Cho, Y., Go, M., Kim, Y., Heo, J., Oh, J., Ban, H., Yoon, D., Lee, M., Kim, D., Park, M., Cha, S., Kim, J., Han, B., Min, H., Ahn, Y., Park, M., Han, H., Jang, H., Cho, E., Lee, J., Cho, N., Shin, C., Park, T., Park, J., Lee, J., Cardon, L., Clarke, G., McCarthy, M., Lee, J., Lee, J., Oh, B., & Kim, H. (2009). A large-scale genome-wide association study of Asian populations uncovers genetic factors influencing eight quantitative traits Nature Genetics, 41 (5), 527-534 DOI: 10.1038/ng.357
A paper just released in Nature Genetics takes the most comprehensive look yet at the genetic factors underlying complex traits in an East Asian population, using a large sample of Korean individuals.
The researchers used a genome-wide association study to look at eight medically relevant traits: body mass index (BMI), waist-to-hip ratio, height, systolic and diastolic blood pressure, pulse rate, and measures of bone density in the arm and leg. The results reveal both similarities and intriguing differences in the genes contributing to trait variation in East Asian and European populations. For instance, while many of the genomic regions associated with BMI and height were the same in this study as those previously identified in European cohorts, there were a number of replicated variants associated with the other traits that have not previously been found in large European studies.
Why the differences? In some cases the answer is obvious: the genetic
variant differs in frequency between Europeans and Koreans, altering
the power of genome-wide studies to detect the association. In other
cases the variant is at a similar frequency in the two
populations, suggesting that the effects of certain genetic variations
may depend on variants elsewhere in the genome (i.e. a so-called
“genetic background” effect).
During the kerfuffle over the value of genome-wide association studies (GWAS) a couple of weeks ago much was made of the fact that even if genetic variants are not strongly predictive, they can still provide insight into the underlying biology of a common disease. The problem with this approach is that genes that play a major role in the disease process will not be picked up by GWAS if they don’t happen to contain a common variant that alters their function in that population.
By surveying multiple human populations by GWAS, however, we increase the pool of common genetic variants – and therefore increase the odds that any given disease pathway gene will be picked up in at least one GWAS. In addition, because different populations show different patterns of association between neighbouring genetic variants (linkage disequilibrium), looking in multiple populations can be extremely useful for dissecting out which variants are actually causative and which are mere bystanders.
With those goals in mind, you can expect to see many more GWAS of non-European populations over the next couple of years, and some explicit comparisons of the differing genetic architecture of complex traits between populations. Exciting times for those of us interested in the genetic and evolutionary basis of between-population differences…