How to pronounce“alphafold”in English

- AlphaFold 2 was really a system about designing our deep learning.

AlphaFold 2 was trained on the exact same data with much, much better machine learning as AlphaFold 1.

AlphaFold is a specialized hybrid system.

AlphaFold was nearly 10 years ago now, and it's transforming pharmaceuticals, but there's so much more that can be done, right?

AlphaFold understood the probability distribution over configurations of proteins.

The AlphaFold 2 paper has been cited over 30,000 times.

Like AlphaFold.

As the AlphaFold 2 team searched for better algorithms, they turned to the transformer.

Gone was AlphaFold 1's deep neural network that started with one tower and predicted the other.

- That's why live AlphaFold folding videos can show it doing some weirdly non-physical stuff.

Nobody thinks that AlphaFold, which is this AI system that folds proteins is conscious, because it doesn't speak to us, but it's basically doing

There should be 20 AlphaFolds, right?

DeepMind hoped to change this with AlphaFold.

So Hassabis recruited John Jumper to lead AlphaFold.

We don't really explicitly tell AlphaFold that.

When you came up with AlphaFold, which was, it was obviously a huge, Nobel-worthy scientific breakthrough, probably also of commercial value, and you

"Like figure out how to take AlphaFold 3 "and make a drug that cures cancer "and come back to me once you find something amazing"

Grokking quantum simulations similar to AlphaFold, right?

And the same with the AlphaFold with protein folding.

Its first iteration, AlphaFold 1, was a standard off-the-shelf deep neural network like the ones used for computer vision at that time.

As input, AlphaFold took the protein's amino acid sequence and an important set of clues given by evolution.

These evolutionary tables were an important input for AlphaFold.

With this framework, AlphaFold entered CASP 13 and it immediately turned heads.

Instead, AlphaFold 2's EVO Former builds each tower separately.

The geometrical features learned by this network are passed onto AlphaFold 2's second main innovation, the structure module.

I mean, AlphaFold 3 on the interaction side is surprising, that you can make any kind of progress on that direction.

And I think a great example of this is something called AlphaFold, which is an AI that can predict how a protein is folded purely from its sequence.

And to analyze them, the AlphaFold team built their own version of the transformer called an EVO Former.

- For each triplet of amino acids, AlphaFold applies the triangle inequality.

- For many proteins, AlphaFold 2 predictions were virtually indistinguishable from the actual structures

Then in one fell swoop, AlphaFold came in and unveiled over 200 million of them.

In just a few months, AlphaFold advanced the work of research labs worldwide by several decades.

the proteins so that, because not only was AlphaFold accurate, it was extremely fast, it could fold a protein in a matter of seconds,

We also had depended on public data to train the first versions of AlphaFold, so it only felt right to give back to that community, the structural biology

As output, instead of directly producing a 3D structure, AlphaFold predicted a simpler 2D pair representation of that structure.

So the moment we got back from Seoul, we started the AlphaFold project.

And you're mentioning that some of this work, AlphaFold specifically, comes from DeepMind.

Large language models use attention to predict the most appropriate word to add to a sentence, but AlphaFold also has sequential information,

In December, 2020, DeepMind returned to a virtual CASP with AlphaFold 2, and this time they did it.

And indeed, I've talked to some people at Google and they've said, well, we could use AlphaFold for some of these things and then we'll put it out for free

And AI has the prospect of delivering these things, and we've seen the early shoots, AlphaFold and the protein folding, for example, with the medical innovations.

Some of these breakthroughs include AlphaGo, the first program to defeat a world champion at the game of Go, and AlphaFold, which solved the 50-year grand challenge

'cause, you know, there's three million researchers around the world that use AlphaFold, Pretty much every day.

And then we've tried to ourselves push that downstream with Isomorphic Labs, an Alphabet spinout that is, building, you can think, several more type of AlphaFold

We think about that a lot actually, and that goes back to some examples of that I can give is back to the AlphaFold question,

And what was great, what we did actually in the early days of seeding some collaborations, what you could do with AlphaFold, we worked with the DNDi,

especially with the Alpha X projects, so I'm thinking AlphaGo, of course, AlphaFold.

- Yeah, I think that there are actually a huge class of problems that could be couched in this way, the way we did AlphaGo and the way we did AlphaFold,

I don't fully understand all the incredible stuff that the teams are building-- Gemini or whatever-- are doing, or the team behind AlphaFold.