Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Generative A.I. arrives in the gene editing world of CRISPR

nytimes.com

Post reply on HN
Print view View original

41–50 of 69 posts

Re: Generative A.I. arrives in the gene editing world of CRISPR

#41

Post by JangoSteve »

m3kw9 wrote: post #30

[flagged]

Do you mean for AI to do the entire job of researching, creating, testing, manufacturing, and distributing a cure, or just for AI to be involved? And do you mean completely eradicating a disease, or just producing a cure for it? And do you mean an outright cure, or also a treatment or vaccine? If the latter in all cases, here's an example:

https://www.technologyreview.com/2022/08/26/1058743/i-was-th...

If you mean the former in any case, it'll probably be a while, if ever in our lifetimes.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#42

Post by nyokodo »

scarmig wrote: post #25
r2_pilot wrote: post #24

Earlier quoted context omitted.

Computer viruses run on hardware they do not create but merely hijack. Unless a virus took control of a semiconductor fab, it's hard to argue that they are alive/reproducing in the context of this discussion.

Parasites do the same thing: hijack a piece of hardware and use it to reproduce. Computer viruses have even formed a kind of symbiotic ecosystem with blackhats: the scammer provides resources to help the virus reproduce, and the virus provides access to the scammer in turn. In a way, all life hijacks hardware (the material world) that it doesn't create to reproduce itself.

> Parasites do the same thing

Only analogously. The reality is that what we call computer viruses are merely instructions running on a computer and are not substantially distinct from the computer in the same way that parasites or physical viruses are distinct from biological tissue.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#43

Post by pharmaz0ne »

swamp40 wrote: post #40
paxys wrote: post #23

Earlier quoted context omitted.

You mean exactly what 23andMe tried to do, and failed miserably at.

We are still early. Eventually you'll be able to change your race, gender, add reptile eyes, regrow limbs etc. Has to start somewhere. Need more data.

Exactly. We should start building global database connecting DNA with medical history.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#46

Post by ben_w »

mewpmewp2 wrote: post #31
scarmig wrote: post #25

Earlier quoted context omitted.

Parasites do the same thing: hijack a piece of hardware and use it to reproduce. Computer viruses have even formed a kind of symbiotic ecosystem with blackhats: the scammer provides resources to help the virus reproduce, and the virus provides access to the scammer in turn. In a way, all life hijacks hardware (the material world) that it doesn't create to reproduce itself.

Computer viruses need to get an ability to mutate by themselves, improving the code overtime though.

Simulated evolution is trivial to implement, but my guess is also a bit pointless from the point of view of most people writing viruses — the viruses might mutate to not give them money.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#47

Post by devindotcom »

mxwsn wrote: post #17
miraculixx wrote: post #11

Reading their blog post I wonder if an LLMs is really the best way to do this. If I got it right, they used the LLM to enumerate potential protein DNA sequences. Does that really need an LLM? Enumeration is not novel, nor are LLMs particularily good at it. If you want to computationally parallelize the search in a large enumeration space it would be much easier to simply, well, do that instead of taking a detour via…

It's not an English LLM, but a "protein" language model, where tokens represent amino acids or nucleotides. Learning a transformer language model on such data simply learns a distribution over sequences of tokens. It's a fine approach conceptually that in many ways is the "right" way or most elegant method, and not a stretch at all.

I enjoyed the feeling when I made this connection talking with a startup doing this a while back. It's just a different "language" and although it's not a given that LLMs can operate in it, it's a reasonable thing to try, and it turns out they can.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#48

Post by ben_w »

r2_pilot wrote: post #24
jprete wrote: post #16

Earlier quoted context omitted.

What about computer viruses?

Computer viruses run on hardware they do not create but merely hijack. Unless a virus took control of a semiconductor fab, it's hard to argue that they are alive/reproducing in the context of this discussion.

> Unless a virus took control of a semiconductor fab, it's hard to argue that they are alive/reproducing in the context of this discussion.

It does not seem implausible at this point to imagine a virus which gets to control some currency, uses it place an order for parts to be assembled, delivered to a location, connected to power, etc.

It is, in a sense, taking control by pulling the levers supplied by our society. Is that alive?

I would say no, it is not 'alive'… but I would also paraphrase Dijkstra: "The question of whether an AI or computer virus is 'alive' is no more interesting than the question of whether a submarine can swim."

Re: Generative A.I. arrives in the gene editing world of CRISPR

#49

Post by scarmig »

ben_w wrote: post #46
mewpmewp2 wrote: post #31

Earlier quoted context omitted.

Computer viruses need to get an ability to mutate by themselves, improving the code overtime though.

Simulated evolution is trivial to implement, but my guess is also a bit pointless from the point of view of most people writing viruses — the viruses might mutate to not give them money.

Those mutations would probably be favored by evolution, if anything. They would attract less attention and allow more focus on reproduction.

Re: Generative A.I. arrives in the gene editing world of CRISPR

#50

Post by swamp40 »

pharmaz0ne wrote: post #43
swamp40 wrote: post #40

Earlier quoted context omitted.

We are still early. Eventually you'll be able to change your race, gender, add reptile eyes, regrow limbs etc. Has to start somewhere. Need more data.

Exactly. We should start building global database connecting DNA with medical history.

Not a big fan? It could be provably private. You could have a kit with a random username/password. It could be done. People just have a bad taste from 23andMe.
Post reply on HN