Those of you who have delved into our Gene Journal (now called Health and Traits) feature may have noticed that many of the traits only give genetic data “assuming European ethnicity”. Why is that? It certainly isn’t because people with African or Asian ancestry aren’t susceptible to heart attacks or colorectal cancer. The fact of the matter is that much of the research on the genetic factors that influence diseases is done in countries where people with European ancestry make up a majority of the population. As a result, the studies are done in groups that are mostly or completely Caucasian, and we can really only be sure the study results are only known to be valid in people with the same ancestry (actually the results might not even be valid in all people with European ancestry, but it’s the best we can do with the available science). Yet, you may still wonder, isn’t DNA universal? Why wouldn’t a SNP that makes a Swedish person a good sprinter also make a Korean person a good sprinter? In cases where a SNP directly affects the trait in question, it is very likely to be true that the different versions of the SNP do work the same way in everyone. But in many cases, the SNPs that researchers find in their studies are just stands-ins for DNA changes that actually affect the traits they’re investigating. We say that these stand-ins are “markers”. In the same way a sign on the highway might tell you that there are restrooms and gas stations at the next exit, a marker SNP tells you that a functional DNA change is nearby. (Read on to learn about marker SNPs and ancestry) Now let’s imagine that some prankster steals the highway sign and puts it next to some other exit that has no restrooms or gas stations. Because the sign has been separated from the exit it was originally next to, it is no longer informative. Similarly, if a marker SNP is separated from the functional DNA change it was originally next to, the marker is now uninformative. How could a marker SNP become separated from the functional DNA change it was pointing to? The answer is recombination–the shuffling of DNA that takes place when sperm and eggs are made. (Click here to learn more about recombination) Historically, major populations were reproductively separated from each other. That is, people from one group tended to have children with people from the same group. Since DNA shuffling during recombination occurs pretty randomly, each population developed its own unique shuffling pattern. So it’s possible that a marker SNP and a functional DNA change that are linked together in one population got shuffled away from each other in a different population. This means, for example, that if a version of a SNP is associated with disease susceptibility in people with European ancestry and it turns out to just be a marker SNP, it might be uninformative in people with Asian or African ancestry (if it got shuffled away from the functional change in those populations). The same goes for people with mixed ancestry. A marker SNP may or may not turn out to be informative for them. The difficulty is that scientists usually don’t know if the SNPs they find in their studies are marker SNPs or functional DNA changes. That’s why it’s important to repeat genetic studies in different ethnic groups to see if the SNP associations with certain traits are true for people of different backgrounds. We at 23andMe hope to perform our own research in a diverse set of peoples to help customers the world over understand more about their genetics.