Oliver Clarke points out that VP of research at DeepMind, Pushmeet Kohli, says they hope to release AlphaFold 3 code in 6 months. "We love the excitement & results from the community on AlphaFold 3 and are doubling the AF Server daily job limit to 20. Happy to also share that we're working on releasing the AF3 model (incl weights) for academic use, which doesn’t depend on our research infra, within 6 months. Pushmeet Kohli. 11:27 AM · May 13, 2024" https://x.com/pushmeet/status/1790086453520691657 If that happens we will try to add ChimeraX support for running AlphaFold 3 predictions. Tom

On May 13, 2024, at 1:42 PM, Oliver Clarke <olibclarke@gmail.com> wrote:

Hi Tom, hope all is well! It seems like this stance might change (maybe because of all the pushback from the community?):

https://x.com/pushmeet/status/1790086453520691657

So maybe it will be possible by the end of the year...

Cheers Oli

...

On May 13, 2024, at 4:23 PM, Tom Goddard via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Hi Sohail,

Google did not released the code for AlphaFold 3 so the only way to run it is from their web page. They also have not provide a web service that would allow programs like ChimeraX to submit prediction jobs. I asked a contact at Google DeepMind if they plan in the future to provide a web service or provide the code, and they said they have no plans to do so at this time. Details here

https://www.rbvi.ucsf.edu/trac/ChimeraX/ticket/15135

Google also is not allowing researchers to run predictions with small molecule ligands except for some very common ones like ATP. It appears Google is treating AlphaFold 3 as a commercial product so the most useful capabilities will only be accessible to the drug discovery company Google has formed called Isomorphic Labs.

I will be making the ChimeraX display of predicted aligned error (PAE) work with AlphaFold 3 output. Currently it is broken because AlphaFold 3 now has predicted aligned error for individual atoms of ligands, ions, glycosylation and non-standard residues while AlphaFold 2 only had values for residues of proteins. I expect to put this fix in ChimeraX today or tomorrow and it will be in the ChimeraX daily builds and ChimeraX 1.8 release candidates.

Tom

...

On May 11, 2024, at 6:18 AM, Khawar Siddiqui via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Hi, Greetings! When ChimeraX will get Alphafold 3?

Thanks and best wishes

Dr K Sohail Siddiqui, School of Biotechnology and Biomolecular Sciences (BABS), Room 420, Building D26, The University of New South Wales (UNSW), Randwick, High Street, Sydney, Australia. NSW 2052 Office: +61 2 938 55193 <tel:+61 2 938 55193> email: z3030204@ad.unsw.edu.au; sohailsiddiqui1995@yahoo.com _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/

_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/