Have We Drawn the Wrong Lessons from Epigenetics? | Hacker News Reader

Have We Drawn the Wrong Lessons from Epigenetics?

aeon.co

42 pointsmr_tyzic17 comments

> Our new awareness of epigenetics challenges genetic determinism and throws a new wrench into nature-nurture debates.

It does?

> Epigenetic studies show that genes alone do not determine form and function,

Of course genes alone do not do that - that's the whole point of the nature-nurture debate, and the fact that most things are a mix of nature and nurture. We already knew a long time ago that genes alone do not determine form and function.

> but that the cellular environment matters in making people who they are as biological and social beings.

That also isn't new. Of course the cellular environment matters in making people (for example, fetal alcohol syndrome is a clear example of that, way before epigenetics).

The fascinating thing about epigenetics is something else entirely. It is that there is inheritance of traits via a path that is not DNA. This is incredibly important and interesting, but was not what showed us that genes do not determine everything about organisms, we knew that long before. In other words, we already knew non-genetic factors determined how organisms are formed and behave, but epigenetics showed us that those non-genetic factors can be inherited.

The "genetic determinism" thing is just a strawman. No scientists believe that genes alone determine everything about a person. For example the heritability of height in the US is 90%~, and no one says that it's 100%.

I don't think the author of the article fully understands the concept of epigenetics. It has to do with inheritance of non DNA traits not nature vs nurture.

Very true. I'd like just add to 'non-genetic factors can be inherited': those same non-genetic factors can adapt much quicker than the genetic factors.

It is the combination of inheritable and quickly adaptable, that makes epigenetics such an important finding.

This nature/nurture distinction reminds me of dealing with bugs in large systems.

Some of the time it's the code (nature), sometimes it's the data (nurture).

I'd turn that round ! Code is nurture (can be fixed/altered), data is nature (it's there, and it just don't care).

hmmh... is that why Dutch people are tall? it's not because short people are selected out? If you gave a new generation of Dutch the diet/environment from 50 years ago would they revert to the height from 50 years ago? if not, why not? dumb questions no doubt but I don't really get how "inheritance by a path other than DNA" actually works, where the additional information gets stored.

DNA is data. A blueprint. But new organisms aren't made in a factory, they're made by other organisms, which were made by other organisms, etc. So when environmental factors affect which parts of the blueprint are used in construction of descendant organisms (and we know they can), those organisms can be different even though they share the same blueprint. Think of it as data implicit in code execution state, as children literally fork off the parent process and continue their lives.

Makes sense...what I don't really get is how that means 'heritability' other than through DNA.

Seems to me DNA is kind of like a recipe for beer, and on a humid day the beer comes out a little different with the same recipe. But if you go back to the original weather the beer will come out the same.

So if you sent a Dutch couple of today back in time to the diet and environment of 50 years ago, their kids would be the height Dutchmen would have been 50 years ago, since the DNA and environment are the same. Nothing else changed inside the Dutch that would make their kids taller, and the change in height isn't something that is 'heritable' through other than DNA.

Methylation of cythosines decrease transcription rate. Meaning a gene becomes less active. Methylation state is heritable.

For example, the changes can be other molecules aside from DNA. Like something that attaches to DNA and prevents a gene from being expressed.

What makes such a thing interesting is if it persists through cell divisions, and especially interesting if it is passed down to the next generation. And it turns out that, surprisingly, such things exist.

Would this non-DNA path for inheritance be the proteome rather than the genome?

The article's core point is a misunderstanding of the science and essentially disproved by the fact that you can use epigenetic decorations to fairly accurately measure age [1]. A lot of epigenetics is common reactions to common circumstances, to the point at which the patterns are very similar between people. Sure everyone's a special snowflake, just like every lawnmower rusts in a different way, but we also all age due to exactly the same underlying mechanisms, our metabolisms all operate on the same set of mechanisms, and so forth. Personalized medicine via epigenetics is the sort of thing you focus on when you can't outright cure a medical condition by completely repairing the damage that causes it. It is a consequence of inability to intervene comprehensively, to the point at which the research community scrabbles for any handhold that says something about the condition's progression, looking for comparison points that are mostly similar but just a little different to try to learn what is really going on. But it isn't the path to the next generation of medicine, for all that it is the path to getting your grant funded today.

[1]: http://newsroom.ucla.edu/releases/ucla-scientist-uncovers-bi...

There is no "lesson from epigenetics", and while I'm somewhat sympathetic to the author's politics I'm not about to try to claim they in any way follow from claims that "epigenetic diversity is fundamentally different from genetic diversity or simple plain old phenotypic diversity."

Not everyone is the same height. We've learned over the past century that doing our best to accommodate diversity results in a more interesting, happy and loving world. The source of diversity doesn't matter. We've also learned that giving people the freedom to make their own choices about their own bodies is generally better than giving that power to someone who has less information, like a legislator or a judge.

None of this has anything to do from epigenetics, and there is simply no political lesson to be had from it, any more than there was a political lesson to be had from Darwin, despite claims to the contrary by everyone from Spencer to Marx.

Being friendly with a few full-time computational and bench geneticists, I can tell you that none of them say things like "Epigentics has shown us that..." with great confidence, much less make sweeping claims about multi-generational heritable traits and clear cut environmental determinants.

Good grief.

> Indeed, most of the illnesses and conditions that are the focus of epigenetic research, except for cancer, are contested conditions. One reason they are contested is that many of them have come into being through measurements and statistics that effectively define illness as a deviation from a norm rather than as an underlying pathology. Just as type II diabetes is now defined by elevated blood-glucose levels (rather than a set of symptoms), the Body Mass Index (BMI) clinically diagnoses obesity as a larger than average (or what used to be average) weight-to-height ratio. It does not actually measure adiposity; in fact, a high BMI can be the result of large amounts of bone or muscle as easily as fat. Meanwhile, IQ has long been under attack as a culturally biased and non-objective way to measure intelligence.

Type II diabetes: does the author not believe a disease can be present before noticeable symptoms?

Obesity: the shortfalls of BMI are well-known; I doubt there are a lot of doctors out there who see an athlete with a high BMI and think, "time to burn some of that muscle"

Brain damage: the author is attacking their own choice of metric for the condition, in order to... deny the condition's existence??

Oh wow, this is one of the few instances where I really wish one could downvote submissions. What a whole bunch of blatant misinformation and misrepresentation of the literature. The authors are clearly non-specialists and it shows...

If this is not academic malpractice, I don't know what it is.

Reply on news.ycombinator.com