4 ms·

This is the second article in two days on Hacker News about the topic of people taking genetic testing too seriously. It’s really sad to realize how much damag

This is the second article in two days on Hacker News about the topic of people taking genetic testing too seriously. It’s really sad to realize how much damage Americans have done to themselves based on junk science. Science works, but it appears most people are in no position to figure out how to respond appropriately to scientific findings. I don’t know what the solution is, but the pattern is tragic.

dennis_jeeves 7y ago
>Science works, but it appears most people are in no position to figure out how to respond appropriately to scientific findings. >I don’t know what the solution is, but the pattern is tragic. There is ( very unfortunately) no solution, for the vast majority of people ( tragedy of the masses if you will). But a tiny minority who are more intellectually gifted can team up and pool their resources ( intellectual or otherwise ) to better their health by equipping themselves with useful information ( i.e what is not junk science ) and perhaps even training themselves medically if needed. My personal peeve with this tiny minority is that they are often too individualistic ( by virtue of being independent thinkers) to team up. Side note - the article was behind a paywall, hence could not read it, but genetic tests aside, my understanding is that in mainstream medicine this pattern of over/mis-diagnosis often happens leading to mis/over-treatment, where often the treatment is worse than the supposed disease.
keanebean86 7y ago
A year or so ago my sister took a dna test. Some other family members have also taken the test. The results showed an inconsistency in mitochondrial dna. My sister had a different kind than our cousins. When she first found out she could only find 2 explanations. A super rare, basically impossible, random mutation or my sister and cousins had different female ancestors. My sister hoped for number 2. She imagined our grandpa having an affair that resulted in our mom. To avoid shame our grandma raised the love-child as her own. my sister and I have always felt out of place so this would certainly explain everything. Turns out the dna test is just not great at distinguishing mitochondrial dna.
- stevenwoo 7y ago
  Cursory reading of the mitochondrial DNA test indicates there is a European gene set bias built into some versions of it, and there is a non straightforward hereditability with mtDNA (conceptually like chimera?) https://isogg.org/wiki/Heteroplasmy https://isogg.org/wiki/Heteroplasmy Did yall eliminate those?
ineedasername 7y ago
You're right in general about junk science, but that label doesn't really apply to this sort of genetic testing. Genetic testing is not junk science, it's just that we're still in the early days and so there's much less certitude about some of the correlations that have been observed.
Isinlor 7y ago
I don't see there any evidence that a major methodological mistake was made, or that these women did not receive expert opinion. This seems to be just science. There is a lot to discuss about reproducibility of science. But mistakes are and will be part of the process, because science can not absolutely prove anything. Scientist normally expect that assuming null hypothesis they will gather data falsely pointing something significant in 5% of cases. In reality it seems to be a lot higher and that is not good. But if the type of mistake from the article is indeed rare, I assume less than 5%, then taking chance of that mistake might have reduced the best guess of cancer only by some small procentage points. These women had to take very difficult decision under high uncertainty anyway.
BurningFrog 7y ago
If you only look at the costs but not the benefits of something, it will always look tragic.

dennis_jeeves 7y ago

>Science works, but it appears most people are in no position to figure out how to respond appropriately to scientific findings. >I don’t know what the solution is, but the pattern is tragic. There is ( very unfortunately) no solution, for the vast majority of people ( tragedy of the masses if you will). But a tiny minority who are more intellectually gifted can team up and pool their resources ( intellectual or otherwise ) to better their health by equipping themselves with useful information ( i.e what is not junk science ) and perhaps even training themselves medically if needed. My personal peeve with this tiny minority is that they are often too individualistic ( by virtue of being independent thinkers) to team up. Side note - the article was behind a paywall, hence could not read it, but genetic tests aside, my understanding is that in mainstream medicine this pattern of over/mis-diagnosis often happens leading to mis/over-treatment, where often the treatment is worse than the supposed disease.

keanebean86 7y ago

A year or so ago my sister took a dna test. Some other family members have also taken the test. The results showed an inconsistency in mitochondrial dna. My sister had a different kind than our cousins. When she first found out she could only find 2 explanations. A super rare, basically impossible, random mutation or my sister and cousins had different female ancestors. My sister hoped for number 2. She imagined our grandpa having an affair that resulted in our mom. To avoid shame our grandma raised the love-child as her own. my sister and I have always felt out of place so this would certainly explain everything. Turns out the dna test is just not great at distinguishing mitochondrial dna.

stevenwoo 7y ago
Cursory reading of the mitochondrial DNA test indicates there is a European gene set bias built into some versions of it, and there is a non straightforward hereditability with mtDNA (conceptually like chimera?) https://isogg.org/wiki/Heteroplasmy https://isogg.org/wiki/Heteroplasmy Did yall eliminate those?

ineedasername 7y ago

You're right in general about junk science, but that label doesn't really apply to this sort of genetic testing. Genetic testing is not junk science, it's just that we're still in the early days and so there's much less certitude about some of the correlations that have been observed.

Isinlor 7y ago

I don't see there any evidence that a major methodological mistake was made, or that these women did not receive expert opinion. This seems to be just science. There is a lot to discuss about reproducibility of science. But mistakes are and will be part of the process, because science can not absolutely prove anything. Scientist normally expect that assuming null hypothesis they will gather data falsely pointing something significant in 5% of cases. In reality it seems to be a lot higher and that is not good. But if the type of mistake from the article is indeed rare, I assume less than 5%, then taking chance of that mistake might have reduced the best guess of cancer only by some small procentage points. These women had to take very difficult decision under high uncertainty anyway.

BurningFrog 7y ago

If you only look at the costs but not the benefits of something, it will always look tragic.