# AlphaFold Just Won The Nobel Prize! 🧬

## Метаданные

- **Канал:** Two Minute Papers
- **YouTube:** https://www.youtube.com/watch?v=G7gdOPEd6mU
- **Дата:** 13.09.2022
- **Длительность:** 6:47
- **Просмотры:** 137,315
- **Источник:** https://ekstraktznaniy.ru/video/13447

## Описание

❤️ Check out Cohere and sign up for free today: https://cohere.ai/papers

📝 The paper "Highly accurate protein structure prediction with #AlphaFold" is available here:
https://www.deepmind.com/blog/alphafold-reveals-the-structure-of-the-protein-universe
https://www.nature.com/articles/s41586-021-03819-2
https://www.deepmind.com/blog/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology

Database:
https://www.uniprot.org/

❤️ Watch these videos in early access on our Patreon page or join us here on YouTube: 
- https://www.patreon.com/TwoMinutePapers
- https://www.youtube.com/channel/UCbfYPyITQ-7l4upoX8nvctg/join

🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:
Aleksandr Mashrabov, Alex Balfanz, Alex Haro, Andrew Melnychuk, Benji Rabhan, Bryan Learn, B Shang, Christian Ahlin, Eric Martel, Geronimo Moralez, Gordon Child, Ivo Galic, Jace O'Brien, Jack Lukic, John Le, Jonas, Jonathan, Kenneth Davis, Klaus Busse, Kyle Davis, Lorin Atzb

## Транскрипт

### Intro []

Dear Fellow Scholars, this is Two Minute  Papers with Dr. Károly Zsolnai-Fehér. Today you will see history in the making.   Yes, we are going to have a look at AlphaFold   from DeepMind, perhaps one of the most  important papers of the last few years.    This is a true gift to humanity. And, if  even that is not good enough, it gets better,

### What is AlphaFold [0:17]

because today you will see how this  gift has now become 200 times bigger. So, what is AlphaFold? AlphaFold is an AI that is  capable of solving protein structure prediction,   which we will refer to as protein folding.    Okay…but what is a protein  and why does it need folding? A protein is a string of amino acids, these are  the building blocks of life. This is what goes in,   which in reality, has a 3D structure. And that is  protein folding. Letters go in, a 3D object comes   out. This is hard. Really hard. For instance,  DeepMind’s earlier AI programs were able to play   hard games like Chess, Go, and Starcraft 2, and  none of these AIs were a match to protein folding. Yet, it has become the world’s  best solution at the CASP event,   I’ve heard DeepMind CEO Demis Hassabis  call it the Olympics of protein folding.    If you look at how teams of scientists prepare  for this event, you will probably agree that yes,   this is indeed the Olympics of protein folding.   And what is even more incredible is that it has   reached a score of 90, which means two amazing  things. One, in absolute terms, AlphaFold 2   is considered to be about three times better than  previous solutions. And, it gets even better, the   score of 90 means that we can consider protein  folding as a mostly solved problem. Wow.    This is not something that I thought  I would be able to say in my lifetime. Now, we said that this is a  gift to humanity. Why is that?    Where is the gift part here?   What makes this a gift? Well, a little after publishing the paper,  DeepMind made these 3D structure predictions   available for free for everyone. For instance,  they have made their human protein predictions   public. Beyond that, they already have  made their predictions public for yeast,   important pathogens, crop species, and more. I also said that AlphaFold 2 is not the  end, but a start of something great.    I promised to tell you how it improves over  time, so here are two incredible revelations. One, we noted that AlphaFold will likely help  us fight diseases and develop new vaccines. And   hold on to your papers, because this is not  just talk, this is now a reality today: look,   this group was already able to accelerate their  research work a great deal due to AlphaFold,   in fighting antibiotic resistant bacteria. They  are studying how to fight deadly bacteria that   modify their own membrane structure so that  the antibiotic can’t get in. That is a hugely   important project for all of us. They noted that  they have been working vigorously on this project   for 10 years, and now that they have access  to AlphaFold, they could get a prediction done   in 30 minutes, and this is something that was not  possible to do for 10 years before that. That is   absolutely incredible. This is history in  the making. I love to be able to live through   times like this. So thank you so much  for sharing this amazing moment with me. So this was one project, and what is this here?    Oh yes, these are some more  of these AlphaFold predictions   that were already referenced from other scientific  research papers. This is making big waves. And two, this gift to humanity has now  become bigger. So much bigger. Initially,   the protein prediction database they published  contained 1 million entries. And now, if you have   been holding on to your papers, now squeeze that  paper, because they have grown it from 1 million   to 200 million structures. That is 200x within  less than a year. Wow. So just imagine how many   more of these amazing medicine research projects  it will be able to accelerate. I cannot even

### Conclusion [4:58]

fathom how many lives this AI project is going  to save in the future. What a time to be alive! Here is the distribution of the data  across different categories, and these   small circles still represent what was available  before, and the big ones show what is available   from now on. Absolutely amazing. It seems  clearer and clearer that we are now entering   the age of AI-driven art creation, but it  might also be possible that we will also   soon enter the age of AI-driven medicine. These  tools can help us so much! How cool is that? So, does this get your mind going? What would you  use this for? Let me know in the comments below! Thanks for watching and for your generous  support, and I'll see you next time!