@yeastgenome.bsky.socialNew on SGD gene pages: #EC numbers for enzymes, #AlphaFold structure thumbnails for all in the Protein section! Click through to Protein Details for interactive AlphaFold viewer, domains, modification sites, more! #yeast HOG1: www.yeastgenome.org/locus/hog1#p...
@alexis-verger.cpesr.frLet's hope that deepmind is not using an internal filter to check the millions of submitted AF3 #alphafold predictions on their web server to reanalyze all the good ones with an iptm >0.8 for their next discovery 😅😅😅
@microbesinfect.bsky.social#AlphaFold database of predicted structures for nearly every known protein on Earth is being upgraded to better represent some of the least-known and deadliest species: viruses
www.nature.com/articles/d41... #biochemistry #virology #MicroSky
@artificialscience.bsky.socialAlphaFold just released free 3D structures for 2,800+ human-infecting viruses to support pandemic preparedness. Any lab on Earth can download them.
#AlphaFold #PandemicPreparedness #AIforScience #Virology #DeepMind
@uncrownedaddiction.bsky.socialAn open dataset offers AI-predicted 3D protein complexes for 2,800+ viruses, supporting research on outbreaks, diagnostics, treatments and vaccines.
@antiscripthead.bsky.social
@gltch.ioOver 2,800 viral structures—predicted by AI—are now public via AlphaFold. Focus is on viruses that infect humans, aiding pathogen research. 🦠 #AlphaFold
Source: https://phys.org/news/2026-09-alphafold-database-viral-protein-complexes.html
@rasmusjensen.bsky.social📖 Open access via @cellpress.bsky.social: doi.org/10.1016/j.ce...
Big thanks to @embl.org for making this work possible.
#cryoET #StructuralBiology #Microbiology #AlphaFold #EMBL #TeamTomo
🧵(6/7)
@ebi.embl.orgIs the world prepared for a future pandemic?
AI-predicted protein complex structures for 2,800 viruses known to infect humans are now openly available in the #AlphaFold Database, which could improve how we respond to emerging infectious disease outbreaks.
Learn more: www.embl.org/news/science...
@pdbeurope.bsky.socialHow do pLDDT, ipTM, ipSAE, pDockQ2 and LIS fit together?
See how these metrics work together to assess confidence in an AlphaFold DB complex interface.
🎬 Watch the full tutorial: youtu.be/k4yBqddtPyQ
@training.ebi.embl.orgOur AlphaFold use cases webinar series begins on 7 October. Join us to explore the use of AlphaFold-predicted protein structures for drug discovery against Chagas disease.
Registration is free but essential. www.ebi.ac.uk/training/eve...
@pdbeurope.bsky.social #AlphaFold
🧬🖥️📊
@sal-ai.bsky.socialCheck out the latest article in my newsletter: Any Apparatus for Transmitting Messages www.linkedin.com/pulse/any-ap... via @LinkedIn #ArtificialIntelligence #AIRevolution #AIGovernance #HumanCreativity #TechnologyAndSociety #ScientificDiscovery #AlphaFold #AIResearch #HumanAgency #AI #Future
@pdbeurope.bsky.socialA predicted interaction may involve only a small region of a protein complex.
How can we identify confidence in that local interface when the surrounding regions have high PAE?
Our tutorial uses a PAE matrix to explain how LIS can help 👇
@alexis-verger.cpesr.frNice talk, really perfect for teaching
Everything you've always wanted to know about #AlphaFold, iPTM, PAE, pLDDT, and ipSAE but were afraid to ask
@artificialscience.bsky.social5 pharma firms trained an AI on 20,167 private protein structures, without sharing one. Accuracy: 35.6% → 52.1%. Analysis: #AIDrugDiscovery #AlphaFold #FederatedLearning #CompBio #OpenFold3 #AISB-1-Fed #MELLODDY
artificialscience.org/2026/09/fede...
@freegardener.bsky.socialStructure beats sequence: AlphaFold2 reveals "superdark" GPCR-like proteins hiding in the dark proteome. TM184C links GPCR signaling #Proteomics #AlphaFold #GPCR #DarkProteome
https://freegardner.com/sci_scout/structure-reveals-hidden-gpcr-like-proteins-in-dark-proteome.html
@kirillemedvedev.bsky.social💡New paper! Predicted off-targets for Seladelpar and Zanamivir with AlphaFold3 + MD, then tested the top 3. All inactive. For PPARγ, 4 of 5 AF3 models put the drug ~8 Å off - the only correct pose scored lowest confidence. Stability ≠ function.
link.springer.com/article/10.1...
#alphafold #boltz2
@pdbeurope.bsky.socialSame predicted complex, different confidence scores:
ipTM: 0.38
ipSAE: 0.75
Why can these metrics differ? Full breakdown 👇
@mbadonyi.bsky.socialAlphaFold can now predict protein complexes at unprecedented scale. But once we have a predicted complex, the core question of how much we should trust the interface remains. Existing confidence scores are useful, but their meaning is not always obvious.
#alphafold
@madhurirathr.bsky.socialDeepMind AI Projects: From AlphaGo to AlphaGenome
Explore AlphaGo, AlphaFold, AlphaGenome, AlphaEvolve, Gemini, Co-Scientist, GraphCast, and more in this updated project guide.
www.theofficialalphabet.com/deepmind/dee...
#DeepMind #AIResearch #MachineLearning #AlphaFold #Gemini #theofficialaphabet
