5 ms·

The last few paragraphs describe how they basically created unstoppable super-bacteria that are untargetable by any existing medication or immune response, what

The last few paragraphs describe how they basically created unstoppable super-bacteria that are untargetable by any existing medication or immune response, what could go wrong? Let’s hope their lab safety is uncharacteristically excellent.

andai 3y ago
Hmm... I've heard this one before!
msandford 3y ago
Yeah I hear this and I'm not thinking about the limited upside, I'm worrying about the nearly unlimited downside. It is limited though. They can't do worse than destroying all life as we know it. Not very comforting.
- theseatoms 3y ago
  Have they learned nothing from the past four years? Just that they can get away with it?
- pgeorgi 3y ago
  > They can't do worse than destroying all life as we know it. Won't hear anybody complaining, so...
  - pixl97 3y ago
    PM to AI: "We need to find a way to reduce customer complaints" AI: "Solution found"
robblbobbl 3y ago
They should act very carefully, just like the researchers working in quantum physics and machine learning/artificial intelligence field. There is too much mayhem you can obviously cause with this new kind of technology possibilities/views. And while breaking current limitation boundaries, they should always keep in mind (remember) the 'greater good' for the average popular masses.
antonvs 3y ago
Wait until they figure out super-prions. "Oh you got one in your mouth? It's been nice knowing you, in a few hours you'll be a blob of goo."
- throwawaymaths 3y ago
  That's really not how prions work.
  - antonvs 3y ago
    Not yet.
    - throwawaymaths 3y ago
      No. Prions are not "infectious proteins", they are "transmissible protein folds". The universe of proteins which can be spontaneously misfolded by another protein that is already misfolded is relatively rare. It's also not very difficult to engineer. When I was in grad school some of my classmates tried (not all amyloid "prion" proteins are bad, some are structural, such as pMel which is used to sequester toxic melanin precursors and scaffold them into melanin), and it was difficult to improve upon nature. It's very likely that every "prion-capable" protein is that way for a strategic reason (e.g. Alzheimer's protein is likely that way to ensnare viruses before they can further damage brains). https://www.sciencedirect.com/science/article/pii/S0021925817506031 https://www.sciencedirect.com/science/article/pii/S002192581... Not every protein needs to be capable of misfolding, life would be very metastable if every protein were, so, no, you will not turn into a "puddle of goo" via an engineered prion. It might really suck, and your brain (or pick organ here) would get holes drilled in it over the course of years. But not hours.
      - antonvs 3y ago
        It may be difficult for humans to improve on nature. It may be easier for AIs. Keep in mind that the space explored by evolution is necessarily quite restricted, and very damaging entities would have been selected out if at all possible. What you can imagine based on the evolved structures you've seen may be more limited than you realize. In any case, my previous comments were intended jokingly, they weren't submissions to Methods in Molecular Biology.
candiodari 3y ago
This is already done. Essentially making a medicine goes like this: 1) is it functional? Yes -> step 2 2) is it cytotoxic? (kills everything, for example by dissolving cell membranes) No -> goto step 3 3) is it neurotoxic? (prevents neurons from working) No -> goto step 4 4 to 199) is it ...toxic? 200) proceed to in vitro testing Someone made a paper ... what happens if you just reverse steps 2 through 199? Start with any old protein, and keep changing it to display as many different kinds of toxicity as possible. Answer: This just works. Works very well, since we know computational tests for a whole lot of toxicities, and at least a few of them are differentiable. So you can just assume essentially every country has already done this and tested it. Fortunately this is not such a huge issue. The challenge is not making poisons, the challenge is spreading them effectively (without ...). Some of these proteins would at least need contact between a large drop of the protein and your skin. Most would need to be injected to kill you. So ... And we know plenty of non-protein molecules, much easier and faster to make, available in industrial quantities without much checking, that are about as bad as the worst of these proteins. Also: there are radioactive substances sold on Amazon that'll kill you slowly and painfully just being near you. https://unicri.it/sites/default/files/2021-12/21_dual_use.pdf https://unicri.it/sites/default/files/2021-12/21_dual_use.pd...
michaelbrave 3y ago
I want to say I read a few years back some scientists were studying the possibility of using phages to fight super bacteria, apparently can evolve to be either phage resistant or antibiotic resistant but never both, or at least that was the gist of it I can remember.
z500 3y ago
Did I miss something? There's something in there about using a modified amino acid to enhance the immune response against tumors, and something about bacteriophage resistance, but the last few paragraphs are about synthesizing polymers.

andai 3y ago

Hmm... I've heard this one before!

msandford 3y ago

Yeah I hear this and I'm not thinking about the limited upside, I'm worrying about the nearly unlimited downside. It is limited though. They can't do worse than destroying all life as we know it. Not very comforting.

theseatoms 3y ago
Have they learned nothing from the past four years? Just that they can get away with it?
pgeorgi 3y ago
> They can't do worse than destroying all life as we know it. Won't hear anybody complaining, so...
- pixl97 3y ago
  PM to AI: "We need to find a way to reduce customer complaints" AI: "Solution found"

robblbobbl 3y ago

They should act very carefully, just like the researchers working in quantum physics and machine learning/artificial intelligence field. There is too much mayhem you can obviously cause with this new kind of technology possibilities/views. And while breaking current limitation boundaries, they should always keep in mind (remember) the 'greater good' for the average popular masses.

antonvs 3y ago

Wait until they figure out super-prions. "Oh you got one in your mouth? It's been nice knowing you, in a few hours you'll be a blob of goo."

throwawaymaths 3y ago
That's really not how prions work.
- antonvs 3y ago
  Not yet.
  - throwawaymaths 3y ago
    No. Prions are not "infectious proteins", they are "transmissible protein folds". The universe of proteins which can be spontaneously misfolded by another protein that is already misfolded is relatively rare. It's also not very difficult to engineer. When I was in grad school some of my classmates tried (not all amyloid "prion" proteins are bad, some are structural, such as pMel which is used to sequester toxic melanin precursors and scaffold them into melanin), and it was difficult to improve upon nature. It's very likely that every "prion-capable" protein is that way for a strategic reason (e.g. Alzheimer's protein is likely that way to ensnare viruses before they can further damage brains). https://www.sciencedirect.com/science/article/pii/S0021925817506031 https://www.sciencedirect.com/science/article/pii/S002192581... Not every protein needs to be capable of misfolding, life would be very metastable if every protein were, so, no, you will not turn into a "puddle of goo" via an engineered prion. It might really suck, and your brain (or pick organ here) would get holes drilled in it over the course of years. But not hours.
    - antonvs 3y ago
      It may be difficult for humans to improve on nature. It may be easier for AIs. Keep in mind that the space explored by evolution is necessarily quite restricted, and very damaging entities would have been selected out if at all possible. What you can imagine based on the evolved structures you've seen may be more limited than you realize. In any case, my previous comments were intended jokingly, they weren't submissions to Methods in Molecular Biology.

candiodari 3y ago

This is already done. Essentially making a medicine goes like this: 1) is it functional? Yes -> step 2 2) is it cytotoxic? (kills everything, for example by dissolving cell membranes) No -> goto step 3 3) is it neurotoxic? (prevents neurons from working) No -> goto step 4 4 to 199) is it ...toxic? 200) proceed to in vitro testing Someone made a paper ... what happens if you just reverse steps 2 through 199? Start with any old protein, and keep changing it to display as many different kinds of toxicity as possible. Answer: This just works. Works very well, since we know computational tests for a whole lot of toxicities, and at least a few of them are differentiable. So you can just assume essentially every country has already done this and tested it. Fortunately this is not such a huge issue. The challenge is not making poisons, the challenge is spreading them effectively (without ...). Some of these proteins would at least need contact between a large drop of the protein and your skin. Most would need to be injected to kill you. So ... And we know plenty of non-protein molecules, much easier and faster to make, available in industrial quantities without much checking, that are about as bad as the worst of these proteins. Also: there are radioactive substances sold on Amazon that'll kill you slowly and painfully just being near you. https://unicri.it/sites/default/files/2021-12/21_dual_use.pdf https://unicri.it/sites/default/files/2021-12/21_dual_use.pd...

michaelbrave 3y ago

I want to say I read a few years back some scientists were studying the possibility of using phages to fight super bacteria, apparently can evolve to be either phage resistant or antibiotic resistant but never both, or at least that was the gist of it I can remember.

z500 3y ago

Did I miss something? There's something in there about using a modified amino acid to enhance the immune response against tumors, and something about bacteriophage resistance, but the last few paragraphs are about synthesizing polymers.