Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Dear Tim, I remain enthusiastic about the value to my work, research and service to the crystallographic community, in what I learn using twitter ie X. I also appreciate the learned feedback I get to what I share in my tweets. I do not use X to share my opinions on “wider issues”. I have tried

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

That is very encouraging Tom. Thanks very much. Nice article and talk. Kind regards Arif On Thu, Nov 30, 2023 at 9:36 PM Tom Terwilliger < b6116340b489-dmarc-requ...@jiscmail.ac.uk> wrote: > Hi Structural biologist colleagues! > > Our article that helps you make the case that the experiment

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Hi Mark, responsible people are resigning from X. Cheers, Tim Am 01.12.2023 23:24, schrieb Mark J. van Raaij: just came across this critique of that paper on Twitter: This exciting paper shows AI design of materials, robotic synthesis. 10s of new compounds in 17 days. But did they? This paper h

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

just came across this critique of that paper on Twitter: https://twitter.com/Robert_Palgrave/status/1730358675523424344 Robert Palgrave (@Robert_Palgrave) on X twitter.com https://twitter.com/Robert_Palgrave/status/1730358675523424344but I'm not enough of an expert to judge - perhaps some char

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Adding to that literature list a bit outside : Merchant, A., Batzner, S., Schoenholz, S.S. et al. Quote: "... we show that graph networks trained at scale can reach unprecedented levels of generalization, improving the efficiency of materials discovery by an order of magnitude. " Scaling dee

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Hi Tom, Great to see your new analysis and publication. A wonderful addition to your recent work in this domain/comparisons. Congratulations! For those trying to keep up with "AI Drug Discovery" (including AlphaFold, OpenFold, Gen AI, etc) vs "Conventional Drug Discovery" (primarily crystallograp

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

The Baker group has also reported substantial progress in all atom predictions Krishna et al., Generalized Biomolecular Modeling and Design with RoseTTAFold All-Atom bioRxiv 2023.10.09.561603 doi: 10.1101/2023.10.09.561603 https://www.biorxiv.org/content/10.1101/2023.10.09.561603v1.full Regards,

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Very very interesting analysis - Thankyou! On Fri, 1 Dec 2023 at 18:01, Tom Terwilliger < b6116340b489-dmarc-requ...@jiscmail.ac.uk> wrote: > Hi Roberto, > Thanks for the link to the DeepMind site! I hadn't seen that and now I > read it and the paper that they link to. It looks to me as tho

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Hi Roberto, Thanks for the link to the DeepMind site! I hadn't seen that and now I read it and the paper that they link to. It looks to me as though they are making significant progress on ligands, dna, rna, but that the program is not yet available, so I don't think it can be tested yet! All the

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Based however on an AlphaFold version that I guess will be at some point released. I was wondering if someone had an early access…. Oh well…I guess we will find out how it performs on ligands soon enough. Cheers Roberto Roberto A Steiner www.steinerlab.org https://tw

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

That's a press release, not a tool that can be tested. On 01/12/2023 12:19, Roberto Steiner wrote: Great paper indeed! However quite significant progress seems to have been achieved with ligands as well (and not only)… https://deepmind.google/discover/blog/a-glimpse-of-the-next-generation-of

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Great paper indeed! However quite significant progress seems to have been achieved with ligands as well (and not only)… https://deepmind.google/discover/blog/a-glimpse-of-the-next-generation-of-alphafold/ Has anyone in the community put this to the test? Best wishes Roberto Roberto A Steine

Re: [ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Wonderful paper, well done! On 30/11/2023 21:35, Tom Terwilliger wrote: Hi Structural biologist colleagues! Our article that helps you make the case that the experiment is still very much needed is now out: https://www.nature.com/articles/s41592-023-02087-4 "AlphaFold predictions are valuab

[ccp4bb] The experiment is still very much needed (though AlphaFold helps a lot)

Hi Structural biologist colleagues! Our article that helps you make the case that the experiment is still very much needed is now out: https://www.nature.com/articles/s41592-023-02087-4 "AlphaFold predictions are valuable hypotheses and accelerate but do not replace experimental structure determ