Unfortunately don’t have time to check out the paper, but vaguely I am aware that methyl groups are like an accessory that can be attached or detached (by the cell, normally) to an arbitrary section of DNA within your chromosome. And it can be used in this way for regulation of DNA expression. Because when the methyl group is there it’s harder for that section of DNA to be read (transcribed) because the methyl group is physically in the way. The train can’t run on the track because there’s a big rock on it

DNA is like code, but then there’s this whole meta level of DNA regulation that determines whether each section of code is used, and how much

So by demethylizing some section of the DNA, implicitly that means the scientists made that section of DNA be used more often. And it just so happens that when that section of DNA is used more often it contributes to various processes that ultimately end up with the plant organism as a whole producing more “yield”, or parts of the plant that we like to eat

https://en.m.wikipedia.org/wiki/DNA_methylation

In this case it's RNA not DNA.

Demethylating DNA is like uncommenting a line of code; demethylating RNA is like uncommenting compiled/assembly rather than the source.

Thank you for this incredible analogy.

Oops, thanks

Wow. Thanks for the explanation. It seems to me then that this result is a slam dunk Nobel. Yield improvement without the GMO stigma.

This is GMO. They added a human gene to rice and potatoes to get the improved yields.

People that dislike GMO aren't going to be any happier about adding _human_ genes to plants.

As long as it’s not round up ready or otherwise encourage more pesticid use I’m cool with it

Or making the plant produce pesticide of its own, causing bugs to develop resistance to measures needed by organic farmers, and also dosing humans with what would otherwise be easily rinsed off.

As long as this technology is not used to prevent country farmers to store their own seed whenever they want, and make it only legal to purchase 'Monsanto approved' seeds, I am completely fine with this.

I'm looking forward to the day when we can collectively disentangle GMO from all of the other negative practices of big ag. I don't see a reason why responsible genetic modification can't be a part of a sustainable and equitable future agricultural system.

"Genetically modifying food to increase yield is good for the environment" is a true statement, but you're going to need an awfully good PR firm to persuade the public it's true...

Yeah exactly. And not only yield, but natural pest resistance, etc. I'm also imagining how genetic modifications might eventually be environmentally beneficial in the context of whole-ecosystem engineering if/when we ever move beyond a monocrop model, such as modifying native organisms to counter invasive species, etc. Admittedly a bit sci-fi currently but in principle possible!

Can you explain the link between the methylation and adding that gene?

The gene that was added is translated into a protein that floats around the cell doing the RNA demethylation. The plants already have other genes that produce proteins that did that, but adding another one has an effect.

Edit: correction - thanks Scaevolus, my mistake.

Plants already have proteins that do this [1], but modifying them or adding new ones can affect phenotype as shown by this study.

[1] https://www.frontiersin.org/articles/10.3389/fpls.2019.00500...

I'm surprised the plant lives at all in that state... One would imagine RNA methylation is used for all kinds of logic and mechanisms in plant cells, so arbitrarily removing them all would break stuff.

It would be like a programmer saying "let's just comment out every line of code starting with "if" in the windows source code, and then see if it boots faster!

Now it makes sense, thanks!

How is removing (obfuscation!?) smth from the DNA not genetic modification?

In dealing with methylation one doesn't add to or remove the "code" to synthesize any particular protein from the cell(s) of the organism, the DNA/genes per se remains intact. The process inhibits / allows synthesis of particular proteins from the genes/DNA.

Think of it like software -- by flipping an A/B Switch in software configuration you don't add to or remove from a program, you merely turn on/off certain features. Same with methylation -- it's the A/B Switch for protein synthesis, FWIU.

No, they are changing DNA, they are adding DNA code that does the demethylation of RNA.

It's like adding code to the compiler so it produces different assembly from the same higher level code. But it's still added code because the compiler itself is in the codebase.

I think most lay people would view this as genetic modification, or at least "genetic manipulation", which might be even scarier.

Having said that, I'm pro-GMOs and always roll my eyes when brands go out of their way to say that they proudly don't use GMOs.

Technically correct might be the best kind of correct, but i doubt it will change hearts and minds.

It's enough to circumvent regulations for marking products as GMO.

The EU definition is "organisms in which the genetic material (DNA) has been altered in a way that does not occur naturally by mating or natural recombination". https://publications.jrc.ec.europa.eu/repository/handle/JRC5...

I'm honestly not sure which side of the definition this falls under. It certainly doesn't occur naturally.

Yeah, seems like a case for the lawyers.

Because it’s epigenetics! No novel gene sequences are introduced. Just a change in what natural genes are expressed and at what levels.

No. The FTO gene from humans was inserted into the genome of these plants. That gene produced a protein that performs RNA demethylation, which isn't even epigenetics, because epigenetics is the set of extra stuff that happens on DNA, not RNA.

Oh my bad I didn’t read the full article. Of course, insertion of a gene to do demethylation would be genetic modification.

How novel is the idea then?

Methylation groups are considered meta-genetic I thought. Sort of like dog-earring a base pair.

edit: epigenetic, not metagenetic

I can't explain the paper, but if DNA is genetic ROM, methylation is like one bit of RAM available on each digit [*].

Most of the cellular machinery tries pretty damn hard to make exact copies of DNA. The methylation bits, by contrast, get deliberately flipped on and off all the time.

[*] Technically only A and C can accept methylation, but in a double helix you'll always have either an A or a C on one side or the other at every position (A pairs with T, C pairs with G).

Deliberately flipped based on what?

That is an area of very active research.

I mean the exact chemical mechanism for the flip is some enzyme, and the flip certainly affects gene expression. As to why the cellular systems flip these bits, great question :)