AlphaFold Protein Structure Database

moyix wrote: post #3

Anyone else getting a 403 Forbidden? If so it might be better to link to the paper instead: https://www.nature.com/articles/s41586-021-03828-1

Works fine for me. Must have been a temporary glitch.

Re: AlphaFold Protein Structure Database

#6

Post by ricksunny » Thu, Jul 22, 2021, 3:44 PM UTC

I’m sorry but why don’t tbey just release the ability for a user to enter a known real-world sequence’s accession number from Genbank / GISAID, and generate the protein structure from that? Why do they have to abstract the user from the process by only exposing a completed database of the protein structures the Alphafold researchers decided would be worth producing?

Re: AlphaFold Protein Structure Database

#7

Post by sherjilozair » Thu, Jul 22, 2021, 3:50 PM UTC

ricksunny wrote: post #6

I’m sorry but why don’t tbey just release the ability for a user to enter a known real-world sequence’s accession number from Genbank / GISAID, and generate the protein structure from that? Why do they have to abstract the user from the process by only exposing a completed database of the protein structures the Alphafold researchers decided would be worth producing?

DeepMind has already released the open source code and model parameters. The database makes it easier to access the predictions.

Re: AlphaFold Protein Structure Database

#8

Post by sveme » Thu, Jul 22, 2021, 3:51 PM UTC

ricksunny wrote: post #6

I’m sorry but why don’t tbey just release the ability for a user to enter a known real-world sequence’s accession number from Genbank / GISAID, and generate the protein structure from that? Why do they have to abstract the user from the process by only exposing a completed database of the protein structures the Alphafold researchers decided would be worth producing?

I'd guess the ad-hoc simulation of the structure is computationally quite expensive and takes a while, though that's just a guess and I haven't read the original paper yet.

Re: AlphaFold Protein Structure Database

#9

Post by tazjin » Thu, Jul 22, 2021, 3:52 PM UTC

ricksunny wrote: post #6

I’m sorry but why don’t tbey just release the ability for a user to enter a known real-world sequence’s accession number from Genbank / GISAID, and generate the protein structure from that? Why do they have to abstract the user from the process by only exposing a completed database of the protein structures the Alphafold researchers decided would be worth producing?

You can use the open-source code, and we also have a Colab notebook for that: https://bit.ly/alphafoldcolab

More info: https://deepmind.com/blog/article/putting-the-power-of-alpha...

Re: AlphaFold Protein Structure Database

#10

Post by sdbrown » Thu, Jul 22, 2021, 3:56 PM UTC

This is a fabulous convenience! The reach of this ready-to-go data will be much larger (in some directions) than the model and CASP results themselves.