4 ms·

His fear is about a self-spreading superplant, which is what's supposed to make GMO a special type of danger. Without self-spreading the worst you can get is a

His fear is about a self-spreading superplant, which is what's supposed to make GMO a special type of danger. Without self-spreading the worst you can get is a failed product. If it was that easy to make a superplant, evolution would have already figured it out. The point of GMO is to engineer organisms for artificial environment under human supervision, and humans aren't going to protect harmful plants. Eg. nitrogen fixing may sound "obvious" but it takes energy, and empirically in most environments such plants were outcompeted by those that didn't spend energy this way. It's only net cheaper for humans, because the cost of manufacturing, transporting etc the fertilizer has to be included. It's the same story for everything. It's going to take a really, really long time before human genetic engineering can hope to beat evolution on its own turf. By a rough estimate, at any second on Earth there's close to 2^105 ribosomes manufacturing proteins from RNA, only once simulation capabilities start to come close genetically engineered organisms start to become a systemic risk. Funny story that confirms this view: https://www.nature.com/articles/s41598-019-49660-6 https://www.nature.com/articles/s41598-019-49660-6 >However, it is clear from the data in Garziera et al.6 that the effectiveness of the release program began to break down after about 18 months, i.e., the population which had been greatly suppressed rebounded to nearly pre-release levels. Ultimately, it was just a fun, harmless experiment, and as a bonus anyone arrogant enough to expect current human genetic engineering to beat evolution got a lesson in humility.

phkahler 7y ago
>> If it was that easy to make a superplant, evolution would have already figured it out. Dont be so sure. It seems like nitrogen fixing plants could get out of control, and evolution hasn't produced them yet. Sexual reproduction allows improvements from diverging lineages to combine in a single offspring. But only if they haven't diverged too far. Genetic engineering may bring beneficial feature from vary different species, or in this case kingdoms. If a nitrogen fixing plant took over the world an obvious way to target it would be to aim at the nitrogen pathway, but then we might kill most of the bacteria our food supply depends on. Having said that I think this work is a good idea, but if I can conceive of a way for it to go wrong I'm sure nature could come up with many more.
- DuskStar 7y ago
  > Sexual reproduction allows improvements from diverging lineages to combine in a single offspring. But only if they haven't diverged too far. Genetic engineering may bring beneficial feature from vary different species, or in this case kingdoms. As stated elsewhere, gene transfer also happens cross-kingdom by way of virus.
- riffraff 7y ago
  > It seems like nitrogen fixing plants could get out of control, and evolution hasn't produced them yet. what about legumes? What's the difference between "built-in" nitrogen fixing and symbiotic nitrogen fixing that would make the former much more dangerous?
abraae 7y ago
> If it was that easy to make a superplant, evolution would have already figured it out. Is that the case when plants are now living in an unnatural environment,e.g with co2 at elevated levels compared to the levels that today's plants evolved in? The game today is entirely new, and who's to say what sort of things will thrive in these new conditions.
josinalvo 7y ago
> Ultimately, it was just a fun, harmless experiment, and as a bonus anyone arrogant enough to expect current human genetic engineering to beat evolution got a lesson in humility. Why is is any more "trying to beat evolution" than, say, using antibiotics? In a 18 month timespan, there is not much useful mutation (afaik!). The problem was that there was already a variant that could survive the attack. How can we quantify the risk a given attack will fail/calculate how many such interventions would be needed?

phkahler 7y ago

>> If it was that easy to make a superplant, evolution would have already figured it out. Dont be so sure. It seems like nitrogen fixing plants could get out of control, and evolution hasn't produced them yet. Sexual reproduction allows improvements from diverging lineages to combine in a single offspring. But only if they haven't diverged too far. Genetic engineering may bring beneficial feature from vary different species, or in this case kingdoms. If a nitrogen fixing plant took over the world an obvious way to target it would be to aim at the nitrogen pathway, but then we might kill most of the bacteria our food supply depends on. Having said that I think this work is a good idea, but if I can conceive of a way for it to go wrong I'm sure nature could come up with many more.

DuskStar 7y ago
> Sexual reproduction allows improvements from diverging lineages to combine in a single offspring. But only if they haven't diverged too far. Genetic engineering may bring beneficial feature from vary different species, or in this case kingdoms. As stated elsewhere, gene transfer also happens cross-kingdom by way of virus.
riffraff 7y ago
> It seems like nitrogen fixing plants could get out of control, and evolution hasn't produced them yet. what about legumes? What's the difference between "built-in" nitrogen fixing and symbiotic nitrogen fixing that would make the former much more dangerous?

abraae 7y ago

> If it was that easy to make a superplant, evolution would have already figured it out. Is that the case when plants are now living in an unnatural environment,e.g with co2 at elevated levels compared to the levels that today's plants evolved in? The game today is entirely new, and who's to say what sort of things will thrive in these new conditions.

josinalvo 7y ago

> Ultimately, it was just a fun, harmless experiment, and as a bonus anyone arrogant enough to expect current human genetic engineering to beat evolution got a lesson in humility. Why is is any more "trying to beat evolution" than, say, using antibiotics? In a 18 month timespan, there is not much useful mutation (afaik!). The problem was that there was already a variant that could survive the attack. How can we quantify the risk a given attack will fail/calculate how many such interventions would be needed?