Crosslinks can sometimes improve AlphaFold3 models. Crosslinking is an experimental method often used as a source of distance restraints in modeling. In AlphaFold3, instead of using restraints, crosslinks can be included as covalent ligands. Thanks to Agnieszka Obarska-Kosinska for the idea.
December 4, 2024 at 9:40 AM
Crosslinks can sometimes improve AlphaFold3 models. Crosslinking is an experimental method often used as a source of distance restraints in modeling. In AlphaFold3, instead of using restraints, crosslinks can be included as covalent ligands. Thanks to Agnieszka Obarska-Kosinska for the idea.
New Experimental Google Colab Notebook now integrates AlphaFold3 with OpenFold3 and py2Dmol:
colab.research.google.com/github/sokry...
(1/3)
colab.research.google.com/github/sokry...
(1/3)
June 15, 2026 at 2:11 PM
New Experimental Google Colab Notebook now integrates AlphaFold3 with OpenFold3 and py2Dmol:
colab.research.google.com/github/sokry...
(1/3)
colab.research.google.com/github/sokry...
(1/3)
📢 New preprint:
Experiment-guided AlphaFold3 resolves accurate protein ensembles.
doi.org/10.1101/2025...
AlphaFold3 is incredible, but has crucial limitations: it typically collapses to a single conformation, ignoring the inherent dynamics of proteins. And it can be wrong. Here's a solution. 🧵👇
Experiment-guided AlphaFold3 resolves accurate protein ensembles.
doi.org/10.1101/2025...
AlphaFold3 is incredible, but has crucial limitations: it typically collapses to a single conformation, ignoring the inherent dynamics of proteins. And it can be wrong. Here's a solution. 🧵👇
October 18, 2025 at 6:59 PM
📢 New preprint:
Experiment-guided AlphaFold3 resolves accurate protein ensembles.
doi.org/10.1101/2025...
AlphaFold3 is incredible, but has crucial limitations: it typically collapses to a single conformation, ignoring the inherent dynamics of proteins. And it can be wrong. Here's a solution. 🧵👇
Experiment-guided AlphaFold3 resolves accurate protein ensembles.
doi.org/10.1101/2025...
AlphaFold3 is incredible, but has crucial limitations: it typically collapses to a single conformation, ignoring the inherent dynamics of proteins. And it can be wrong. Here's a solution. 🧵👇
Team science preprint, exploring the capabilities and limitations of Alphafold3 across different application areas, including protein-RNA, protein-lipid, ubiquitination,TCR and antibody recognition with @ninjani.bsky.social @labvanni.bsky.social @dgfeller.bsky.social
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
Capabilities, specificity gaps and training-data dependence of AlphaFold3 across diverse application areas
Structure prediction models have moved from single proteins to assemblies that include diverse biomolecules and their modifications. AlphaFold3 (AF3) and related models extended structural modelling v...
www.biorxiv.org
July 14, 2026 at 5:37 AM
Team science preprint, exploring the capabilities and limitations of Alphafold3 across different application areas, including protein-RNA, protein-lipid, ubiquitination,TCR and antibody recognition with @ninjani.bsky.social @labvanni.bsky.social @dgfeller.bsky.social
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
New OpenFold3 preview out! (OF3p2)
It closes the gap to AlphaFold3 for most modalities.
Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9
It closes the gap to AlphaFold3 for most modalities.
Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9
March 13, 2026 at 3:00 PM
New OpenFold3 preview out! (OF3p2)
It closes the gap to AlphaFold3 for most modalities.
Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9
It closes the gap to AlphaFold3 for most modalities.
Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9
November 23, 2024 at 10:15 PM
We have started a project trying to predic the interactions/structures of all yeast protein pairs using an AlphaFold pooling approach. We are making the current dataset open and we welcome collaborations.
www.evocellnet.com/2026/03/mapp...
www.evocellnet.com/2026/03/mapp...
Mapping the yeast atructural interactome with AlphaFold3: an open call for collaboration
We are excited to announce the early-stage release of our S. cerevisiae structural interactome mapping project. Using AlphaFold3 (AF3), w...
www.evocellnet.com
March 4, 2026 at 10:36 AM
We have started a project trying to predic the interactions/structures of all yeast protein pairs using an AlphaFold pooling approach. We are making the current dataset open and we welcome collaborations.
www.evocellnet.com/2026/03/mapp...
www.evocellnet.com/2026/03/mapp...
Happy to see our paper out today: Guiding AlphaFold3 with experimental data. We are totally excited about the possibility to leverage the power of AlphaFold3 AND of experiments!
www.nature.com/articles/s41...
It has been a fantastic collaboration with great colleagues in the Bronstein team at ISTA.
www.nature.com/articles/s41...
It has been a fantastic collaboration with great colleagues in the Bronstein team at ISTA.
June 29, 2026 at 12:58 PM
Happy to see our paper out today: Guiding AlphaFold3 with experimental data. We are totally excited about the possibility to leverage the power of AlphaFold3 AND of experiments!
www.nature.com/articles/s41...
It has been a fantastic collaboration with great colleagues in the Bronstein team at ISTA.
www.nature.com/articles/s41...
It has been a fantastic collaboration with great colleagues in the Bronstein team at ISTA.
Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!
November 17, 2024 at 4:20 PM
Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!
Nature research paper: Precise DNA base editing using AlphaFold3-based contact modelling
go.nature.com/4fQq9yo
go.nature.com/4fQq9yo
Precise DNA base editing using AlphaFold3-based contact modelling - Nature
ContactSeek is an AlphaFold3-driven model that can improve the precision of genome-editing tools.
go.nature.com
July 24, 2026 at 4:10 PM
Nature research paper: Precise DNA base editing using AlphaFold3-based contact modelling
go.nature.com/4fQq9yo
go.nature.com/4fQq9yo
Looks like OpenFold3 has been formally released in a public "preview". Not quite on parity with AlphaFold3 on a few benchmarks shown, in particular for antibody interactions. All info on the github link. I am sure we will hear more about this from the developers github.com/aqlaboratory...
GitHub - aqlaboratory/openfold-3: OpenFold3: A fully open source biomolecular structure prediction model based on AlphaFold3
OpenFold3: A fully open source biomolecular structure prediction model based on AlphaFold3 - aqlaboratory/openfold-3
github.com
October 28, 2025 at 9:30 AM
Looks like OpenFold3 has been formally released in a public "preview". Not quite on parity with AlphaFold3 on a few benchmarks shown, in particular for antibody interactions. All info on the github link. I am sure we will hear more about this from the developers github.com/aqlaboratory...
ColabFold 1.6.3 is out! 2.5x faster, pip-installable, ipSAE+pDockQ2 scores. Thanks Choonghwan Lee, Marielle Russo, Gyuri Kim
🐍pip install colabfold[alphafold]
CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2…
🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"
🐍pip install colabfold[alphafold]
CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2…
🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"
September 16, 2026 at 5:14 PM
ColabFold 1.6.3 is out! 2.5x faster, pip-installable, ipSAE+pDockQ2 scores. Thanks Choonghwan Lee, Marielle Russo, Gyuri Kim
🐍pip install colabfold[alphafold]
CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2…
🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"
🐍pip install colabfold[alphafold]
CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2…
🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"
Why did Nature publish AlphaFold3 without its code? #AcademicSky 🧪💻🧬 www.nature.com/articles/d41...
AlphaFold3 — why did Nature publish it without its code?
Criticism of our decision to publish AlphaFold3 raises important questions. We welcome readers’ views. Criticism of our decision to publish AlphaFold3 raises important questions. We welcome readers’ v...
www.nature.com
May 22, 2024 at 2:01 PM
Why did Nature publish AlphaFold3 without its code? #AcademicSky 🧪💻🧬 www.nature.com/articles/d41...
BioRxiV is very slow, so this appeared in Authorea before www.authorea.com/users/436643...
AlphaFold3 at CASP16
The CASP16 experiment provided the first opportunity to benchmark AlphaFold3. In contrast to AlphaFold2, AlphaFold3 can predict the structure of non-protein molecules, and according to the benchmark p...
www.authorea.com
April 16, 2025 at 9:44 AM
BioRxiV is very slow, so this appeared in Authorea before www.authorea.com/users/436643...
AlphaFold3 still tends to overfit regions that AF2 multimer predicts to be disordered. Almost all disordered regions become helices in AF3.
December 15, 2024 at 2:03 AM
AlphaFold3 still tends to overfit regions that AF2 multimer predicts to be disordered. Almost all disordered regions become helices in AF3.
Latest preprint from the lab, many years in the making!
By combining #cryoEM with #AlphaFold3 modelling, we propose that norovirus NS3 forms a transmembrane RNA translocase.
This could have big implications for our understanding of viral replication & assembly (🧵)
www.biorxiv.org/content/10.1...
By combining #cryoEM with #AlphaFold3 modelling, we propose that norovirus NS3 forms a transmembrane RNA translocase.
This could have big implications for our understanding of viral replication & assembly (🧵)
www.biorxiv.org/content/10.1...
June 18, 2025 at 6:32 AM
Latest preprint from the lab, many years in the making!
By combining #cryoEM with #AlphaFold3 modelling, we propose that norovirus NS3 forms a transmembrane RNA translocase.
This could have big implications for our understanding of viral replication & assembly (🧵)
www.biorxiv.org/content/10.1...
By combining #cryoEM with #AlphaFold3 modelling, we propose that norovirus NS3 forms a transmembrane RNA translocase.
This could have big implications for our understanding of viral replication & assembly (🧵)
www.biorxiv.org/content/10.1...
A nifty trick our lab is using to improve structure prediction of viral membrane proteins in #AlphaFold3! 👇
Amazingly, this gave us a high-confidence transmembrane channel for both mouse and human NS3, with oleic acids forming a bilayer-like pattern.
Our group now call this the “fatty acid hack.”
The #alphafold predictions with and without fatty acids shown below 👇
Our group now call this the “fatty acid hack.”
The #alphafold predictions with and without fatty acids shown below 👇
June 18, 2025 at 1:43 PM
A nifty trick our lab is using to improve structure prediction of viral membrane proteins in #AlphaFold3! 👇
Some important restrictions to consider when using #AlphaFold3 weights.
It's out! But note: 'To request access to the AlphaFold 3 model parameters...complete this form. Access will be granted at Google DeepMind’s sole discretion...You may only use AlphaFold 3 model parameters if received directly from Google...subject to these terms of use.' github.com/google-deepm...
GitHub - google-deepmind/alphafold3: AlphaFold 3 inference pipeline.
AlphaFold 3 inference pipeline. Contribute to google-deepmind/alphafold3 development by creating an account on GitHub.
github.com
November 11, 2024 at 7:47 PM
Some important restrictions to consider when using #AlphaFold3 weights.
I added support for Boltz1 so we can calculate ipSAE and other scores for pairwise protein-protein interactions on AF2, AF3, and Boltz1 models. Will add nucleic acids eventually. github.com/DunbrackLab/...
March 8, 2025 at 8:26 AM
I added support for Boltz1 so we can calculate ipSAE and other scores for pairwise protein-protein interactions on AF2, AF3, and Boltz1 models. Will add nucleic acids eventually. github.com/DunbrackLab/...
Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles - @istaresearch.bsky.social @alex-bronstein.bsky.social @paulschanda.bsky.social @sankethvedula.bsky.social www.nature.com/articles/s41...
Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles - Nature Biotechnology
A generative model using AlphaFold3 as prior infers protein conformational ensembles consistent with measured experimental data.
www.nature.com
June 29, 2026 at 3:10 PM
Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles - @istaresearch.bsky.social @alex-bronstein.bsky.social @paulschanda.bsky.social @sankethvedula.bsky.social www.nature.com/articles/s41...
Unexpectedly, @jurgjn.bsky.social found that running Alphafold3 predictions for protein interactions can yield ipTM scores that are more predictive of true interactions when run in pools of proteins instead of pairwise predictions. Presumably, this reflects some sort of "competition effect".
July 22, 2025 at 2:13 PM
Unexpectedly, @jurgjn.bsky.social found that running Alphafold3 predictions for protein interactions can yield ipTM scores that are more predictive of true interactions when run in pools of proteins instead of pairwise predictions. Presumably, this reflects some sort of "competition effect".
Finally, the code for anyone wanting to run alphafold3 locally with openfold3 weights (including convert script) can be found here:
github.com/sokrypton/al...
github.com/sokrypton/al...
GitHub - sokrypton/alphafold3: AlphaFold 3 inference pipeline.
AlphaFold 3 inference pipeline. Contribute to sokrypton/alphafold3 development by creating an account on GitHub.
github.com
June 15, 2026 at 2:11 PM
Finally, the code for anyone wanting to run alphafold3 locally with openfold3 weights (including convert script) can be found here:
github.com/sokrypton/al...
github.com/sokrypton/al...
AI model using AlphaFold3-predicted contact probabilities maps and reduces genome editor off-targets, enabling highly precise DNA base editing with improved specificity.
#NBThighlight www.nature.com/articles/s41...
#NBThighlight www.nature.com/articles/s41...
Precise DNA base editing using AlphaFold3-based contact modelling - Nature
ContactSeek is an AlphaFold3-driven model that can improve the precision of genome-editing tools.
www.nature.com
July 23, 2026 at 9:14 AM
AI model using AlphaFold3-predicted contact probabilities maps and reduces genome editor off-targets, enabling highly precise DNA base editing with improved specificity.
#NBThighlight www.nature.com/articles/s41...
#NBThighlight www.nature.com/articles/s41...
so now deepmind has gone and released alphafold3 code as well. must have been feeling the competition.
source code here: github.com/google-de...
🧪🧬🖥️
#alphafold3 #bioinformatics #ai #ml
source code here: github.com/google-de...
🧪🧬🖥️
#alphafold3 #bioinformatics #ai #ml
GitHub - google-deepmind/alphafold3: AlphaFold 3 inference pipeline.
AlphaFold 3 inference pipeline. Contribute to google-deepmind/alphafold3 development by creating an account on GitHub.
github.com
November 13, 2024 at 3:35 AM
so now deepmind has gone and released alphafold3 code as well. must have been feeling the competition.
source code here: github.com/google-de...
🧪🧬🖥️
#alphafold3 #bioinformatics #ai #ml
source code here: github.com/google-de...
🧪🧬🖥️
#alphafold3 #bioinformatics #ai #ml