Disclosure: None of this blog post is intended to be, nor should it be construed to be, medical advice. If you need medical advice, please consult a medical professional who is knowledgeable about your specific question.
It appears 23andMe is testing the waters with a survey to see if its users would pay $749 for a “Premium Service.” Saw this article: https://www.cnbc.com/2018/09/05/23andme-considering-749-dollar-premium-service.html.
The article doesn’t indicate if it’s a Whole (or Full) Genome Sequence (WGS) test or a Whole Exome Sequence (WES). Considering Dante Labs offers both for less than $749 – https://us.dantelabs.com/collections/our-tests (WGS – $499; WES – $449 with a more detailed WGS for $999), and the expectation that WGS testing will eventually sell for as low as $99 at some point in the not too distant future, I think getting it for $749 (be it WGS or WES) from 23andMe would be a huge waste of money. If you have the money to burn, no big deal, but I don’t see a lot of people going for it.
Something that almost everybody doesn’t get is how limited medical DNA knowledge is right now. Sadly, the worst offenders are the medical researchers and scientists who think they can predict diseases, traits, medicine responses, etc. on a handful of genes, SNPs, alleles, STRs, etc. In many cases, they report that only one or a handful of genes, SNPs, alleles, STRs, etc. are responsible for a particular medical condition, trait, or medicine response. A good example that should get these experts re-thinking this theory is Schizophrenia. It’s estimated by some experts that around 250 markers are involved in determining if you will get Schizophrenia. Other experts claim it’s only one gene or a handful of markers. It’s a rare enough condition that it’s more likely that 250 is closer to the mark than one gene or a handful of markers. Plus, consider the fact that some of the markers may actually reduce your odds of getting it and others may be chance markers that appear to affect your odds of getting the condition without actually playing any part either way, and who knows how many other markers may play a part that aren’t currently part of the known or suspected markers. When you consider our DNA is composed of over 3 billion base pairs, that should give medical researchers and scientists pause that they should be looking at more markers. From my old 23andMe reports, they only looked at two SNPs (out of around 250 markers they should have used) to determine my odds of developing Schizophrenia (I don’t have the condition). In other cases, they looked at markers they acknowledged aren’t known to cause the condition, but are in the same genetic area as markers known to cause them.
The other thing to consider is the desire to make things statistical. I see this on the genealogy side as well as the medical side. To quote Mark Twain: Using 23andMe‘s Lactose Tolerance results for me, it estimates the following:
…, based on your genetics, you are not likely to be lactose intolerant.
Am I lactose intolerant? Yes, it’s not bad as I can drink a quart of milk in a short period of time and get anywhere from no intolerance to a moderate level of intolerance. It has gotten worse as I got older. When I was young, I could drink a gallon of milk, eat a large amount of other dairy products, and have no problems with lactose intolerance.
What is 23andMe basing my lactose tolerance on? One SNP, called a genetic marker below, but 23andMe only tests SNPs.
In looking at my 23andMe (v3 chip) old reports and the new reports, there are numerous reports where they only look at one SNP, sometimes a couple of SNPs to determine my risk factor for diseases, medical conditions, traits, medicine responses, etc. In at least one case, they acknowledged the SNP they were using to determine my risk factor wasn’t known to cause the medical condition, but it was in an area with known markers to cause the condition In my opinion, that’s nowhere near reliable enough for 23andMe or any other company or researcher to reach valid conclusions.