Personal Genetics: Testing for Susceptibility
Ask Science talks about genetic susceptibility. What does it mean to be susceptible to a disease? Can susceptibility really be tested? Click to learn more about personal genetic testing.
Lee Falin, PhD
Listen
Personal Genetics: Testing for Susceptibility
In last week’s episode, we laid some of the groundwork for understanding the kinds of information you can and can’t get from a personal genetic testing service. This week, we finish up by looking at what genetic susceptibility means, and why these types of tests can’t tell you the whole story.
If you haven’t read or listened to last week’s episode, make sure you do so before continuing..
Confounding Factors
Let’s say that you are a scientist trying to run a GWAS to find SNPs associated with a certain disease. You find a couple of SNPs that seem to consistently occur in the group of people that have the disease, and which don’t occur in the group of people that don’t have the disease. Then you do some more research and discover that all of the people with those SNPs are of Russian ancestry.
So are the SNPs you found really associated with the disease, or are they just associated with a haplotype that is common to people of Russian ancestry? Could there by other SNPs in that haplotype associated with the disease that your test didn’t measure? Could there be other factors that people of Russian descent might have in common, such as diet, environment, or lifestyle?
All of these issues are confounding factors that make it nearly impossible for a GWAS alone to say whether or not a particular SNP is really associated with a particular disease.
Genetic Susceptibility and Odds Ratios
All of these issues are confounding factors that make it nearly impossible for a GWAS alone to say whether or not a particular SNP is really associated with a particular disease.
Typically, when we report on SNPs being associated with a particular condition, such as a disease, we do so using something called an “odds ratio.” Let’s say that we do a study of people that have loud sneezes. We find a SNP that seems to be associated with this group of loud sneezers and we look at the possible nucleotides.
We discover that of all of the people with loud sneezes, a large portion of them have a certain nucleotide at this SNP (scientists would say a large portion of them have a certain allele.) However, not all of the loud sneezers have that allele, and not everyone with that allele has a loud sneeze. So we can’t say for certain that this allele causes loud sneezes – we can just say that it is associated with loud sneezes.
So now let’s say you’re looking at your genome report and you see that you have the loud sneeze allele. You start to worry that your sneezes might be louder than you thought. Should you see a doctor about your sneezes? Maybe try some alternative sneeze therapies?
Then you notice something on the sneeze report called an “odds ratio.” What does that mean? Well, in this case, an odds ratio tells you the ratio of the odds of being a loud sneezer if you have the allele compared with the odds of being a loud sneezer if you don’t have the allele. The higher the odds ratio, the more likely someone with this allele is to be a loud sneezer.
Unfortunately, there’s no way for an odds ratio to tell you if your loud sneezing is really because you have a certain allele. Maybe you’re of Scandinavian descent, and maybe that allele just happens to be part of a Scandinavian haplotype.
Or, there could be some other SNP within that haplotype that is the real cause of loud sneezes. Perhaps you have that allele, and perhaps you don’t. Your genetic testing service might not even test that allele, because maybe at the time they designed the test, it didn’t seem important. Unfortunately, there’s no real way to be sure.
Other Factors
One of the big things you need to know about DNA is that while your DNA controls a lot of what happens inside your cells, it isn’t the only thing. First off, some other factors can actually control your DNA – things like age, diet, environment, lifestyle, and even the bacteria you’re exposed to can actually result in certain parts of your DNA being turned on and off. These interactions (collectively called “epigenetics”) are so complex that we’re just beginning to understand them.
Another thing to keep in mind is that GWAS odds ratios can only be interpreted properly in the context of a specific population; this is the only way we have of dealing with the haplotype problem, and even this isn’t perfect.
For example, a certain allele might make you more likely to have a certain disease, but we can usually only make that claim in a statement like, “If you are of Asian ancestry and have this allele, you are this much more likely than other people of Asian ancestry who happened to sign up for a particular genetic study to have this certain disease.” But even then, it’s an association – which is just another word for correlation. And as we all know, correlation is not causation.
Conclusion
So now you know more about some of the caveats of these kinds of genetic testing services.
If you have a question that you’d like to see on a future episode, send me an email at everydayeinstein@quickanddirtytips.com.create new email
If you liked today’s episode, you can become a fan of Ask Science on Facebook or follow me on Twitter, where I’m @QDTeinstein.
Photos of blood sample and sneezing man courtesy of Shutterstock.
Â
Â
