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

AlphaFold 3 predicts the structure and interactions of life's molecules

blog.google

Post reply on HN
Print view View original

61–70 of 504 posts

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#61

Post by adw »

moconnor wrote: post #32

Stepping back, the high-order bit here is an ML method is beating physically-based methods for accurately predicting the world. What happens when the best methods for computational fluid dynamics, molecular dynamics, nuclear physics are all uninterpretable ML models? Does this decouple progress from our current understanding of the scientific process - moving to better and better models of the world without human-int…

> What happens when the best methods for computational fluid dynamics, molecular dynamics, nuclear physics are all uninterpretable ML models?

A better analogy is "weather forecasting".

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#62

Post by nybsjytm »

Important caveat: it's only about 70% accurate. Why doesn't the press release say this explicitly? It seems intentionally misleading to only report accuracy relative to existing methods, which apparently are just not so good (30%, 50% in various settings). https://www.fastcompany.com/91120456/deepmind-alphafold-3-dn...

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#63

Post by hahaurfunny »

Jerrrry wrote: post #45
thomasahle wrote: post #41

Earlier quoted context omitted.

> Stepping back, the high-order bit here is an ML method is beating physically-based methods for accurately predicting the world. I mean, it's just faster, no? I don't think anyone is claiming it's a more _accurate_ model of the universe.

Collision libraries and fluid libraries have had baked-in memorized look-up tables that were generated with ML methods nearly a decade ago. World is still here, although the Matrix/metaverse is becoming more attractive daily.

[dead]

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#64

Post by jeffreyrogers »

moconnor wrote: post #32

Stepping back, the high-order bit here is an ML method is beating physically-based methods for accurately predicting the world. What happens when the best methods for computational fluid dynamics, molecular dynamics, nuclear physics are all uninterpretable ML models? Does this decouple progress from our current understanding of the scientific process - moving to better and better models of the world without human-int…

I asked a friend of mine who is chemistry professor at a large research university something along these lines a while ago. He said that so far these models don't work well in regions where either theory or data is scarce, which is where most progress happens. So he felt that until they can start making progress in those areas it won't change things much.

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#65

Post by bbor »

moconnor wrote: post #32

Stepping back, the high-order bit here is an ML method is beating physically-based methods for accurately predicting the world. What happens when the best methods for computational fluid dynamics, molecular dynamics, nuclear physics are all uninterpretable ML models? Does this decouple progress from our current understanding of the scientific process - moving to better and better models of the world without human-int…

This is exactly how the physicists felt at the dawn of quantum physics - the loss of meaningful human inquiry to blindly effective statistics. Sobering stuff…

Personally, I’m convinced that human reason is less pure than we think it to be, and that the move to large mathematical models might just be formalizing a lack-of-control that was always there. But that’s less of a philosophy of science discussion and more of a cognitive science one

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#66

Post by dekhn »

JackFr wrote: post #31
dekhn wrote: post #23

Earlier quoted context omitted.

AlphaFold very explicitly (unless something has changed) removes NMR structures as references because they are not accurate enough. I have a PhD in NMR biomolecular structure and I wouldn't trust. the structures for anything.

Sorry, I don’t mean to be dense - do you mean you don’t trust AlphaFolds structures or NMRs?

I don't trust NMR structures in nearly all cases. The reasons are complex enough that I don't think it's worthwhile to discuss on Hacker News.

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#67

Post by 11101010001100 »

dekhn wrote: post #18
dopylitty wrote: post #13

This reminds me of Google’s claim that another “AI” discovered millions of new materials. The results turned out to be a lot of useless noise but that was only apparent after actual expert spent hundreds of hours reviewed the results[0] 0: https://www.404media.co/google-says-it-discovered-millions-o...

The alphafold work has been used across the industry (successfully, in the sense of blind prediction), and has been replicated independently. The work on alphafold will likely net Demis and John a Nobel prize in the next few years. (that said, one should always inspect Google publications with a fine-toothed comb and lots of skepticism, as they have a tendency to juice the results)

Depending on your expected value of quantum computing, the Nobel committee shouldn't wait too long.

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#68

Post by krzat »

moconnor wrote: post #32

Stepping back, the high-order bit here is an ML method is beating physically-based methods for accurately predicting the world. What happens when the best methods for computational fluid dynamics, molecular dynamics, nuclear physics are all uninterpretable ML models? Does this decouple progress from our current understanding of the scientific process - moving to better and better models of the world without human-int…

We will get better with understanding black boxes, if a model can be compressed into simple math formula then it's both easier to understand and to compute.

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#69

Post by bmau5 »

qwertox wrote: post #44

> Thrilled to announce AlphaFold 3 which can predict the structures and interactions of nearly all of life’s molecules with state-of-the-art accuracy including proteins, DNA and RNA. [1] There's a slight mismatch between the blog's title and Demis Hassabis' tweet, where he uses "nearly all". The blog's title suggests that it's a 100% solved problem. [1] https://twitter.com/demishassabis/status/1788229162563420560

Marketing vs. Reality :)

Re: AlphaFold 3 predicts the structure and interactions of life's molecules

#70

Post by fabian2k »

dekhn wrote: post #59
fabian2k wrote: post #56

Earlier quoted context omitted.

I think it is very plausible that they don't use NMR structures here, but I was looking for a specific statement on it in the paper. I think your guess is plausible, but I don't think the paper is clear enough here to be sure about this interpretation.

Yes, thanks for calling that out. In verifying my statement I actually was confused because you can see they filter NMR out of the eval set (saying so explicitly) but don't say that in the test set section (IMHO they should be required to publish the actual selection script so we can inspect the results).

Hmm, in the earlier AlphaFold 2 paper they state:

> Input mmCIFs are restricted to have resolution less than 9 Å. This is not a very restrictive filter and only removes around 0.2% of structures

NMR structures are more than 0.2% so that doesn't fit to the assumption that they implicitly remove NMR structures here. But if I filter by resolution on the PDB homepage it does remove essentially all NMR structures. I'm really not sure what to think here, the description seems too soft to know what they did exactly.

Post reply on HN