Hacker News new | past | comments | ask | show | jobs | submit login
Atomic Gardening (wikipedia.org)
188 points by jeffwass 4 months ago | hide | past | favorite | 96 comments

This is a pretty brute force method for genetic engineering -- and yet it works! It's essentially just pressing fast forward on evolution.

But this is wildly inefficient. Plant genomes are often ridiculously large (e.g., the onion genome is 4X the size of the human genome), so you either have to throw a bunch of mutagen on the plant (which can cause off-target, toxic mutations) and/or plant a bunch of seedlings to get the plant you're looking for.

New methods often include introducing new genes with mutations through agrobacteria, gold nanoparticle bombardment, etc. without needing to actually mutagenize a plant genome. Do your mutagenesis on a specific gene in bacteria, and then test those random genetic variants in plants.

But for me, what is more exciting is a relatively new shift toward targeted mutagenesis directly in plants. For example, CRISPR can be used to target mutations to specific genes in situ (and not insertions or deletions -- useful point mutations through the use of base editors) [0]. I think the directed evolution of plants will be a pretty fruitful (lol) area of research in the future, and I'm excited to see where that field goes. Gotta love new plants!

[0] https://www.nature.com/articles/s41580-020-00288-9 -- see section "Mutagenesis and directed evolution"

Getting the mutation is not an issue, its understanding what the mutation does to the organism that's the hard part, and that still usually means planting a bunch of seedlings to screen for favorable progeny.

> plant a bunch of seedlings to get the plant you're looking for

This happens anyway, based on my understanding of fruit tree breeding at US land-grant universities.

Hop breeding works that way, plants are actively bred. The ones that are agronomically useful will see bigger and bigger plantings until they are tried at market. When they either fail or establish themselves (to the point of completely overtaking the market like citra did).

With wine it's similar so I suspect its just how most of this works.


It always amuses me to find that people are against GMO foods yet almost everything we eat is GMO depending on how you define it.

Almost every crop and livestock has been genetically modified through selective breeding (domestication).

Mutagenesis, which atomic gardening falls under, has been around since the 1920's using x-rays and chemicals and currently doesn't normally fall under most GMO definitions because its "traditional".

There is also profile guided selection, which is just selective breeding while using modern DNA profiling to make it more effective by watching genetic outcomes directly.

Then there is genetically engineered targeted gene insertion which is what most people consider "GMO" but hasn't been shown to cause any health issues, its just scary because it's done in a modern lab.

It's not that I'm against being cautious with genetically engineered organism (testing etc), but what issue are we solving by labeling an GE organism "GMO" vs a mutagenic one, or a selectively bred one? Is there any data to suggest one is more dangerous than the other? I seem to have a hard time finding it.

> It always amuses me to find that people are against GMO foods yet almost everything we eat is GMO depending on how you define it.

> Almost every crop and livestock has been genetically modified through selective breeding (domestication).

I agree with your broader point that much of what we eat is GMO. But isn't selective breeding a categorically different process of modifying an organism? My understanding is that selective breeding starts with two things that we already are familiar with, and tries to maximize or minimize some characteristic in the next generation via traditional reproduction. On the other hand, a modern definition of GMO includes genetic engineering -- manipulating DNA directly using technology.

Is it naive to worry that the latter process might introduce unexpected and potentially harmful characteristics, in ways that the former wouldn't? Additionally, if selective breeding happens slowly over generations, it seems to me that it would allow us to more carefully track how the modification of particular traits affects the surrounding ecology of the organism.

It’s interesting to read Greenpeace and other environmental groups’ writings to see what they are objecting to.

It seems they are primarily concerned about two types of issue; one is why I’d call “safety”: gene transfer/escape into wild species, unintended ecological consequences, human safety of consuming GM plants. Tho other group is around “licensing”: essentially a deep concern about going down a path where plant seeds are licensed instead of bought, and all the “Monsanto is evil” concerns.

I think on the safety point there is some FUD about what kind of gene transfer is actually biologically possible, but you can make a more general rebuttal that selective breeding has the same risks, it’s just harder to get positive changes into your genotype. Maybe Greenpeace would argue that we’re already endangering the biosphere a bit with selective breeding, and going any faster would dramatically increase the danger.

On the licensing point, I’m very sympathetic, but I think that is an argument for IP law reform or public investment in GM, not an argument against GM itself. But of course if you believe the safety concerns then you don’t want public or private research to happen because you think it’s prima facie harmful.

> Maybe Greenpeace would argue that we’re already endangering the biosphere a bit with selective breeding, and going any faster would dramatically increase the danger.

This is an interesting point. The practice of using GMOs, including the broad category of selective breeding, could ultimately lead to much less biodiversity which is, at best, not fully understood by us, but seems to be quite bad for life.

Already has led, although I guess the bigger effect is agriculture itself – basically giving those few genetically homogeneous cultivars a vast selective advantage by converting biologically diverse natural land into monocultural fields on a vast scale.

Conversely, it could lead to much greater biodiversity as we control the process.

Agreed. In principle we could introduce deliberate biodiversity, and as it becomes cheaper to create new variants, it would become easier to find multiple genotypes that achieve the same desired adaptation. You could even imagine doing so programmatically, perhaps even with explicit inputs describing the known issues that need to be optimized around (e.g. Panama Disease in bananas).

Unfortunately I am not optimistic that this is an inevitable outcome, at least not in the short-to-medium term; if you look at the software ecosystem or even just corporate structure in general, you see a strong pressure towards consolidation and monoculture rather than an emphasis on robustness to systemic risks.

However I think it's encouraging that advances in GM technology provide us with a potential long-term solution to the problem of monoculture, without asking us to give up the benefits that the market selected for in our current state. Of course we could solve monoculture in our current system by giving up the cultivars that the market prefers, and instead switching to more-localized and less-productive cultivars, but this doesn't seem like a realistic way of feeding 7-8B people.

The Cavendish banana is a common and well known example of this.

GMO plants are modified to make them more difficult to propagate and produce seeds, in order to protect intellectual property. The ecological consequences of those variations being released into the wild is devastating and has adversely affected farmers who do not grow GMOs.

> GMO plants are modified to make them more difficult to propagate and produce seeds

Can you name a single commercially available GMO that has this property, aside from being a hybrid (hybrids don't breed true, but that's because they're hybrids, not because they're GMOs)? It is my understanding that there's no real need for any such technology anyway, as legal protection suffices.

I thought this was common knowledge, but I looked for a citation and couldn’t find one.

https://www.seattletimes.com/business/dupont-to-pick-up-wher... suggests that roundup-ready soy seeds are viable. So perhaps they rely solely on IP for protection.

I think if you drill down on many (all?) anti-GMO arguments you will find the evidence for them really isn't there. It's remarkable how the whole ideological edifice is built on nothing.

Occam's razor suggests the best strategy - if available - is to suppress natural reproduction, if you were in possession of a superior genetic strain, in order to stay ahead of the market. Suppressing natural reproduction is the same general principle as suppressing others using your idea/strategy/copyright.

I'm sincerely curious why this seems so far fetched to you. If I were the CEO of a company having the option to nudge my researchers into this, I probably would, seen from a ethics-detached point of view.

By the way, I'm not anti-GMO; in fact I think it's the only way to sustain the whole of humanity until we've passed the next great filter. Still I don't buy into that line of thought.

> I'm sincerely curious why this seems so far fetched to you. If I were the CEO of a company having the option to nudge my researchers into this, I probably would, seen from a ethics-detached point of view.

Not the person you're responding to, but there are three main responses I can give:

1. By and large, for most of the food crops that we eat, what we are eating is itself the seed of the crop. Tinkering with the reproductive elements of the crop has a higher chance of screwing up desirable properties of the food part of the crop than, say, introducing resistance to glyphosate.

2. In a related topic, genetic engineering is rarely so competent as being able to do something like "prevent reproduction." Seeing genetic manipulation as being basically like designer babies is pure science fiction; the more complex the demand is, the further in the realm of speculative nonsense it is.

3. On a related topic, there's the impact it has on the production costs of seedcorn. The easiest (cheapest) way to make anything biological is to use natural biology to make it, not try to create it in vitro. If I'm designing seeds, I'd rather make a couple of seeds using the expensive in vitro steps, plant them in a plot for a few generations, and sell their offspring, instead of trying to sell all the seed designing it in vitro. Inhibiting natural reproduction inhibits this method of making the product.

No one is holding a gun to farmers' heads forcing them to buy GMO seeds. If a particular kind of seed doesn't increase their profit they won't buy it. The older non-GMO seeds will still be around. They won't be outlawed.

So, the scenario you seem to be thinking of is that GMOs will drive down the price of agricultural commodities, so that farmers who don't avail themselves of them will be unable to compete. But this argument "proves too much": it's an argument not against GMOs, but against ANY agricultural technology that boosts yields and reduces costs. Presumably the Green Revolution is not something we should object to because food prices declined.

I think you are on to something in understanding the anti-GMO motivation in Europe, though. Farmers don't want to have to compete against the more efficient large scale farmers in the US. Anti-GMOism is a kind of camouflaged protectionism.

I think the worry is that you get the biased against reproduction genes that make a lot of sense for a for profit GMO plant that hop to normal plants and reduce their ability to reproduce.

Not saying that's the concern just my impression.

If it reduced the plants' ability to reproduce, surely that would not spread in a population. So that seems no different from any other deleterious mutation. Such mutations happen all the time in any population.

Your argument there does apply to non-Mendelian inheritance ("gene drive"), so banning modifications to crops that would be engineered to cause that seems prudent (although I'm not sure why anyone would want to do that). Modifications to pests, on the other hand...

The argument against IP protection for GMOs makes no sense to me at all. If there were ever a technology for which IP protection was appropriate, it's this one. The plants literally manufacture themselves, so if the IP wasn't protected the investment in the creation of the variety could not be recouped. There would be one season of sales then the market would be gone.

I think it’s immensely problematic that a seed can fall in your farm, and you can be sued for illegally growing Monsanto’s IP. More generally I think it’s a bad societal outcome for one company to use IP laws to monopolize the food production economy. I think allowing this will lead to a much less robust society.

I’m all for free markets for most markets, but things get very bad if people don’t have food. So I think there is a good case for the government funding research and open-sourcing the work here, and also potentially restricting eligibility of patents in this area too.

I think your argument is sound in a free-market maximalist / small-government libertarian framework, I’m just less convinced that the current IP framework produces good results in this case.

> I think it’s immensely problematic that a seed can fall in your farm, and you can be sued for illegally growing Monsanto’s IP.

It would be problematic if that's the case. That is not the case however. What you're probably thinking of is https://en.wikipedia.org/wiki/Monsanto_Canada_Inc_v_Schmeise..., where the farmer originally claimed to have had accidental contamination of the Monsanto seeds (this defense was dropped before the actual trial itself, and if you read the actual depositions, it's really hard to actually believe this claim). It's difficult to find any other case that comes close to falling under the claimed situation here (due to the notoriety of this one case), but I don't think that any of them actually do.

Incidentally, I should point out that the underlying issues of plant patents are actually far older than GMO cases (early-mid 20th century). It's not anything actually new.

That scare scenario has never occurred, and it's legally doubtful that any such lawsuit by someone like Monsanto could succeed. Monsanto itself explicitly said it would never sue anyone for accidental contamination.

What Monsanto DID do was sue someone whose field was contaminated, and who then repeatedly sprayed the field with glyphosate to kill all but the GMO plants, so he could selectively concentrate the interlopers to continue to propagate. It's this last step that got the farmer in trouble.

> but things get very bad if people don’t have food

GMO IP protection does nothing to prevent farmers from continuing to grow previously existing varieties. So this is another scare argument without foundation.

> I think your argument is sound in a free-market maximalist / small-government libertarian framework, I’m just less convinced that the current IP framework produces good results in this case.

I think the anti-IP argument is an underhanded way to try to ban GMOs without actually saying you want to do that.

There have been hundreds of lawsuits by Monsanto. I don't immediately recognise which case you're talking about. Do you have any links please?

> GMO IP protection does nothing to prevent farmers from continuing to grow previously existing varieties.

If they buy seed from grain elevator which comingles GMO and non-GMO seed (which is what happened in the Bowman case) the farmer can't sell what they've grown from that seed.

Here's the case I was thinking of (it was canola, not soybeans):


As for Bowman


Bowman didn't just do that. He applied glyphosate to the field. So, he bought seeds without a license to grow them, with the intent to exploit the fact they were resistant to glyphosate. This violated the patent.

I don’t think you’re engaging with me in good faith here. I’m making my own case, and clearly; I’m not trying to make an underhanded other point.

I’ll tap out here. Have a nice day.

I view the entire anti-GMO position as being not in good faith, so I have little patience when those making the anti-GMO arguments trot out the same well-debunked (or a priori nonsensical) talking points.

Again mutagenesis I would consider "manipulating DNA directly using technology" its just using a less targeted and more random approach.

Every variation of life today including poisonous and deadly plants and animals came from random mutation over time.

For example tomatoes and eggplants and peppers are closely related to tobacco and to deadly nightshade. How likely would a random mutation create a poisonous version, or one that is just poison to some people. Many people are allergic to them already.

Artificial selection with artificial mutation speed that process up greatly, targeted gene editing even further but in a more controlled way.

Which process is more likely to produce harmful side effects? Is there data to warrant engineered organisms are more likely or is it just unwarranted fear?

> Again mutagenesis I would consider "manipulating DNA directly using technology" its just using a less targeted and more random approach.

Right, I would put mutagenesis in my second category of "modern GMO". (Edit: I interpreted your use of mutagenesis as the controlled use, as in the original topic of this post. Maybe that's not what you meant?)

> Which process is more likely to produce harmful side effects? Is there data to warrant engineered organisms are more likely or is it just unwarranted fear?

That's essentially what my question was, for traditional artificial selection (selective breeding) versus genetic engineering. I wasn't making a point one way or another; it was a genuine question.

The point about random mutations occurring naturally is fair. Perhaps part of what scares people is simply the speed of the process, combined with not really knowing or understanding what technology is being used in the labs.

Selective breeding can also introduce harmful characteristics. Many modern cultivars that have been selectively bread for commercially favorable traits for decades lack the hardiness found in landraces, for instance. Species also are mutating all the time randomly.

I'm guessing it's more about transparency. Monsanto using a proprietary genome for the sake of increasing yield at scale seems fundamentally different than a local farmer who selectively breeds to maintain some heirloom crop. I think the former is much more of a black box and has more potential for something unexpected to occur than the latter.

Would both of your examples be labeled as gmo?

GMO implies a lab process with specific alteration of genes, while traditional plant breeding to traits only exposes phenotypes as indication of results, and relies on the pre-existing variance in species to provide new traits.

>It always amuses me to find that people are against GMO foods yet almost everything we eat is GMO depending on how you define it.

What I find problematic about many "anti-progress" stances is the (very false) mythos around a return to a purer past when we didn't do things in such supposedly "corrupt" ways. That stance is certainly prevalent in much of the anti-GMO following but I would hesitate to tar over it all with the same brush.

The second falsity is that the primary concern with GMO is individual human health.

Much of the more informed anti-GMO stances consist of the following two views:

1. The view that the past practices you mention in your comment were damaging to ecosystems and while the horse has bolted with many of the products of those packages (and reaping the fruits of those now is not too problematic) the negative impacts of those practices can be learned from to avoid future damage.

2. That it's a nuanced topic fundamentally about making informed decisions about the lifecycle impact of GMO based on current scientific knowledge rather than a blanket dogmatic opposition to the concept as a whole.

The second point in particular is absolutely not taken into consideration in current GMO practices: widespread corruption and disregard for impact analyses is well documented.

Techniques like CRISPR are at least 1000 times more powerful than selective breeding. That's why we use them.

Given the food industry don't give a shit about nutrition and have spent decades pushing sugar and fat and 101 similar things, why would you trust them with tools 1000 times more powerful than they already have?

I'm opposed to gmo not because I think it will magically make me grow another head or something. I'm opposed to it when it doesn't offer anything good for me as a consumer.

There’s two types of people who are against GMOs. Those that are afraid of eating genetically modified organisms, and those who take issue with how GMOs are used in practice. In practice, the most common GMOs (I believe) are Monsanto’s Roundup Ready crops. The point of that crop is that farmers can spray huge quantities of the herbicide glyphosate in to the soil to kill all plants in the field without damaging the crop. Monsanto swears this is safe but there are many reasons why this practice puts workers and eaters at risk and seriously damages ecosystems.

So I would love to know what plants are GMO and specifically what strain is used and why. This would help me avoid GMOs that are used to damage farm ecosystems. This is also why I look for certified regenerative organic products - I want the food I eat to regenerate the soil, not destroy it.

> Monsanto swears this is safe...

Not anymore. The lawsuits challenging its safety ended to the tune of a 10 Billion dollar loss.

I'm in your boat. Gene's are letters in a story. To say "That's a salmon gene! It doesn't belong in my tomato!" is puzzling, like "That letter belongs to another book! Keep it out of this book!"

Anyway, good ol' paleo food was whatever nature provided, most of what was happy to kill you. Humans tamed 1000's of plants over 50,000 years changing them beyond recognition so they wouldn't hurt us with alkalis and poisons. And so the tasty bits were bigger and easier to get at. All by genetics.

Sure we can do it faster now. But we're aware of what we're doing. Grandparents with crossbred tomato didn't have any idea if she was reintroducing a poison or raising the acidity to dangerous levels. They were blundering in the dark.

I see no problem with turning on the light, and moving confidently.

This is oversimplifying in the opposite direction in my opinion. The important thing is not the letters, it's how they manifest -- the way they're interpreted as instructions. So the objection is to taking a paragraph of Ulysses and putting it into The Grapes of Wrath. Or even better, selecting a step from the middle of How to Build a Birdhouse and inserting it into the directions for crocheting a tea cozy.

If properly chosen, it could certainly have a beneficial effect. But DNA is also a much more complex system than a book.

> Gene's are letters in a story. To say "That's a salmon gene! It doesn't belong in my tomato!" is puzzling, like "That letter belongs to another book! Keep it out of this book!"

Using the same reasoning we could pick some pieces from an apple computer, mount it in a PC and expect that will fit in place and work flawlessly because, well... plastic is plastic. Right?.

It does not work like that. Organisms has been designed and polished for millions of years to work as a whole, all expensive useless stuff has been pruned for good reasons, and yes, this salmon gene -most probably- will not belong to a tomato... because is part of a salmon.

-> Organisms have been...

Since I share at least 60% of the DNA with about anything I eat I don't worry too much about mixing one organism with another genetically bad things can happen mixing close relatives or distant ones and I am not sure one poses a higher risk than the other.

DNA is DNA, be it fragments or full genes, and regardless of the sequence, our bodies are pretty damn good at breaking down DNA we eat without directly incorporating it into our genomes.

Would be pretty interesting if that weren't the case though; quite literally "you are what you eat."

> Mutagenesis, has been around since the 1920's using x-rays and chemicals and currently doesn't normally fall under most GMO definitions because its "traditional".

Not, is because is a different technique. Hybridizing two closely related species is a thing. Introducing the gen of a fish into a tomato is another totally different, and mutating the gen of one species (often with poor results) is a third thing. GMO mix the genetic code of species that can be even in different kingdoms.

There is also an international consensus into not including selective breeding in the same category as genetically modified, for similar reasons. Would be confusing.

The so called atomic gardening had produced several fancy and really ugly cacti, but not much more really. Plays in a much lower league than hybridization.

> Mutating the gen of one species, often with poor results, is clearly different than introducing the gen of a different species into other that can be even classified in a different kingdom.

Genes aren’t inherently kingdom specific, and mutagenesis can produce genes that don’t even exist in nature; moreover, it produces changes all over the genome; when you then select for a desired trait, you still have unrelated changes all over, whereas there is a whole lot more control and knowledge of what you have with transgenics.

Yeah, they are different techniques, but it is insane that mutagenesis (especially in its modern forms) is not treated as even a moderate concern by the same people who hold up transgenics as an existential threat.

The statistical probability of a crop randomly mutating a gen that would express a new complex alkaloid from scratch is close to zero (We know it because hasn't happened often in the last 3000 years)... but can be easily done with GMO.

To blend all those different techniques under the same term would be useless and confusing to both science and consumers.

> The statistical probability of a crop randomly mutating a gen that would express a new complex alkaloid from scratch is close to zero

That's true if most mutations produced through and then artificially selected for in modern mutagenesis based techniques, as well as most of the mutations irrelevant to the targeted trait that come along for the ride.

> We know it because hasn't happened often in the last 3000 years

3,000 years isn’t all that long and we don’t know that, what we do know is that no mutation that does that has occurred where it has been in net of cost to the organism useful enough to be selected for an retained, which is a very different thing.

> To blend all those different techniques under the same term would be useless and confusing to both science and consumers

So? The post you respond to acknowledges that they are different but notes “it is insane that mutagenesis (especially in its modern forms) is not treated as even a moderate concern by the same people who hold up transgenics as an existential threat.”

That it is comparatively easy to do a precise, targeted change where you know exactly what new thing will be expressed and change little else through transgenics isn’t in dispute, its the whole reason for that statement.

> Genes aren’t inherently kingdom specific

Some not. We share common mechanisms with other life beings, but most of the time, Yes, they are. You will not find a single animal able to photosynthesize by its own means.

> mutagenesis can produce genes that don’t even exist in nature

Because they are broken. DNA "per se" is useless unless it can be read and used to build something useful. Mutated genes are like changing a paragraph from a book to a language that does not exist and can't be translated. Most of the times the result will not made any sense. Evolution will eventually clean the mess, as usual.

GMO in the sense most people mean it (direct genetic modification) almost always also means pesticides.

I think that

That's not true for bt-corn or golden rice.

Organic farming also means pesticides. All farming means pesticides.

Oh, are you an expert on farming?

I have never heard of that connection. I was under the assumption anti gmo folk were against modified rice that is super drought resistant.

Sure there are some anti-science people out there, but most people I've talked to are worried about pesticides.

GMO plants are designed to take much higher doses. That is a concern both environmentally and as a potential health issue.

One use if GMO is to engineer a plant, that is resistant to a specific herbicide, which is typically sold by the same company (see e.g. Monsanto and Glyphosate).

Another use is to engineer plants that produce insecticides themselves, which raises the concern if those insecticides affect other organisms as well.

So is mutagenesis "direct genetic modification"?

How does GMO mean pesticides? I thought that was more under "organic" which is again a very ambiguous and problematic label.

I think the poster may mean that some plants have been modified to resist pesticides sprayed on them and folks may not want to eat these plants.

The real problem with GMO is entirely political. Plants with proprietary DNA engineered to ignore a soil-persistent wide-band herbicide. Short term crop productivity rises, forcing farmers to follow suit. Next, prices drop, costs rises, but you can't go back because fields now can't grow anything else for years. It's agricultural lock-in. Add insecticides that destroys pollinating insects for maximum effectiveness...

Let me do my best to convey some potential concerns. To say that they’re all GMO so what’s the problem is sort of either an ignorant or disingenuous overloading of the term. Sexual reproduction in a way sort of fits with that category (I mean it’s just selective breeding crowdsourced). But if we restrict GMO to only mutations, yeah there are actually legitimate issues.

Mutations occur naturally, and overtime nature has evolved methods to withstand this. Like the one thing that all organisms share, DNA-wise, is the code to faithfully copy DNA. You literally never see mutations in this code because nothing survives. Beyond that we do see random mutations all the times, but they’re usually harmful or at best don’t do much. Then occasionally there are beneficial mutations but they’re very rare and they happen over long periods of time. Atomic gardening is using this exact process albeit speeding it up bu increasing the speed that mutations occur.

On the other hand, genetic engineering is selective in a way that there is no natural precedent for. We literally have no idea if it’s not harmful or not. I’m not trying to say this to sound alarmist I just mean philosophically speaking you can’t prove that.

I’ll add as an aside that I don’t think the condescension that you see all the time regarding GMO is a food faith argument. There are plenty of scientific reasons to give pause to testing this stuff out in production.

As to concrete, practical reasons why we should give pause before engineering life with such wild abandon, I’d say recombinant DNA is one, but also so is the danger of just good ok’ mono-crops.

Now let me explain that I’m not personally and categorically anti-GMO. I think it’s wonderful technology that we can do so much with. I just think it’s dumb for people who don’t have degrees in genetics to go on the internet and say that people who have any concerns are “amusing” or stupid or whatever because science. The truth is it all depends abd there’s absolutely zero reason to think the GMO can’t go horribly wrong as some kind of rule.

I’m on my phone so some of this may read weird but I hope my point comes across as I meant it. Basically science is a philosophical branch rooted in skepticism.

One last thing I’ll say is there’s a bit of selection bias happening. You never hear from scientist who raise legitimate concerns because industry has a huge influence in people’s ability to do research and this their careers.

So in summary, mutations happen all the time in nature. Some “good” some “bad”. There are dangers but there’s been billions or years of evolution to buffer against it. Targeted genetic engineering like CRISPR is only the same un that there’s mutations but that’s where the similarities end. Like natural mutations they can be good or bad. There’s nothing unscientific or amusing about giving a second thought if maybe there are additional differences yet to be seen downstream. To say otherwise is categorically unscientific. We should absolutely proceed with caution and there are absolutely problems like the ones I listed. I mean, many of the people here are programmers. How many of us trust any code?

It’s a spirit of temperance and skepticism that im putting fourth, and I actually think the burden of proof should lie on a case by case basis.

Does that make sense?

I'm against GMO because 99% of the time it means I'm eating the pesticides that the plant was engineered to resist.

The problem with GMO is that you end up with varieties that don’t naturally come into balance with people and the environment, especially since the reality is that GMO implies large scale industrial farming with crop monocultures. People don’t want to experiment with their lives and discover decades later that some obscure ingredient in a pesticide has some long term effect. Nor do they want to have issues like water tables dropping and aquifers getting depleted because the high yield crops are not sustainable with a region’s water cycle. Cooks are worried about losing the diversity of crops that support their culinary diversity. Farmers are worried about getting bullied and economically enslaved by seed manufacturers. These are all risks - and your evaluation or perception or appetite for risks may be different - but I don’t think it’s illogical for people to be wary of GMO or want to take it slow, especially when operating in a world where humans regularly get things wrong and where human processes (like government approvals) can be corrupted.

We have large scale industrial farming and monocultures without GMOs. We could have GMOs without large scale industrial farming and monocultures. So what you are saying really doesn't make sense.

These sorts of scattershot incoherent arguments are a red flag that the position being defended wasn't arrived at by honest reasoning.

I heard about this on the radio and was surprised I hadn't heard of Atomic Gardening before. Basically, farmers and gardeners would expose large quantities of seeds and plants to varying degrees of radiation, and look for useful mutations.

Rio Red grapefruits, accounting for 3/4 of Texan-produced grapefruit, were originally created by Atomic Gardening! From the Grapefruit wiki, which links to the Atomic Gardening one :

"Using radiation to trigger mutations, new varieties were developed to retain the red tones that typically faded to pink. The 'Rio Red' variety is a 2007 Texas grapefruit with registered trademarks Rio Star and Ruby-Sweet, also sometimes promoted as Reddest and Texas Choice. The 'Rio Red' is a mutation-bred variety that was developed by treatment of bud sticks with thermal neutrons. Its improved attributes of mutant variety are fruit and juice color, deeper red, and wide adaptation.[21]"

[20] https://www.nytimes.com/2007/08/28/science/28crop.html?_r=1&...

[21] https://mvd.iaea.org/

The Rio red variety was not developed in 2007. It was developed in the 1980s, in, you guessed it: Rio, Brazil. When I first heard about this some handful of years ago I was naturally curious about how a researcher in Rio in the 1980s had access to thermal neutrons without a fission reactor. After a lot of reference digging and discussion on physics forums we came to the conclusion it was most likely a penning trap style fusor with some thick moderator. The Mars Curiosity rover also had/has a penning style trap for generating neutrons to probe for underground water sources.

What makes you think "Penning Trap"? Neutron generator tubes have been used since forever in the oil well logging industry. They use an electric field to accelerate deuterium ions into a solid target containing deuterium or tritium.

As far as I know, fusors have never been commercially successful as neutron generators.

Also, why would one use neutrons for plant mutagenesis? Photons are far easier to make. An x-ray tube will produce a given (biologically effective) dose at a very small fraction of the cost of a neutron generator.

The Curiosity neutron source is indeed a conventional pulsed sealed-tube neutron source, not a Penning trap.


Sorry I wasn't clear. The DAN and other pulsed sealed tube neutron sources use a penning trap for their ion sources. But yeah, it was an incomplete and inaccurate wording.

How likely is a random mutation in a plant to be dangerous to the eater? E.g. a protein sequence changed such that a toxic molecule is produced. If subtle or slow acting that could pass a lot of expensive testing and make it to market. But is it risky enough to worry about?

Sure it could occur with natural mutations too but at a much lower rate.

I think the most likely scenario would be the plant expressing a new (or modified) allergenic protein, and it seems like there are some decent methods to look for those things already in use.


Also likely is that part of the genome that produces toxins in one part of the plant that isn't normally eaten gets turned on in edible part.

For example: asparagus berries are toxic, but the toxin doesn't appear in the stalk. The mechanism to produce the toxic chemical is there, just turned off in those cells. Enabling it is the sort of thing that small mutations might do.

So The Simpsons episode with “tomacco” wasn’t entirely unrealistic!

Tomatoes and tobacco are actually reasonably closely related already: same Family, Solanaceae. (I wouldn't be surprised if that fact inspired the episode.) They might even be cross-breedable as is.

I kinda remember reading an article about someone in England that was doing this with tomatoes and potatoes. They do it with a graft though instead of genetic modification.


Here are more pictures of such a garden:


My understanding is plants are fairly radiation resistant compared to animals... A small fault in the leaf of a plant just kills that leaf, but a small fault in your heart kills you...

If that's the case, I'd be a little worried even being in a plane flying over that garden...

The source might be in a container that is shielded in such a way that it only allows radiation to escape sideways which will then eventually end up absorbed in the dike wall.

Even today, within what could be called the “fancy houseplant” community of garden enthusiasts, we hear stories about Indonesian farmers purposely and repeatedly putting houseplants through airport X-Ray machines to try to induce variegation in their offspring. There are also plants like the Pink Congo, where growers use ethylene gas and auxin hormones to temporarily create pink foliage: https://www.wired.com/story/pink-princess-plantfluencers-pin...

Where does the fancy houseplant community hangout on the internet? I’m intrigued

Instagram, Facebook, and YouTube, believe it or not!

Instagram has been a big driver of houseplant sales; people see other people showing off their plants and their “shelfies” (plant shelf selfies), in gorgeous glamour photos, and it drives demand of the plants and discussion in the comments. Instagram is also a direct sales channel for Indonesian, Thai, and Sri Lankan growers, who often don’t speak English well but know how to tag the photos of their crops in English, so you can search the hashtags and get daily photos of ready stock. And buyers and sellers then use Google Translate to complete the sales in Instagram messages. I usually reach out to multiple growers in Instagram messages to get their latest price lists, compare them to the latest photos in the feeds, and then do a little arbitrage.

Facebook groups are huge for houseplant geeks. And while FB may have a reputation for older folks, the people in these groups trend much younger, many in their twenties. There are groups divvied up by care tips per specific species, and “BST” (Buy, Sell, Trade) groups that are for commerce but not really discussion, and even regional groups that build up local communities. For example, the Bay Area has BART: Bay Area Rare Tropicals, but also Bay Area Hoya. People post what they’re looking for, and do sales and swaps and even arrange transportation for newly purchased plants (i.e. if you’re going from Livermore to Palo Alto mid-week, can you bring my philodendron). These groups tend to have large and active administrator teams, and tracking reputation within the communities/groups helps cut way down on scammers. If you sell someone a crappy plant, or conversely if you waste a seller’s time and flake out on your purchase, word will get around.

Finally, there is some seriously great plant content on YouTube. Detailed how-to videos, propagation experiments, tours of rare plant nurseries and stores, and many people use that content to drive traffic to their own store or to promote their favorite stores. Rare plant unboxing videos are a whole genre! And they’re extra suspenseful because a decent number of wholesale deliveries from overseas do arrive dead.

Outside of those three, Etsy has become a major sales channel for rare houseplant sales. And just so we’re clear, rare houseplants can go for hundreds of dollars. Price inflation isn’t just for lumber, it’s philodendrons, too.

Thanks for the detailed reply! Can you drop some names of content creators you enjoy as a starting point?

Surely the best song (mostly) with this title, Bad Religion's "Atomic Garden": https://www.youtube.com/watch?v=AhzhiQA6-Aw

According to this article,[1] cultivars produced with this method include Calrose 76 rice and Todd's Mitcham peppermint, which respecively account for roughly half of California's rice crop and a majority of the world's peppermint

[1] http://www.ediblegeography.com/strange-and-beautiful-seeds-f...

This reminds me of an article [1] I read a long time ago on erowid about people using gout medication to turn cannabis seeds into polyploid 'superweed'.

[1] - https://erowid.org/plants/cannabis/cannabis_cultivation14.sh...

Another helpful technique for stimulating genetic changes is colchicine, which induces polyploidy.

If I take colchicine for gout is there an extra risk of polyploidy in my little swimmers?

I'm immediately curious what percentage of non-GMO certified produce are cultivars that were produced this way. This approach strikes me as being much more likely to accidentally produce unexpectedly harmful effects than modern gene recombination.

Why do you think it would be more prone to do that compared with newer approaches?

With gene recombination, you're inserting specific, known genes into the organism's DNA. The possible range of consequences of that modification is at least reasonably predictable.

With atomic gardening, you're changing the genetic code at random, and I would guess that you really have no decent way of knowing exactly what's changed and what effects it might have. So you're only going to notice things that are really obvious, or that you specifically looked for. And you can't just do a dragnet search for every possible effect, because that would get you sent to the 4th circle of scientific hell. [1]

Personally I'm not worried about whether Rio Star grapefruit is safe to eat. (About as safe as any food can be, that is. This morning I stumbled across something about french press coffee being bad for cholesterol levels, and can't help but think that, if you look hard enough, you can probably find some way in which every food is killing you. Which isn't to say that nothing matters, but one of the things that does matter is effect size. But I digress.) It's been around for a long time and people have been eating it with no apparent ill effects. But I'm pretty sure I'm not using the same risk calculus as folks who belong to the non-GMO certification program's target market.

[1]: https://journals.sagepub.com/doi/10.1177/1745691612459519

Yeah, you're basically scrambling things across an entire genome with no way to account for all of the random and unspecific changes that may occur.

One thing to consider though is that our bodies have evolved to eat all sorts of DNA - fragments and full genes - without directly incorporating it into our own genomes.

I'd personally be more concerned about plants expressing new allergenic proteins or that sort of thing, but it seems like there are already some decent methods in use to detect that.

The Simpsons Tomacco episode comes to mind: “ugh, this is terrible... I want more!!”

Should I try this in my back yard?

Yeah, where do I go to get my tomato seeds irradiated? Is there something on Amazon I can buy for that?

Thinking about this actually makes me sick to the stomache.

Same thing happens naturally all the time.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact