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

31–40 of 504 posts

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

#31

Post by JackFr »

dekhn wrote: post #23
joshuamcginnis wrote: post #17

Earlier quoted context omitted.

Accuracy can be assessed two main ways: computationally and experimentally. Computationally, they would compare the predicted structures and interactions with known data from databases like PDB (Protein Database). Experimentally, they can use tools like x-ray crystallography and NMR (nuclear magnetic resonance) to obtain the actual molecule structure and compare it to the predicted result. The outcomes of each approa…

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?

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

#32

Post by moconnor »

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-interpretable theories and mathematical models / explanations? Is that even iteratively sustainable in the way that scientific progress has proven to be?

Interesting times ahead.

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

#33

Post by candiodari »

s1artibartfast wrote: post #2

The article was heavy on the free research aspect, but light on the commercial application. I'm curious about the business strategy. Does Google intend to license out tools, partner, or consult for commercial partners?

I wonder what the license for RoseTTAFold is. On github you have:

https://github.com/RosettaCommons/RoseTTAFold/blob/main/LICE...

But there's also:

https://files.ipd.uw.edu/pub/RoseTTAFold/Rosetta-DL_LICENSE....

Which is it?

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

#34

Post by tepal »

Metacelsus wrote: post #4

From: https://www.nature.com/articles/d41586-024-01383-z >Unlike RoseTTAFold and AlphaFold2, scientists will not be able to run their own version of AlphaFold3, nor will the code underlying AlphaFold3 or other information obtained after training the model be made public. Instead, researchers will have access to an ‘AlphaFold3 server’, on which they can input their protein sequence of choice, alongside a selection of…

Or OpenFold, which is the more literal reproduction of AlphaFold 2: https://github.com/aqlaboratory/openfold

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

#35

Post by fabian2k »

dekhn wrote: post #23
joshuamcginnis wrote: post #17

Earlier quoted context omitted.

Accuracy can be assessed two main ways: computationally and experimentally. Computationally, they would compare the predicted structures and interactions with known data from databases like PDB (Protein Database). Experimentally, they can use tools like x-ray crystallography and NMR (nuclear magnetic resonance) to obtain the actual molecule structure and compare it to the predicted result. The outcomes of each approa…

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.

Looking at the supplementary material (section 2.5.4) for the AlphaFold 3 paper it reads to me like they still use NMR structures for training, but not for evaluating performance of the model.

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

#36

Post by dekhn »

ranger_danger wrote: post #26
Metacelsus wrote: post #4

From: https://www.nature.com/articles/d41586-024-01383-z >Unlike RoseTTAFold and AlphaFold2, scientists will not be able to run their own version of AlphaFold3, nor will the code underlying AlphaFold3 or other information obtained after training the model be made public. Instead, researchers will have access to an ‘AlphaFold3 server’, on which they can input their protein sequence of choice, alongside a selection of…

Not just unfortunate, but doesn't this make it completely untrustable? How can you be sure the data was not modified in any way? How can you verify any results?

You determine a crystal structure of a known protein which does not previously have a known structure, and compare the prediction to the experimentally determined structure.

There is a biennial (biannual?) competition known as CASP where some new structures, not yet published, are used for testing predictions from a wide range of protein structure prediction (so, basically blind predictions which are then compared when the competition wraps up). AlphaFold beat all the competitors by a very wide margin (much larger than the regular rate of improvement in the competition), and within a couple years, the leading academic groups adopted the same techniques and caught up.

It was one of the most important and satisfying moments in structure prediction in the past two+ decades. The community was a bit skeptical but as it's been repeatedly tested, validated, and reproduced, people are generally of the opinion that DeepMind "solved" protein structure prediction (with some notable exceptions), and did so without having the solve the full "protein folding problem" (which is actually great news while also being somewhat depressing).

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

#37

Post by dekhn »

COGlory wrote: post #29
renonce wrote: post #5

> What is different about the new AlphaFold3 model compared to AlphaFold2? > AlphaFold3 can predict many biomolecules in addition to proteins. AlphaFold2 predicts structures of proteins and protein-protein complexes. AlphaFold3 can generate predictions containing proteins, DNA, RNA, ions,ligands, and chemical modifications. The new model also improves the protein complex modelling accuracy. Please refer to our paper…

>So the criticism towards AlphaFold 2 will likely still apply? For example, it’s more accurate for predicting structures similar to existing ones, and fails at novel patterns? Yes, and there is simply no way to bridge that gap with this technique. We can make it better and better at pattern matching, but it is not going to predict novel folds.

alphafold has been shown to accurately predict some novel folds. The technique doesn't entirely depend on whole-domain homology.

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

#38

Post by cgearhart »

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 a neat observation. Slightly terrifying, but still interesting. Seems like there will also be cases where we discover new theories through the uninterpretable models—much easier and faster to experiment endlessly with a computer.

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

#39

Post by rolph »

renonce wrote: post #5

> What is different about the new AlphaFold3 model compared to AlphaFold2? > AlphaFold3 can predict many biomolecules in addition to proteins. AlphaFold2 predicts structures of proteins and protein-protein complexes. AlphaFold3 can generate predictions containing proteins, DNA, RNA, ions,ligands, and chemical modifications. The new model also improves the protein complex modelling accuracy. Please refer to our paper…

problem is biomolecules, are "chaperoned" to fold properly, only specific regions such as, alpha helix, or beta pleatedsheet will fold de novo.

Chaperone (protein)

https://en.wikipedia.org/wiki/Chaperone_(protein)

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

#40

Post by fnikacevic »

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 can only hope the models will be sophisticated enough and willing to explain their reasoning to us.
Post reply on HN