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

321–330 of 683 posts

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

#321

Post by kgwgk »

deeviant wrote: post #304
kgwgk wrote: post #139

Earlier quoted context omitted.

I agree. If a newspaper published a headline "Dr. Whatever cured cancer (... in some of her patients)" we would find it misleading.

If there was a headline, "Company X with Product Y cured cancer" and it turns out that product Y actually only cured 90% of cancer, I'm pretty sure most people would be happy the headline. Oh, and to be a true parallel example, in this case the remaining 10% of cancers might not even be cancers, as experimental accuracy of protein structures is only ~90% accurate, the model could very well be more accurate than our c…

I really interpreted that headline as "found a general solution to the protein-folding question" not as the also interesting but not that much "can be used to solve protein-folding problems".

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

#322

Post by purpleidea »

Does this produce the various different foldings that each protein can often "sit" in?

Can it take temperature and other environmental conditions into account?

Can you specify that a particular ligand or electrical current is present so that you can see the resultant shape change?

Is all the source code for this available so that other scientists can build on top of this, or will we have to go through a paid or SaaS google API to use it?

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

#323

Post by uoaei »

astroalex wrote: post #272
uoaei wrote: post #256

No, they didn't. They approximated a solution to protein folding. The two are different concepts -- this isn't the typical HN pedantry. "Solving" the problem would entail developing an interpretable algorithm for taking a string of amino acids and determining the 3D structure once folded. Approximating a solution would entail simulating that algorithm, which is what their neural network is doing. It is of course usua…

The distinction you're making between "solved" and "closely approximated" makes logical sense to me. However, if I'm interpreting the AlphaFold results correctly, this distinction isn't practically significant, right? If you can approximate an algorithm with error that is "below the threshold that is considered acceptable in experimental measurements" (to quote another HN comment), then you have something as good as…

Practically there is little difference if all you're interested in is determining folds from protein sequences.

The difference comes in developing a theory for generalizing the study of protein folding as a scientific pursuit.

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

#324

Post by visarga »

mensetmanusman wrote: post #55

This is amazing, if we can simulate multi-protein interactions, you could imagine in our lifetimes being able to see a fully computation driven simulation of a human blood cell. That would be a huge breakthrough.

What amazed me most was that they used hundreds of millions of unlabelled protein scans. This means we can collect massive data in a new modality, besides the usual suspects: images, video, audio, text, lidar and sensors. Soon I expect neural implant data to be massive as well.

They surely did unsupervised training on raw data and then fine-tuning on the 170K labelled sequences. I expect the data volume could be increased by orders of magnitude in the next couple of years and we'll see a GPT-3 like jump.

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

#325

Post by uoaei »

deeviant wrote: post #295
uoaei wrote: post #256

No, they didn't. They approximated a solution to protein folding. The two are different concepts -- this isn't the typical HN pedantry. "Solving" the problem would entail developing an interpretable algorithm for taking a string of amino acids and determining the 3D structure once folded. Approximating a solution would entail simulating that algorithm, which is what their neural network is doing. It is of course usua…

> this isn't the typical HN pedantry. Then launches into what can only be recognized as an exercise in pedantry.

"Pedantry" implies that the distinction is not meaningful.

This is true if you're only paying attention to how this system can be utilized to answer questions posed to it.

This achievement by itself, however, does not do much to push the science of protein folding much further. Those advances will come when people poke, prod, and break the model to develop a unified theory for protein folding.

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

#326

Post by enchiridion »

breatheoften wrote: post #182

Anyone care to muse about appropriate investment strategies based on the not previously feasible research approaches that might now be possible? Should we expect to see faster progress in large well capitalized bioscience companies -- or a sudden increase in the viability of smaller biotech and/or biotech startups ...? Are we gonna see top talent fleeing the old biotech companies to start their own ventures with a ne…

I agree. I think a company that is working on large scale automated bio experiments would be well positioned to take advantage of something like this.

What companies are doing that work?

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

#327

Post by FrojoS »

breatheoften wrote: post #314
martinpw wrote: post #305

Earlier quoted context omitted.

Why is the graph not monotonically increasing? Does the complexity of the problem to be solved increase each time? If so, does that make the relative improvement from the previous result even more impressive?

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 the top score would be, that previous winners are not re-submitted unchanged. For instance AlphaFold v1 seems to not have been submitted to the latest competition.

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

#328

Post by deeviant »

sundarurfriend wrote: post #114
asdfasgasdgasdg wrote: post #68

Earlier quoted context omitted.

> the outstanding question as far as to whether or not this is useful for evaluating novel proteins That is not an outstanding question. The test on which DeepMind scored high marks is a test of how well the algorithm folds novel proteins -- proteins whose ground-truth structure has not yet been published.

You missed the actual outstanding question in their comment: > the outstanding question ... is going to be how good is the confidence metric at telling the user to trust or not trust the results.

You don't generally look at neural network output like that.

There is generally a threshold, less than X, not the class, equal or more, is the class. Then you run the network with the same threshold on a known data set and compute a confusion matrix, which tells you about the error, I don't even want to know what a confusion matrix analogue for 3D geometry would look like but I'm sure they have something.

This is literally the process that one does in taking part of the this. And the error rate (specifically the lack of errors) is what is everybody is talking about. 90 is just as accurate as we can get with experimental measurement. It's likely at this point the source of error is in the data set (we can only train on data we experimentally measure and these are not perfect measurements). It's also possible, at this point, the model generalized so well that when it deviates from experimental measurements it's actually correct and the experimental value was the one that was wrong.

So no, the outstanding question is not "is going to be how good is the confidence metric at telling the user to trust or not trust the results.". Nobody is going to be looking confidence values when it model is giving an output, they are going to be looking at the overall error rate across a broad spectrum of proteins to get a sense of it's accuracy.

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

#329

Post by diab0lic »

misnome wrote: post #258
penagwin wrote: post #212

Earlier quoted context omitted.

This. I believe technically just linear regression could be considered "machine learning".

I've seen people at bio conferences actively calling linear regression machine learning.

This is likely because linear regression meets most widely accepted definitions of machine learning. [0][1] It is simple and very effective when learning in linear space.

[0] https://en.wikipedia.org/wiki/Machine_learning

[1] https://www.cs.cmu.edu/~tom/mlbook.html

Post reply on HN