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

AlphaFold: a solution to a 50-year-old grand challenge in biology

deepmind.com

Post reply on HN
Print view View original

371–380 of 683 posts

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#371

Post by pmastela »

ramraj07 wrote: post #236
mncharity wrote: post #64

Additional commentary in Science: https://www.sciencemag.org/news/2020/11/game-has-changed-ai-... (submitted by furcyd : https://news.ycombinator.com/item?id=25254888 ).

The most amazing part: > The organizers even worried DeepMind may have been cheating somehow. So Lupas set a special challenge: a membrane protein from a species of archaea, an ancient group of microbes. For 10 years, his research team tried every trick in the book to get an x-ray crystal structure of the protein. “We couldn’t solve it.” > But AlphaFold had no trouble. It returned a detailed image of a three-part pro…

Like the old Arthur C. Clark quote goes: “Any sufficiently advanced technology is indistinguishable from magic” -- unless it might be cheating in which case throw them a curve ball.

Kudos to the DeepMind team for making magic happen.

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#372

Post by TomJansen »

verroq wrote: post #74

So who will have access to this? DeepMind never publishes their models.

According to [1], they must release enough information for others to replicate the AI model: "As a condition of entering CASP, DeepMind—like all groups—agreed to reveal sufficient details about its method for other groups to re-create it. That will be a boon for experimentalists, who will be able to use accurate structure predictions to make sense of opaque x-ray and cryo-EM data."

[1]: https://www.sciencemag.org/news/2020/11/game-has-changed-ai-...

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#374

Post by TomJansen »

lacksconfidence wrote: post #315
curiousllama wrote: post #311

Earlier quoted context omitted.

Actually, no! Or at least, the budget they used ( In other words, it's less like GPT3 and more like ImageNet.

Is the cost to train really the relevant metric for developing this? It seems like the salary's involved are probably at least 10x whatever they spent on hardware.

>Is the cost to train really the relevant metric for developing this?

Yes, because they must release sufficient information for others to recreate the AI model, according to the rules of entering CASP.

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#375

Post by the8472 »

layer8 wrote: post #361
WhompingWindows wrote: post #181

Earlier quoted context omitted.

This reminds me of AlphaGo and AlphaZero. DeepMind was able to produce a very solid model on their first attempt, at both protein folding and at Go (and Starcraft2 as well). Their second models, however, seemed to blow their first out of the water. This bodes extremely well for the future of computational biology, I'm very excited thinking about the prospects. If we know how a protein folds, we know its shape, meanin…

One difference to AlphaZero though, if my understanding is correct, is that AlphaFold is trained on a predetermined data set and hence didn’t learn how “arbitrary” proteins fold in general, but just how the kinds of proteins fold for which we already know how they fold. To work more like AlphaZero, AlphaFold would have to be able to synthesize arbitrary proteins and run the experiments on them to verify and correct i…

> AlphaFold would have to be able to synthesize arbitrary proteins and run the experiments on them to verify and correct its predictions.

Could this lead to a virtuous cycle where AlphaFold is used generate a ton of random sequences where it has low confidence, those are then screened for ease of synthesis, measured and the results used to improve the model?

Edit: nevermind, according to another comment[0] there are still plenty of real proteins without experimental data left to explore.

[0] https://news.ycombinator.com/item?id=25255601

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#376

Post by kolinko »

matsemann wrote: post #275
xral wrote: post #233

Earlier quoted context omitted.

AlQuraishi's tweet [0] about this: > CASP14 #s just came out and they’re astounding—DeepMind looks to have solved protein structure prediction. Median GDT_TS went from 68.5 (CASP13) to 92.4!!!! Cf. their 2nd best CASP13 struct scored 92.8 (out of 100). Median RMSD is 2.1Å. I think it's over https://predictioncenter.org/casp14/zscores_final.cgi [0]: https://twitter.com/MoAlQuraishi/status/1333383634649313280

What does that Å mean? Never seen our letter been used in a scientific context.

0.1nm - approximately a size of an atom - used in organic chemistry often.

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#377

Post by _RPL5_ »

harperlee wrote: post #12

Not knowing a lot about biotechnology, I read the article and it sounds great, but how big is this as a gamechanger? Can someone comment on how big are the implications of this in, let’s say, 5 years from now, on day to day life? Does this mean that biotech is going to explode? Or just that drugs will come to market faster, perhaps cheaper for rare diseases, but from the same industry structure as always?

The industry process will not change. You still need industrial biologists to generate and validate AphaFold structures, interpret the results as part of the bigger picture, and to finally design the drugs. And, then, of course you still need to validate the drugs in experimental systems (first the test tube, then mice, then humans).

So your second guess is correct - one of the steps is much cheaper now, which marginally improves the entire pipeline. As a result, drugs should now arrive to the market faster.

As a side note, I am curious what happens to the field of structural biology in 10 to 15 years from now. Every research university has a large structural biology department with super expensive Xray/NRM/Cryo-EM machines, and armies of students who routinely spend 4-6 years of their PhD trying to solve a structure of a single protein. If AlphaFold works as advertised, NIH will gradually shift funding to other problems.

(It was predicted that it'd be taxi drivers, not professors, that AI got first. Ironic.)

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#378

Post by smt1 »

dekhn wrote: post #343
mjn wrote: post #341

Earlier quoted context omitted.

That's a good point; this system certainly didn't come from nowhere! The protein datasets they used also mostly came out of various NIH-funded projects. What I meant to focus on was that I think DeepMind has less of a pure money/scale advantage in this area than in some others. In something like Go or Atari game-playing, there are many academic groups researching similar things, but their resources are laughably smal…

Personally I think a major part of the secret sauce is Google's internal compute infrastructure. When I was an academic, 50% of my time went to building infra to do my science. At Google, petabytes of storage, millions of cores, algorithms, and brains were all easily tappable within a common software repo and cluster infrastructure. That immediately translates to higher scientific productivity.

I agree. What's doubly interesting is google's internal transparency, and open source first policy. I think it's probable that that effect spreads and creates fly wheel effects for life, natural sciences, and behavioral sciences. Keep in mind that they've also absorbed effectively the R&D side of Bell Labs from a computer science /distributed computing point of view, gopher is pretty much that, and also in effect interesting from a sociological p.o.v, "this is the shifting the resources of the polyad network problem", or problems caused by rapid commercializations of the World Wide Web rather than physics like was originally ideated @ CERN) and moving to effective effort in other fields, even if it doesn't happen @ Alphabet. Hell, they could be dismantled (given the FTC complainants), and probably the resultant companies would rebuild like paperclips sort of like Pa'Bell did post-1984.

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#379

Post by gmorainbows »

flobosg wrote: post #301
gmorainbows wrote: post #288

Earlier quoted context omitted.

The method relies on multiple-sequence-alignment (MSA) of homologous proteins. This cannot fold arbitrary proteins, only biologically relevant ones that have high quality MSAs available. It's also worth pointing out that the gold-standard for validating MSAs relies on PDBs of folded proteins. This is exciting work that will assist NMR and XRay crystallographers, but it's not a panacea of protein folding. https://gith…

In their CASP abstract[1] they mention alternatives to typical co-evolution features which improve performance in shallow MSA depths. [1]: https://predictioncenter.org/casp14/doc/CASP14_Abstracts.pdf...

It doesn't matter so much how they perform the feature extraction, so much as what their inputs to the feature extraction are.

This model requires a collection of wild-type proteins in an accurate MSA. Producing an accurate MSA is hard even if you have many homologs.

They require protein homologs which means they can "only" do this for wild-type proteins. This work is useless with mutant and synthetic proteins. This is a big advancement that will assist crystallographers and NMR structural biologists with difficult wild-type proteins, but it doesn't "solve protein folding" by any stretch of the imagination.

Re: AlphaFold: a solution to a 50-year-old grand challenge in biology

#380

Post by breatheoften »

FrojoS wrote: post #327
breatheoften wrote: post #314

Earlier quoted context omitted.

That's quite interesting ... I believe the test set size is not constant year to year but rather a function of how many new structures have been experimentally discovered since the last contest? Does seem like the contest structure could include quite a bit of risk for hiding the effect of overfitting ... I wonder if there is anything inherent about the problem that reduces that risk ...?

My understanding is, that it's always 100 new structures, which is a small fraction of the total structures identified in that year. The reason why the top score in one year, can be lower than in the previous year, is that the test (the 100 structures to guess) is always new and different, so it can end up being 'harder' than the year before. Luck will also play a small role. Another explanation for a reduction in th…

Only 100 new structures each test cycle? That seems a very small test set size ...

Is it really possible to select 100 new structures which together are likely to represent a meaningful increase in the sample generalization versus the prior years test set ...?

Post reply on HN