- The bottom line is that even though the error rate for any individual SNP call is very very low (<0.01% error), with a million SNPs being called there is (almost) certainly at least one error somewhere in the genotype. In fact, assuming a conservative error rate leading to an average of 100 errors per genotype, the probability that a 23andme genotype has no errors is less than 10^(-40).
- From the estimates of error rates provided above, and using the back of an envelope, it stands to reason that about 1/3 of 23andme tested individuals have an error at one of their “interesting” SNPs.
Also a neat perspective from Mike Eisen, more from the regulatory point of view. He suggests that instead of trying to shove these genetic tests into the standard regulatory bins, the FDA should establish a new set of guidelines, specific to predicting phenotype from genotype: Sure, the Illumina arrays might count as a medical device, but you don't run clinical trials to verify genomics analysis. It's a whole other ball game to maintain your confidence in the results of a screen with a million separate markers.The FDA wants to classify genetic tests like those offered by 23andme as medical devices, and to apply the appropriately strict criteria used for medical devices to genetic tests. But the problem with this is that contemporary genetic tests will almost certainly fail to meet these criteria, and I don’t see who benefits from that scenario. ... The FDA and companies like 23andme need to come up with standards for accurately and honestly describing the current state of knowledge for genotype-phenotype linkages and their application to individual genotypes.