5 ms·

>This “protein-folding problem,” as it came to be known, baffled scientists until this year, when the papers we’ve deemed the 2021 Breakthrough of the Year were

>This “protein-folding problem,” as it came to be known, baffled scientists until this year, when the papers we’ve deemed the 2021 Breakthrough of the Year were published. This is a straight lie and the author should be embarrassed. There's not one scientist alive that can explain how proteins fold from their amino acid sequence. There's not one algorithm in existence that can predict it de novo. I'm so tired of seeing this lie repeated over and over. AlphaFold did not gain us any new information on how proteins fold. It just got really good at modelling them from existing data where existing data is available. Which is great. But still ultimately limited by the same categorical limitations of homology modeling. It simply has not solved protein folding. Yes it's amazing. Yes it's worthy of lots of praise. No, it did not solve protein folding, ESPECIALLY in the way that is described here.

epgui 5y ago
> There's not one scientist alive that can explain how proteins fold from their amino acid sequence. As a biochemist, what I understood from this editorial was totally reasonable: - scientists understand that electrostatic forces guide protein folding, and steric factors limit the conformations a protein can take - scientists understand that the primary structure determines the secondary and tertiary structures of proteins - scientists know that protein folding is aided by chaperone proteins - scientists understand how thermodynamics and entropy affect protein folding Despite all of the above, scientists cannot solve protein folding de novo. New tools now allow us to make better, cheaper predictions which are testable. Sounds reasonable to me.
- COGlory 5y ago
  As a structural biologist, I have to ask, how did you draw that from this? > seemed possible that the three-dimensional structure of a protein could be predicted based on the sequence of its amino acids. That simply is not what AlphaFold (or any homology modeling) is doing, and certainly not: >This is a breakthrough on two fronts. It solves a scientific problem that has been on the to-do list for 50 years. And just like Fermat’s Last Theorem or gravitational waves, scientists kept at it until it was done.
  - epgui 5y ago
    > That simply is not what AlphaFold (or any homology modeling) is doing I don't know if we're getting hung up in semantic differences, but how would you describe homology modelling, or what AlphaFold is doing? Because for me, it takes as input an AA sequence and gives as output a 3D structure prediction.
    - COGlory 5y ago
      It's not a semantic difference. That's the problem. There is a difference between knowledge of protein folding (which is what the article is talking about), i.e. understanding proteins well enough to predict how they will fold, and modeling a folded structure (which is what AlphaFold is doing, based off of other pre-existing experimental structures). AlphaFold isn't simulating folding at all. It's modeling what the final structure should look like based off other data, not knowledge of how a protein folds. That's what is frustrating me so much with this topic. It's being sold as "we can predict how proteins will fold" as opposed to "we can model a folded protein based on its sequence". There's a huge difference in our understanding of protein folding, between those two points.
      - epgui 5y ago
        It does sound like a semantic difference. To me, and I believe to most in the field, the protein folding problem was always pretty much about figuring out the final 3D structure (the final state), and not necessarily about the act of folding (ie.: the gerundive) itself. There is a lot of value in learning more about folding (in the gerundive sense), but the focus has always been on finding the folded structure, because as you very well know, structure is function.
        t_serpico 5y ago
        The “protein folding problem” consists of three closely related puzzles: (a) What is the folding code? (b) What is the folding mechanism? (c) Can we predict the native structure of a protein from its amino acid sequence? (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2443096/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2443096/). It seems the purist definition of solving protein folding means specifying the sequence of actions to arrive at the native structure for a given input sequence. This implies your solution is grounded in a physical reality. Such approaches (i.e MD simulations) have not been very successful to my knowledge and are computationally intractable for most proteins. Even if you are able to do this, how do you verify that your folding mechanism is in fact correct? We don't have ground truth experimental data at the temporal resolution one would want to verify such a prediction. For now, the only practical framing of the protein folding problem is to predict a native structure given an input sequence. This is simply because we have the requisite data...

epgui 5y ago

> There's not one scientist alive that can explain how proteins fold from their amino acid sequence. As a biochemist, what I understood from this editorial was totally reasonable: - scientists understand that electrostatic forces guide protein folding, and steric factors limit the conformations a protein can take - scientists understand that the primary structure determines the secondary and tertiary structures of proteins - scientists know that protein folding is aided by chaperone proteins - scientists understand how thermodynamics and entropy affect protein folding Despite all of the above, scientists cannot solve protein folding de novo. New tools now allow us to make better, cheaper predictions which are testable. Sounds reasonable to me.

COGlory 5y ago
As a structural biologist, I have to ask, how did you draw that from this? > seemed possible that the three-dimensional structure of a protein could be predicted based on the sequence of its amino acids. That simply is not what AlphaFold (or any homology modeling) is doing, and certainly not: >This is a breakthrough on two fronts. It solves a scientific problem that has been on the to-do list for 50 years. And just like Fermat’s Last Theorem or gravitational waves, scientists kept at it until it was done.
- epgui 5y ago
  > That simply is not what AlphaFold (or any homology modeling) is doing I don't know if we're getting hung up in semantic differences, but how would you describe homology modelling, or what AlphaFold is doing? Because for me, it takes as input an AA sequence and gives as output a 3D structure prediction.
  - COGlory 5y ago
    It's not a semantic difference. That's the problem. There is a difference between knowledge of protein folding (which is what the article is talking about), i.e. understanding proteins well enough to predict how they will fold, and modeling a folded structure (which is what AlphaFold is doing, based off of other pre-existing experimental structures). AlphaFold isn't simulating folding at all. It's modeling what the final structure should look like based off other data, not knowledge of how a protein folds. That's what is frustrating me so much with this topic. It's being sold as "we can predict how proteins will fold" as opposed to "we can model a folded protein based on its sequence". There's a huge difference in our understanding of protein folding, between those two points.
    - epgui 5y ago
      It does sound like a semantic difference. To me, and I believe to most in the field, the protein folding problem was always pretty much about figuring out the final 3D structure (the final state), and not necessarily about the act of folding (ie.: the gerundive) itself. There is a lot of value in learning more about folding (in the gerundive sense), but the focus has always been on finding the folded structure, because as you very well know, structure is function.
      - t_serpico 5y ago
        The “protein folding problem” consists of three closely related puzzles: (a) What is the folding code? (b) What is the folding mechanism? (c) Can we predict the native structure of a protein from its amino acid sequence? (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2443096/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2443096/). It seems the purist definition of solving protein folding means specifying the sequence of actions to arrive at the native structure for a given input sequence. This implies your solution is grounded in a physical reality. Such approaches (i.e MD simulations) have not been very successful to my knowledge and are computationally intractable for most proteins. Even if you are able to do this, how do you verify that your folding mechanism is in fact correct? We don't have ground truth experimental data at the temporal resolution one would want to verify such a prediction. For now, the only practical framing of the protein folding problem is to predict a native structure given an input sequence. This is simply because we have the requisite data...