AlphaFold Unofficial
@alphafold.bsky.social
Unofficial account exploring the intersection of biology, molecules, science, AI and protein folding with AlphaFold.
Pinned
By far the best scientific community on the planet. Feels great leaving Mordor for Bluer Skies! #science
Reposted by AlphaFold Unofficial
You're getting mad at straw men. I follow tons of anti-GenAI people and I haven't seen any of them complain about AI being used for things like Alphafold. People hate GenAI because it stole the works of thousands of artists/writers and is being used to drive those same creatives out of work
December 21, 2025 at 7:29 PM
You're getting mad at straw men. I follow tons of anti-GenAI people and I haven't seen any of them complain about AI being used for things like Alphafold. People hate GenAI because it stole the works of thousands of artists/writers and is being used to drive those same creatives out of work
Reposted by AlphaFold Unofficial
You ignore the progress that is being made in science by non scientists with AI. Alphafold, and material sciences use AI with a lot of success. It's not an Oracle that replaces your work. It can be a great enhancement, and produce slop just like most humans ,😄. Yes its here and will change stuff.
December 21, 2025 at 6:11 PM
You ignore the progress that is being made in science by non scientists with AI. Alphafold, and material sciences use AI with a lot of success. It's not an Oracle that replaces your work. It can be a great enhancement, and produce slop just like most humans ,😄. Yes its here and will change stuff.
Reposted by AlphaFold Unofficial
Can we please please please start articles about AI with a disambiguation about what kind exactly we are talking about?
AlphaFold is very much different from generative AI, both in the underlying technology and in domain it is used.
AlphaFold is very much different from generative AI, both in the underlying technology and in domain it is used.
December 21, 2025 at 3:32 AM
Can we please please please start articles about AI with a disambiguation about what kind exactly we are talking about?
AlphaFold is very much different from generative AI, both in the underlying technology and in domain it is used.
AlphaFold is very much different from generative AI, both in the underlying technology and in domain it is used.
Reposted by AlphaFold Unofficial
Alphafold, the 2020-era Google AI innovation, has owned protein structure, and AF3, its newest iteration, predicts interactions with nucleic acids DNA, RNA, and it increasingly docks small molecules (drugs).
THAT is public interest, unlike 99% of what AI does.
THAT is public interest, unlike 99% of what AI does.
December 20, 2025 at 8:01 PM
Alphafold, the 2020-era Google AI innovation, has owned protein structure, and AF3, its newest iteration, predicts interactions with nucleic acids DNA, RNA, and it increasingly docks small molecules (drugs).
THAT is public interest, unlike 99% of what AI does.
THAT is public interest, unlike 99% of what AI does.
Reposted by AlphaFold Unofficial
December 20, 2025 at 12:08 AM
Scientists can never fully be replaced by machines. AI can never replace nature, the brain and the love-felt chambers of the human heart.
December 20, 2025 at 7:55 PM
Scientists can never fully be replaced by machines. AI can never replace nature, the brain and the love-felt chambers of the human heart.
Reposted by AlphaFold Unofficial
Using sequence alignment and AlphaFold modeling, the Carman lab at Rutgers identified four conserved motifs in yeast Pah1 essential for catalytic activity and lipid balance. go.njmicrobe.org/lmCbZg #Microbiology #YeastResearch #LipidMetabolism #Pah1 #AlphaFold #Lipin #MajeedSyndrome #Rhabdomyolysis
Deciphering the active site: Why motif conservation in Pah1 matters for yeast lipid homeostasis
The enzyme Pah1 is a central figure in the lipid metabolism of Saccharomyces cerevisiae, acting as a Mg2+-dependent phosphatidate (PA) phosphatase that converts PA into diacylglycerol. This catalytic ...
go.njmicrobe.org
December 20, 2025 at 3:34 PM
Using sequence alignment and AlphaFold modeling, the Carman lab at Rutgers identified four conserved motifs in yeast Pah1 essential for catalytic activity and lipid balance. go.njmicrobe.org/lmCbZg #Microbiology #YeastResearch #LipidMetabolism #Pah1 #AlphaFold #Lipin #MajeedSyndrome #Rhabdomyolysis
Reposted by AlphaFold Unofficial
How Alphafold Has Changed Biology Research, 5 Years On
NPR science Friday podcast
@scifri.bsky.social
Very educational 🙏👏
NPR science Friday podcast
@scifri.bsky.social
Very educational 🙏👏
December 20, 2025 at 12:05 AM
How Alphafold Has Changed Biology Research, 5 Years On
NPR science Friday podcast
@scifri.bsky.social
Very educational 🙏👏
NPR science Friday podcast
@scifri.bsky.social
Very educational 🙏👏
Reposted by AlphaFold Unofficial
The ONLY ‘AI’ I will allow are specialist models, like Alphafold. Things that are actively working to improve lives. Not taking jobs or profiting off of others’ work. Not GenAI or LLMs.
December 19, 2025 at 10:28 PM
The ONLY ‘AI’ I will allow are specialist models, like Alphafold. Things that are actively working to improve lives. Not taking jobs or profiting off of others’ work. Not GenAI or LLMs.
Reposted by AlphaFold Unofficial
AlphaFold was of course a version of that too, even if it couldn't score well here. But it wasn't general. These capabilities appear to be general intelligence, e.g. according to Francois Chollet's well considered framing here arcprize.org/arc-agi
ARC Prize - What is ARC-AGI?
The only AI benchmark that measures AGI progress.
arcprize.org
December 19, 2025 at 9:02 PM
AlphaFold was of course a version of that too, even if it couldn't score well here. But it wasn't general. These capabilities appear to be general intelligence, e.g. according to Francois Chollet's well considered framing here arcprize.org/arc-agi
Reposted by AlphaFold Unofficial
Join us for our first #webinar of 2026, introducing EMBL-EBI data resources.
We will be joined by five resources representatives, and you will have the opportunity to pre-submit questions as well as ask them live.
Registration is free but essential: www.ebi.ac.uk/training/eve...
🖥️🧬
We will be joined by five resources representatives, and you will have the opportunity to pre-submit questions as well as ask them live.
Registration is free but essential: www.ebi.ac.uk/training/eve...
🖥️🧬
December 19, 2025 at 10:39 AM
Join us for our first #webinar of 2026, introducing EMBL-EBI data resources.
We will be joined by five resources representatives, and you will have the opportunity to pre-submit questions as well as ask them live.
Registration is free but essential: www.ebi.ac.uk/training/eve...
🖥️🧬
We will be joined by five resources representatives, and you will have the opportunity to pre-submit questions as well as ask them live.
Registration is free but essential: www.ebi.ac.uk/training/eve...
🖥️🧬
Reposted by AlphaFold Unofficial
In search of the ER guardian of oxygen galaxy: how AlphaFold and discovery of a conserved heme-binding protein can help
CIHD Virtual Seminar December 16th, 2025 - Dengke Ma, Ph.D.
University of Utah CIHD
youtu.be/wsj46I6Im9Y?...
CIHD Virtual Seminar December 16th, 2025 - Dengke Ma, Ph.D.
University of Utah CIHD
youtu.be/wsj46I6Im9Y?...
CIHD Virtual Seminar December 16th, 2025 - Dengke Ma, Ph.D.
YouTube video by University of Utah CIHD
youtu.be
December 19, 2025 at 6:04 AM
In search of the ER guardian of oxygen galaxy: how AlphaFold and discovery of a conserved heme-binding protein can help
CIHD Virtual Seminar December 16th, 2025 - Dengke Ma, Ph.D.
University of Utah CIHD
youtu.be/wsj46I6Im9Y?...
CIHD Virtual Seminar December 16th, 2025 - Dengke Ma, Ph.D.
University of Utah CIHD
youtu.be/wsj46I6Im9Y?...
Reposted by AlphaFold Unofficial
LLMs are ML too. This is built on the Google DeepMind GraphCast foundation model. Yes, the same DeepMind that built the Gemini LLM, and won a Nobel Prize last year for AlphaFold.
December 18, 2025 at 6:35 AM
LLMs are ML too. This is built on the Google DeepMind GraphCast foundation model. Yes, the same DeepMind that built the Gemini LLM, and won a Nobel Prize last year for AlphaFold.
Reposted by AlphaFold Unofficial
The strongest argument is that this is just what people who actually do drug development say.
www.prospectmagazine.co.uk/ideas/techno...
www.prospectmagazine.co.uk/ideas/techno...
DeepMind has predicted the shape of every protein known to science. How excited should we be?
The AI-based breakthrough, like the sequencing of the human genome, may not turbocharge the development of new drugs as some hope. But as a resource f...
www.prospectmagazine.co.uk
December 18, 2025 at 2:32 AM
Reposted by AlphaFold Unofficial
AlphaFold is a good example of what AI is actually good for. the creators of it won a nobel prize a few years ago, massive leap in medical technology. i know AI is also used for a lot of advanced simulations in astrophysics and stuff but i forget the specific details on that.
December 17, 2025 at 5:41 PM
AlphaFold is a good example of what AI is actually good for. the creators of it won a nobel prize a few years ago, massive leap in medical technology. i know AI is also used for a lot of advanced simulations in astrophysics and stuff but i forget the specific details on that.
Reposted by AlphaFold Unofficial
But why should we anticipate this? There is no evidence for it so far. They cite, among other things, AlphaFold. AF is very useful, but has not transformed drug development and will not. /2
December 17, 2025 at 1:10 PM
But why should we anticipate this? There is no evidence for it so far. They cite, among other things, AlphaFold. AF is very useful, but has not transformed drug development and will not. /2
Reposted by AlphaFold Unofficial
Pragya Parashara, Teresa Gunn, Jennifer Kong @jennkong.bsky.social and colleagues discover that the E3 ubiquitin ligase MGRN1 targets melanocortin receptors MC1R and MC4R via interactions with transmembrane adapters.
journals.biologists.com/jcs/article/...
journals.biologists.com/jcs/article/...
December 17, 2025 at 9:41 AM
Pragya Parashara, Teresa Gunn, Jennifer Kong @jennkong.bsky.social and colleagues discover that the E3 ubiquitin ligase MGRN1 targets melanocortin receptors MC1R and MC4R via interactions with transmembrane adapters.
journals.biologists.com/jcs/article/...
journals.biologists.com/jcs/article/...
Reposted by AlphaFold Unofficial
Is anyone else having Alphafold server jobs fail a lot lately ("stuck in PENDING for over 24 hours" or other error) and seeing the message "Current estimated wait time for new jobs submitted is up to 6 hours" on the submission page? Wondering if it's just me or a general issue.
December 17, 2025 at 12:21 AM
Is anyone else having Alphafold server jobs fail a lot lately ("stuck in PENDING for over 24 hours" or other error) and seeing the message "Current estimated wait time for new jobs submitted is up to 6 hours" on the submission page? Wondering if it's just me or a general issue.
Reposted by AlphaFold Unofficial
AlphaFold is a diffusion model that works the same way AI image generators work.
December 16, 2025 at 9:03 PM
AlphaFold is a diffusion model that works the same way AI image generators work.
Reposted by AlphaFold Unofficial
Right? I found the first example really off putting. The author gives credit to AlphaFold for doing the hard work of winning a Nobel. Yes, it's important tech, but to then down play the work of the people that developed and continue to run, check, and interpret the work is messed up.
December 16, 2025 at 8:08 PM
Right? I found the first example really off putting. The author gives credit to AlphaFold for doing the hard work of winning a Nobel. Yes, it's important tech, but to then down play the work of the people that developed and continue to run, check, and interpret the work is messed up.
Reposted by AlphaFold Unofficial
We started out with AlphaFold predictions and #cryoEM structures that show the interactions between the endonuclease XPF-ERCC1, the scaffold protein SLX4, and the DNA repair factor SLX4IP. This allowed us to pinpoint and test the molecular interactions that govern complex assembly.
December 16, 2025 at 4:33 PM
We started out with AlphaFold predictions and #cryoEM structures that show the interactions between the endonuclease XPF-ERCC1, the scaffold protein SLX4, and the DNA repair factor SLX4IP. This allowed us to pinpoint and test the molecular interactions that govern complex assembly.
Reposted by AlphaFold Unofficial
AlphaFold is one of those things that is both a genuine scientific triumph and also massively over touted for its abilities, if you talk to any experimental structural biologist they will be able to tell you where it's useful and where it's not, but not enough people are asking them!
December 16, 2025 at 4:19 PM
AlphaFold is one of those things that is both a genuine scientific triumph and also massively over touted for its abilities, if you talk to any experimental structural biologist they will be able to tell you where it's useful and where it's not, but not enough people are asking them!
Reposted by AlphaFold Unofficial
One thing about AlphaFold (and presumably all of these, alphafold is just the one i know) is that it actually has a lot of limitations! Even if it was perfect, it takes actual scientists to come up with the ideas for how to use it in the first place!
None of the examples in this article are “restarting the world’s idea machine.” They are AI tools helping scientists automate rote or routine tasks.
December 16, 2025 at 4:17 PM
One thing about AlphaFold (and presumably all of these, alphafold is just the one i know) is that it actually has a lot of limitations! Even if it was perfect, it takes actual scientists to come up with the ideas for how to use it in the first place!
Reposted by AlphaFold Unofficial
You’ve got a nanobody that binds a small molecule, but you’re not quite sure about its structure or its binding site? Can you trust AlphaFold 3? Or is it better to use Boltz-2 or Chai-1? Does it even matter if your nanobody sequence is similar to ones used during the training of these tools?
December 16, 2025 at 3:39 PM
You’ve got a nanobody that binds a small molecule, but you’re not quite sure about its structure or its binding site? Can you trust AlphaFold 3? Or is it better to use Boltz-2 or Chai-1? Does it even matter if your nanobody sequence is similar to ones used during the training of these tools?
Reposted by AlphaFold Unofficial
The premise also seems to be “The X-Ray machine itself deserved to receive its own Nobel Prize alongside Wilhelm Röntgen”
December 16, 2025 at 12:26 PM
The premise also seems to be “The X-Ray machine itself deserved to receive its own Nobel Prize alongside Wilhelm Röntgen”