An Epigenetics Controversy
May 7, 2016 12:07 AM   Subscribe

The somewhat hysterical denunciation of The New Yorker as another head on the hydra of postmodernism seems to me to weaken that piece.
posted by thelonius at 1:23 AM on May 7, 2016 [11 favorites]

The somewhat hysterical denunciation of The New Yorker as another head on the hydra of postmodernism seems to me to weaken that piece.

Yeah I thought that too. Maybe it's... a magazine trying to tell a story (which may well be inaccurate). I figured out why he has a chip on his shoulder about this - he's a scientist who wants to do philosophy!
posted by atoxyl at 2:01 AM on May 7, 2016

He [in the "searing criticism" link] may be right that the article has errors or is misleading about the state of that scientific field, but the sudden pivot to a broader critique of the magazine as being insufficiently scientistic confuses that issue.
posted by thelonius at 2:12 AM on May 7, 2016 [1 favorite]

Field in its infancy + wild assed speculation = scientist grar!!!

When somebody first created a Bose Condensate, I thought they were the most brilliant person on the Earth, not because they had created a Bose Condensate, but because, when some interviewer asked what it was good for they pointed out all the things folks said a laser was good for in the 50's and 60's and what we actually use them for today and then said, "So, to answer your question, I don't know."
posted by Kid Charlemagne at 6:01 AM on May 7, 2016 [25 favorites]

I'm biased by my loyalty to the Tribe of Journalists (in which capacity Mukherjee wrote his New Yorker piece). But I'm persistently surprised at how scientists criticizing popular portrayals of their work seem not to reflect upon the possibility a) that their scientific expertise doesn't automatically make them experts in narrative journalism; and b) that their scientific expertise might actually mean they're among the worst people to objectively assess its presentation in popular articles.

Jerry Coyne's contributions to this controversy are characteristically intemperate, but the point-by-point critique by Ptashne and Greally is more interesting in this regard.

Their criticism opens by assuming that the phrase "Allis began to tackle a problem that had long troubled geneticists and cell biologists" means that nobody had tackled it previously, which makes no sense. They spend a significant amount of words criticizing Mukherjee for things clearly attributed to his subject, Allis, as if Mukherjee must be fully endorsing them by quoting them. They specifically attribute the phrase "epigenetic code" to "the author", when it appears between quote marks attributed to Allis. And their final "criticism" is actually an agreement with Mukherjee that parts of this topic are highly speculative. Finally, Coyne appends an addendum about how this kind of journalistic speculation annoys him. He is of course fully entitled to that subjective viewpoint, but it shouldn't be confused with expert criticism.
posted by oliverburkeman at 7:00 AM on May 7, 2016 [19 favorites]

some interviewer asked what it was good for

Gladstone once asked Faraday what electricity was good for.

"Some day, Sir, you shall tax it."
posted by Segundus at 7:52 AM on May 7, 2016 [15 favorites]

huh, I read all of that, totally thinking of Jerry Fodor in my head instead of Jerry Coyne. I thought it was weird that Fodor looked so young, and that he would have a blog about why evolution is true.
posted by obliterati at 9:07 AM on May 7, 2016

I'm as annoyed by Coyne as anyone else here but, bottom line, a New Yorker piece that peddles Lamarckism is a problem. It came up in another thread a couple of days ago and I was all, whaaat, where the hell is this person getting this stuff from? Now I know.
posted by Ivan Fyodorovich at 9:19 AM on May 7, 2016

From Mukherjee's article
My visit with Allis had ended on a cautionary note. “Much about the transmission of epigenetic information across generations is unknown, and we should be careful before making up theories about the kind of information or memory that is transmitted,” he told me. By bypassing the traditional logic of genetics and evolution, epigenetics can arouse fantasies about warp-speeding heredity: you can make your children taller by straining your neck harder. Such myths abound and proliferate, often dangerously. A child’s autism, the result of genetic mutation, gets attributed to the emotional trauma of his great-grandparents. Mothers are being asked to minimize anxiety during their pregnancy, lest they taint their descendants with anxiety-ridden genes. Lamarck is being rehabilitated into the new Darwin.

These fantasies should invite skepticism. Environmental information can certainly be etched on the genome. But such epigenetic scratch marks are rarely, if ever, carried forward across generations. A man who loses a leg in an accident bears the imprint of that accident in his cells, wounds, and scars, but he does not bear children with shortened legs. A hundred and forty generations of circumcision have not made the procedure any shorter. Nor has the serially uprooted life of my family burdened me, or my children, with any wrenching sense of estrangement.
If anyone is "peddling Lamarckism" it's not Mukherjee. He is one of the most level-headed and accurate popular scientific writers I've ever come across, and it doesn't hurt that he is an amazing prose-writer. I think the main criticisms of his article is that he doesn't discuss the experimental work of Jacob and Monod on transcription factors, and instead uses the shorthand of genes being "turned on and off" because to go into TF would require a great deal of extra explanation, making the article longer and losing potential audience with every paragraph.

These are the environmental hazards of scientific journalism - you can't go into every last bit of background. Sometimes when I try to discuss what I do with a relative I have to think really hard about what on earth any normal person knows about genetics and cell biology. Hint, it's not that much. You have to simplify or you're never going to explain anything. These scientists are just being curmudgeons. There are plenty of other much more bombastic science journalists writing untruths about epigenetics that can and should be attacked for their wild speculation.
posted by permiechickie at 9:34 AM on May 7, 2016 [14 favorites]

Well, obviously it would help things a lot if I RTFA.

posted by Ivan Fyodorovich at 9:57 AM on May 7, 2016 [3 favorites]

oliverburkeman & permiechickie, I agree with your points, but there's a much more general issue with this article that Coyne mentions (see Geoffrey North's comment). When you talk about the types of variation described in the article (twins, wounding, cell types, ant castes), you are talking about of the amazing things that gene regulation can do. There are many ways to affect gene regulation, which roughly break down into the genetics (stuff binding DNA and transcribing it, which can then be modulated) or the structure (physical changes to DNA affecting binding and transcription). The article makes it seem as if all of gene regulation is due to the structural changes. This is baffling, as the mention of Yamanaka factors affecting cell type is clear example of regulation by transcription factors, NOT structural changes. It's simply a misleading depiction of gene regulation. Given the general focus of the article (twins, wounding, cell types, ant castes), it should have been about gene regulation more generally; it could have only required a brief description up top that listed the major types of gene regulation in addition to structural changes. Alternatively, it could have been written solely about structural changes (clearly what the author intended), but then it should have been much more focused to the subset of things we know are clearly due to changes in DNA structure, like imprinting or this experiment where they made soldier ants forage.

For fun, I'll just plug some relevant work of my labmates - I'd be delighted to send the full text to anyone who does not have access! Just like the rest of biology, there is a dedicated subfield of people studying structural DNA changes to explain variation in ants (as mentioned in the article). A number of papers have said that one type of structural change, DNA methylation, varies between reproductive (i.e., queen) and non-reproductive (i.e., worker) ants. This raises the possibility that workers and queens have different morphology and behavior because of methylation-induced changes in gene expression. Lured by this possibility, an extremely talented postdoc, Romain Libbrecht, did the most thorough study of DNA methylation in ants to date - for the first time using high quality controls and large numbers of independent replicates. Guess what he found? There are no methylation differences at all between reproductive and non-reproductive ants in the species we study. This was a huge surprise, as people have reported massive differences in other species! What Romain did next was amazing: he got access to the raw data for the other papers and found that they were using really questionable statistical methods with small numbers of replicates. Since he had collected so much data, he could take a subset of his dataset and run it through other people's statistics, and, viola, it would show that there are differences. But when we took the whole dataset, or used better statistical methods, all of the differences went away. Of course I'm biased, but I'm so proud to work with the people who wrote this paper. To me it's real scientific progress: you work your ass off to do the most thorough study to date, you find something unexpected, and then re-analyze everyone else's data and show that the whole field is built on sloppy work. It's sad, because he worked so hard and he didn't figure out what makes worker and queen ants different, but at least we now know it's probably not methylation. This type of thing happens all the time in these flashy new fields like epigenetics, which definitely contributes to the disgruntled reaction of researchers to articles like this one.
posted by Buckt at 1:12 PM on May 7, 2016 [24 favorites]

Epigeneticists, once a subcaste of biologist nudged to the far peripheries of the discipline, now find themselves firmly at its epicenter.

I presume he thinks he is stylin' it like V.S. Naipaul here but that is a rude sentence.

Writers be gentle to your readers!
posted by bukvich at 2:32 PM on May 7, 2016 [1 favorite]

This is baffling, as the mention of Yamanaka factors affecting cell type is clear example of regulation by transcription factors, NOT structural changes. It's simply a misleading depiction of gene regulation.

I've gone back to the paragraph about Yamanaka three times, and I don't think that's an accurate representation of the rhetoric.

What Mukherjee is doing in that paragraph is making a distinction (warranted or not) between proximal and distal cause of identity. This is a philosophical argument going on, not a purely scientific one. His thesis in the article is that one can biologically locate individuality in epigenetic state, as opposed to the other processes, namely gene regulation and cell metabolism, in that paragraph. This is the fairest way to read that paragraph.

Journalistically, the worst part of that paragraph is the sentence saying Yamanaka was inspired by the above reduction argument, of identity to epigenetic structure. Either Yamanaka was, or he wasn't. But there's no information backing this up—did the author meet Yamanaka? Was this from a paper discussing implications or motivations for doing a piece of research? None of this is given. That's why all of this becomes problematic. What part is the author putting forward his own views, versus journalistically tying together speculations and themes from certain groups of researchers? For lay-readers, this lack of disclosure can be problematic.

To the author's credit, the personal relational stories were quite poignant (at least for me), and the good news is that they don't need the speculations to be true to have that emotional force. There's little discussion of that aspect of the writing.
posted by polymodus at 2:54 PM on May 7, 2016

I have some questions about the criticisms:
And there is no evidence that coiling and uncoiling of DNA has a causal effect on gene activity.
What about Barr bodies in X-inactivation? Aren't they a pretty straightforward example of this?
There is no evidence, despite years of research, that nucleosome states can be “copied” for transmission to daughter cells.
I'm not an expert; most of my knowledge comes from a couple of readings of Alberts. But Alberts seems pretty clear in the section "The Mechanisms of Eucaryotic Chromosome Duplication Ensure That Patterns of Histone Modification Can Be Inherited" that exactly this happens.

And DNA methyltransferase, which copies methyl marks when DNA is copied, is also a thing, right? Why would it exist if it's pointless?

What I'm wondering: Is what we're seeing here two sides in an internal scientific fight that hasn't been settled, between "regulatory factorists" and "histonists", or has the science been settled since my 2008 edition of Alberts in favour of the "regulatory factorists"? And if it has been settled, what is all the machinery that copies histone and methylation marks for?
posted by clawsoon at 3:44 PM on May 7, 2016 [2 favorites]

polydomus, I agree with you about this thing of Yamanaka's inspiration, who knows what it was. I also agree that it's an emotionally impactful article - definitely no criticism there.

As far as this question of what the paragraph is about, it says "Perhaps the most startling demonstration of the power of epigenetics to set cellular memory and identity arises from an experiment performed by the Japanese stem-cell biologist Shinya Yamanaka in 2006 ... he and his colleagues figured out a way to erase a cell’s memory ... Most important, epigenetic marks were erased and rewritten, resetting the landscape of active and inactive genes." That's really inaccurate! What does "most important" even mean? Is there some experiment I don't know about that shows that epigenetic changes are more important than the Yamanaka factors that actually set off the whole cascade?

To this question of identity, that's a testable hypothesis! The test would be to take a cell of one type and give it all of the epigenetic marks of another type. Would it switch identity? No one knows - we don't currently have the technology to do that experiment. But what we do know is that transcription factors can change identity. So to say that identity resides in epigenetic and not transcription factors is really silly.

clawsoon, those are fun questions, but I don't know this field well enough to answer. I think part of it is a question of talking about histones vs. DNA methylation. Coiling and nucleosome states refer to histone modifications, which is a newer and perhaps less well-supported area. And yeah, methyltransferases are there, but their function is much more diverse than just gene regulation. But this gets at the debate a bit: we actually don't know what they do in insects, because they're mostly absent from Drosophila melanogaster, the main model organism for insect genetics. The fact that species can lack DNA methyltransferases are a mark against their primal importance.

As far as this being an internal scientific debate, I think you're right. There's a vocal minority who believe that structural changes to DNA really are the most important regulatory elements. I (and I think most people) take a holistic view: the whole cell is the cell. Epigenetic marks are important, as is the PH of the cytoplasm and the cell age and shape and transcription factors and the location of the cell and messages coming in from outside, etc.
posted by Buckt at 11:02 AM on May 8, 2016 [1 favorite]

« Older Kenm explains his craft   |   Radio discussion with Viet Thanh Nguyen, author of... Newer »

This thread has been archived and is closed to new comments