Prions don't build new prions form smaller parts, like virus do. Prions just change the shape of a complete protein that is already build. But each prion can change the shape of only one type of protein, not any protein that is floating around.
Interestingly enough, looking at the powdered form of retinal is a visual quine. (it is a chemical in your eyes that is produced during visual stimulation) https://arbesman.substack.com/p/a-visual-quine
This can't be true, many bioscience people have commented assuring us that because Deep Fold doesn't literally spit out new proteins, its nothing new, and saying otherwise is misleading
What do you mean? It certainly can fold new proteins that were not part of its training set, otherwise it is useless. Maybe you mean that it cannot fold proteins in new ways that were never seen in its training set though ? But new proteins might fold in known ways, right?
It's sarcasm, every time DeepMind does something revolutionary in protein folding, some nominal biochemist jumps in and says "oh no, because it didn't literally automate 100% of my job today, this is overhyped"
Well it hasn't improved the human condition one bit yet (except for researchers desperate for paper citations), so whether it lives up to the "hype" remains to be seen.
DeepMind has a website highlighting some of the applications of AlphaFold (https://unfolded.deepmind.com), and apparently it has improved at least some lines of research.
Yes, again very good for the company's PR and padding researcher CVs. But no actual benefit to humankind has happened - again, the hype remains to be justified (or not). Notice the carefully vapid wording: doing [some nebulous thing] to "fight" [some terrible sounding thing], or "this could do [some holy grail amazing thing]".
Agreed, like the James Web telescope, a load of hype about the beautiful images and how much would be discovered as a result of it.
Sure we will be able to see further into the universe, but I am still trying to work out what practical application it will have beyond pretty pictures.
Anyway, the hype is over, no one in my social media circle who were amazed by it have gone on to post a second picture. Meanwhile a drunk scientist put out a picture of a piece of chorizoand trolled them all. I see it as a merger of "I support the current thing" with "I fucking love science".
actually, tehre are a number of papers published now using alphafold predictions to benefit. No, nobody cured the world of all diseases, but it does represent a sea-change in prediction quality.
According to TA, a novel protein nanoparticle is part of some COVID vaccine. It's not far-fetched to assume that such nanoparticles will be designed with machine learning tools in the future.
And then if that happens, people will say, oh but that innovation was discovered a long time ago, these companies haven't done anything earth-shattering recently.
I would pay skepticism to entrepreneurial commentaries on AlphaFold (CEO of a company based on competing technology has hot take, who would have guessed?), but I would mind the caveats and areas of improvement pointed out by domain experts i.e. Baker, AlQuraishi, etc. Keeping in mind that the latter usually spend more time praising AlphaFold than not -- and yes, they have more domain knowledge than pretty much anyone on HN:
It's almost entirely positive. There was a slice in time where I was effectively paid to find as many gaps in AlphaFold as I could, and, even then, I couldn't help but be impressed.
Snark aside, this is not AlphaFold per se, but sort of a reverse AlphaFold, the same way protein design can be defined as the inverse protein folding problem.
Tangent: Has AlphaFold and other "ML"/"AI" programs made progress on describing how proteins fold? Or is it solely for final (post-folding) coordinates/dihedral angles? Has there been any progress on understanding folding? Results from folding-at-home? Is it treated as too tough of a problem?
> This can't be true, many bioscience people have commented assuring us that because Deep Fold doesn't literally spit out new proteins, its nothing new, and saying otherwise is misleading
From the article...
> But when they instructed microorganisms to make their creations in the labs, none of the 150 designs worked. “They didn’t fold at all: they were just gunk at the bottom of the test tube,” says Baker.
For the record, David Baker (and other groups) have been successfully designing new proteins just from their own understanding of protein folding, for years. So "Deep Fold" still has yet to catch up to our own limited human understanding.
Those 150 designs were probably early proofs of concept hallucinated using e.g. RoseTTAFold; Baker has been thinking about protein hallucination for a while already.
Indeed, a simple spell for animating an arbitrary number of broomsticks to fetch an amount of water given by argument. [0] Oh, and it was given to us by a demon whose output we don't really understand. What could go wrong?
Thankfully, if I'm understanding the article properly, the broomsticks didn't properly animate, though both the demon and human researchers figured they would. I'm all for demon-assisted magical research, but I very much doubt that the next action taken will be "abandon this avenue of research until we understand the demon's process better". We should probably understand this kind of thing if we're going to be messing with it.
> But when they instructed microorganisms to make their creations in the labs, none of the 150 designs worked. “They didn’t fold at all: they were just gunk at the bottom of the test tube,” says Baker.
When it works properly (when they fold), what do you see in the test tube instead? Does it no longer look like “gunk”? What I mean is, is this difference visible to the naked eye?
When you express recombinant proteins in bacteria, you grow them and then let them produce your protein of interest. Then you destroy the bacteria and separate the soluble parts (where you hope your protein stays) from the insoluble ones (protein aggregates, cell membrane components, etc.), usually by ultracentrifugation. The soluble portion (or supernatant) is a clear liquid, while the insoluble fraction (or precipitate) is a whiteish, gunky solid.
So, to answer your question, soluble proteins would look clear. However, to actually check for the presence of your protein of interest in each fraction you load them in a polyacrylamide gel and pass current through it to separate all proteins in the mixture, using a method called electrophoresis[1].
SDS-PAGE is the normal way, at least for an initial test expression. For a Western blot or ELISA you need antibodies specific to your protein, which you probably don’t have. An easier way to identify, for example, would be to cut the SDS-PAGE band and perform mass spectrometry on it to try to elucidate its protein sequence.
To follow on, typically proteins produced this way have a tag, or handle, to specifically pull it out from all the other proteins. A typica tag is 6 x Histidine amino acids at the beginning of the protein. These histidines bind to nickel, which is immobilized on a resin. You can use antibodies against the 6xHis to Western for your protein of desire without needing a protein specific antibody.
As an aside, in the first real purification and isolation of a protein (Rho), there is a figure of a test tube with what just looks like liquid, and a caption "Rho protein in solution".
I imagine some soluble proteins actually make the solution look colorful (autofluorescence) but at the concentrations required you'd be pushing up against aggregation
Cofactor- or chromophore-binding proteins are usually colored, which is quite useful during expression and purification. The quintessential example is GFP or GFP-tagged proteins.
Nothing. If they fold well and are water soluble, the solution remains transparent because the protein remains solvated. If they don't fold well, they expose hydrophobic regions that then bind to the hydrophobic regions on neighboring proteins, causing a chain reaction that results in a solid chunk of protein that isn't dissolved.
Not all proteins are water soluble in nature, however. Plenty of membrane proteins and other proteins exist that don't fold well when expressed in a micro-organism.
As a solution to the inverse folding problem, this is THE enabling technology for molecular nanotechnology. I suggest anyone interested in this read Drexler’s Radical Abundance which talks about what is possible with atomically precise manufacturing built with custom designed proteins.
I'm gonna check that book, I bet I will learn a lot.
I sometimes wonder how we are going to solve the scaling problems around silicon and energy for this kind of AI applications. Think about this: a life-saving cancer drug may need to be custom-designed for a patient by using a few Giga-joules of energy to feed a data-center.
Most of the energy goes into building these models. Applying the model takes orders of magnitude less energy, and can run on efficient hardware that is optimized for that purpose. It's not going to take a gigaojoules of energy to infer custom solutions from a prompt.
Good luck- this has been a long-term interest of mine, but one that has not really led to any forward progress. Lithography continues to be a far more approachable technology.
This news is so shocking I created an account here after lurking for like a decade. I've yet to read the paper itself but if this is legitimate and not "New battery chemistry is 1000x better!!" type clickbait then holy shit. I'm surprised the article doesn't mention nanotechnology. The potential applications of this are insane and go well beyond medicine.
Two attempts. The first failed, the second succeeded.
>When Baker and his team applied this second network to their hallucinated protein nanoparticles, it had much greater success making the molecules experimentally. The researchers determined the structure of 30 of their new proteins using cryo-electron microscopy and other experimental techniques, and 27 of them matched the AI-led designs2. The team’s creations included giant rings with complex symmetries, unlike anything found in nature. In theory, the approach could be used to design nanoparticles corresponding to almost any symmetric shape, says Lukas Milles, a biophysicist who co-led the effort. “It is electrifying to see what these networks can do.
What are the usecases for designer proteins outside of medicine? The article mentions pollution cleanup. Anybody have any resources to learn about the mechanisms involved in a use case like that?
One goal of protein design is to create bespoke enzymes that catalyze reactions of interest, such as the degradation of target compounds. Keep in mind that the vast majority of de novo designed proteins to date lack a defined function and have been used to showcase structure rather than functionality.
Any bio-processes that you can think of. For example, spiders fabricate ultra-strong fibers using a few cubic millimeters. Compare that volume to the smallest Kevlar factory we can set up.
A lot of applications will be about designing or adapting biological processes; so they wouldn't be enabled by just protein design but would need way more "tooling". It's worth noting that the computational viability of much of that tooling is boosted by AI as well. That I think is the main value and the "new thing": we got a new, powerful class of "hacks" that we can use to shortcut and scale in-silico bio-research.
Laundry detergent. Genentech, a biotech company, engineered a version of a protease (a protein that cuts other proteins) to work at high temps, then formed a spinoff with Corning to license that tech for billions to laundry detergent companies. https://en.wikipedia.org/wiki/Genencor
59 comments
[ 0.14 ms ] story [ 120 ms ] threadI think the risk is almost almost almost zero.
So a prion quine would be a self replicating nano machine, a semi popular doomsday scenario.
* Has a lower bond energy than the correct fold (so can't readily be fixed), and * Can cause its correctly-folded version to also misfold.
So far, as far as I'm aware, there have only ever been a handful of misfolds that fit this pattern.
https://en.wikipedia.org/wiki/Ribosome#Structure
Sure we will be able to see further into the universe, but I am still trying to work out what practical application it will have beyond pretty pictures.
Anyway, the hype is over, no one in my social media circle who were amazed by it have gone on to post a second picture. Meanwhile a drunk scientist put out a picture of a piece of chorizoand trolled them all. I see it as a merger of "I support the current thing" with "I fucking love science".
https://moalquraishi.wordpress.com/2021/07/25/the-alphafold2...
It's almost entirely positive. There was a slice in time where I was effectively paid to find as many gaps in AlphaFold as I could, and, even then, I couldn't help but be impressed.
From the article...
> But when they instructed microorganisms to make their creations in the labs, none of the 150 designs worked. “They didn’t fold at all: they were just gunk at the bottom of the test tube,” says Baker.
For the record, David Baker (and other groups) have been successfully designing new proteins just from their own understanding of protein folding, for years. So "Deep Fold" still has yet to catch up to our own limited human understanding.
Thankfully, if I'm understanding the article properly, the broomsticks didn't properly animate, though both the demon and human researchers figured they would. I'm all for demon-assisted magical research, but I very much doubt that the next action taken will be "abandon this avenue of research until we understand the demon's process better". We should probably understand this kind of thing if we're going to be messing with it.
[0] https://www.youtube.com/watch?v=VErKCq1IGIU
When it works properly (when they fold), what do you see in the test tube instead? Does it no longer look like “gunk”? What I mean is, is this difference visible to the naked eye?
So, to answer your question, soluble proteins would look clear. However, to actually check for the presence of your protein of interest in each fraction you load them in a polyacrylamide gel and pass current through it to separate all proteins in the mixture, using a method called electrophoresis[1].
[1]: https://en.wikipedia.org/wiki/Polyacrylamide_gel_electrophor.... In the first picture of the article each band represents one (or several, if you’re unlucky!) protein of a determined molecular weight or size.
I imagine some soluble proteins actually make the solution look colorful (autofluorescence) but at the concentrations required you'd be pushing up against aggregation
Not all proteins are water soluble in nature, however. Plenty of membrane proteins and other proteins exist that don't fold well when expressed in a micro-organism.
Radical Abundance in video form: https://m.youtube.com/watch?v=1bw6Zi17DBI
If anyone is interested in making this happen, contact me.
I sometimes wonder how we are going to solve the scaling problems around silicon and energy for this kind of AI applications. Think about this: a life-saving cancer drug may need to be custom-designed for a patient by using a few Giga-joules of energy to feed a data-center.
>When Baker and his team applied this second network to their hallucinated protein nanoparticles, it had much greater success making the molecules experimentally. The researchers determined the structure of 30 of their new proteins using cryo-electron microscopy and other experimental techniques, and 27 of them matched the AI-led designs2. The team’s creations included giant rings with complex symmetries, unlike anything found in nature. In theory, the approach could be used to design nanoparticles corresponding to almost any symmetric shape, says Lukas Milles, a biophysicist who co-led the effort. “It is electrifying to see what these networks can do.
A lot of applications will be about designing or adapting biological processes; so they wouldn't be enabled by just protein design but would need way more "tooling". It's worth noting that the computational viability of much of that tooling is boosted by AI as well. That I think is the main value and the "new thing": we got a new, powerful class of "hacks" that we can use to shortcut and scale in-silico bio-research.