23andme and accepting the terms

[Azrael]

I reacted especially at one thing today when I accepted the 23andme terms:

Keep in mind that genetic research is not comprehensive and the laboratory process may result in errors

Does that mean 23andme is actually run by lab chimpanzees and they may do something wrong from time to time?

No but seriously how much errors can there be?

I honestly thought the raw data that will be available to me is 97%-100% accurate.

Anyone else felt like me?

 

[Sea]

97% accuracy is a 3% error rate. I don't think their error rate is that high, but there definitely is room for some error.

It is just a statement to cover themselves from being sued. Any gravely important information you learn from your genetic test should be validated by more testing.

 

[Azrael]

I understand it's some sort of cover, im hoping alot on this though to get some insight of what's going on with me.

In my country im not able to get any kind of help from the state healthcare with the problems I have.

When I talk about various tests they start talking about other things.

They ignore me
They are arrogant
They are incompetent
They want me to stay home and just shut my mouth.

Sad but true

 

[Critterina]

@Azrael ,

@Sea is right: anything less than 100% means that there are errors. The methods are good, but not infallible.

I've had similar problems getting help and actually was turned down by 4 specialists and the Mayo Clinic. But the choice is be sick/die or keep trying. I don't know why people are so surprised and impressed by my perseverance. Not a difficult choice if you ask me.

My recommendation is to be fact-based in your appeal. Ask questions about the tests or results. Ask questions as to whether the symptoms could be related (to each other, to drugs/supplements/procedures/practices). Stay rational. And take good notes as to what happens when, keep your lab results in order, and make your questions easy to understand in context. Some of us don't fit the patterns, and some doctors don't want to think outside the 99%. We will have a harder time getting diagnosis and treatment. But what choice do we have? I'm not one to give in.

My timeline from abnormal test result to seeing a specialist was almost 9 months. I barely avoided the emergency room about 7 months in. I lost a coworker to a related issue last January. You have to be your own best advocate. Rest when you need to, then pick the battle back up.

 

[barbc56]

The point is that even if the tests are 100% accurate, it's an almost impossible task to really know what the results mean.

The 23andme test only shows a small portion of your genetic code and it's not known how the genes not seen by the test might affect the overall result. Then you have to factor environmental effects, which make predicting what the results mean even harder.

Arguments rage all over the place on this one, but the short answer is that SNP tests—which is the kind of DNA scan 23andme relies upon— don’t tell us all that much (yet). They analyze a million genes out of three billion total and the impact those million play in long term-health outcomes is still in dispute. For example, the nature/nurture split is normally viewed at 30/70—meaning environmental factors play a far more significant role in long-term health outcomes than genetics

http://www.forbes.com/sites/stevenk...nter-of-the-personalized-genomics-revolution/


To find out the true ramifications you need a geneticist to interpret the results. Even then the diagnostic outcomes are based on probability.
Barb

 

[Critterina]

The point is that even if the tests are 100% accurate, it's an almost impossible task to really know what the results mean.

The 23andme test only shows a small portion of your genetic code and it's not known how the genes not seen by the test might affect the overall result. Then you have to factor environmental effects, which make predicting what the results mean even harder.



http://www.forbes.com/sites/stevenk...nter-of-the-personalized-genomics-revolution/


To find out the true ramifications you need a geneticist to interpret the results. Even then the diagnostic outcomes are based on probability.
Barb

I disagree, Barb. It's not at all impossible to tell what some of the results mean. That would be counter to science, and there would be no point in having a "Genetics Testing and SNPs" forum. If you know you have an elevated or reduced risk of developing a specific condition, that's based on probability. If you have gene mutations that effect the activity of enzymes, that's another, and quite likely quantified.

 

[barbc56]

To know the results you need as much information as possible. That's science.

23andme gives only part of the information. That's not science.

When I get my blood pressure taken, I want the top as well as the bottom number and not just one. Otherwise the number is not useful.

If people want to get the test that's fine but you can't dress it up as science.

IMO

Barb

ETA I had no idea we have a geneticss forum. Thanks.

 

[Valentijn]

23andme gives only part of the information. That's not science.

Actually it is science - it's just incomplete. If your data shows that you have a certain mutation known to cause a certain problem, that's true whether you just have 600K SNPs from 23andMe or every SNP in your genome.

The short-coming with 23andMe data is that it can't completely rule out problems (though perhaps indicate they're less likely) due to not testing every SNP. Similarly, there are certainly SNPs tested by 23andMe (or not tested at all) which are problematic but unknown to be problematic because there's no research yet about those specific SNPs. But when the data does show a problematic SNP, it's extremely likely that it indicates a real problem.

 

[barbc56]

Your right, it's science but the test has little if no diagnostic/treatment value because of the unknown genes, their interactions with known genes from the test and the implication of environment.

. But when the data does show a problematic SNP, it's extremely likely that it indicates a real problem.

This is inaccurate for the same reasons I stated above. Disease states are not as straightforward as one isolated gene directly causing one condition. It's the interaction among all the factors that causes a genetic event.

This means that one little error in a snp can snowball into a hugh statistical error once all the unknown factors such as genes that aren't tested are known.

It's such a complicated field that takes years of study to become a genetist who is the most reliable source for interpretation.

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. Not all of SNPs arising in association studies are related to diseases, but many of them are. I don’t think its unreasonable to postulate that a significant percentage of 23andme customers have some error in a SNP that is medically important. Whether such errors are typically false positives or false negatives is unclear, and the extent to which they may lead to significant odds ratios is another interesting question. In other words, its not good enough to know how frequently warfarin sensitivity is being called incorrectly. The question is how frequently somemedically significant result is incorrect

These are the reasons I would not get the tests but it's a personal choice.

http://liorpachter.wordpress.com/2013/11/30/23andme-genotypes-are-all-wrong/

Take care.

Barb

 

[Valentijn]

This is inaccurate for the same reasons I stated above. Disease states are not as straightforward as one isolated gene directly causing one condition. It's the interaction among all the factors that causes a genetic event.

Actually in a great many cases it is exactly as simple and straight-forward as a single SNP causing a disease. Have a look through the OMIM database sometime - there are probably tens of thousands (maybe hundreds of thousands) of SNPs documented which are pretty well-known to cause disease.

 

[barbc56]

@Valentijn

Interesting. What diseases are you referring that have this one to one correlation assuming, of course, there are no errors in the 23andme test results?

Thanks.

Barb

 

[Valentijn]

Interesting. What diseases are you referring that have this one to one correlation assuming, of course, there are no errors in the 23andme test results?

Off the top of my head - several varieties of Muscular Dystrophy, hemochromatosis, mitochondrial diseases, diseases caused by genetically-induced vitamin transportation/use/etc (B12, D, biotin, folate), diseases caused by genetically-induced enzyme deficiencies (hypotonia, lactic acidosis), etc.

I think many intelligent people get confused due to the way which Yasko and some others (mis)use SNP data. They take a lot of SNPs with very minor, non-disease-inducing effects, then fail to indicate how little effect they have, and/or badly sensationalize the effect which those SNPs have.

Many SNPs have an effect, but only some (tens of thousands?) are known to have a strong pathogenic effect. But for those pathogenic SNPs, it's very much the case of flipping the disease switch from "off" to "on". Whereas the other SNPs with a minor impact are more like turning a dial a couple notches, maybe from 0% to 2% on a dial that can go to 100%. And then other genetic or environmental or infectious factors have to turn the dial the rest of the way to get to a disease state.

It gets even more confusing because things like the Yasko panel combine the SNPs with tiny effects with SNPs with major (turning the dial to 65%) and/or pathological effects, plus a few completely harmless SNPs. But there is an abundance of real science showing quite clearly that certain SNPs (and certain pairs of SNPs on opposite strands of a gene) are 100% disease causing.