubble. Science 331: 861-862.

UPDATE 2.23.2011. Dr. Hawks blogs about this issue here.