Gabriele Corso
gcorso.bsky.social
Gabriele Corso
@gcorso.bsky.social
PhD student @MIT • Research on Generative Models for Biophysics and Drug Discovery
Reposted by Gabriele Corso
Join us today at MIT from 5pm-6pm for a seminar for BoltzGen, which is free and open to the public! Looking forward to seeing you all later today! 🤩 If you plan to attend, don't forget to RSVP so we can plan accordingly: luma.com/7474iho2
#AISky #AcademicSky
October 30, 2025 at 2:36 PM
Reposted by Gabriele Corso
One cool thing about BoltzGen is the flexibility of the conditioning information you can provide (e.g. "constrain the structure of these residues and also these but let their relative placement vary"). I'm really curious to see what the community does with it! github.com/HannesStark/...
GitHub - HannesStark/boltzgen
Contribute to HannesStark/boltzgen development by creating an account on GitHub.
github.com
October 27, 2025 at 2:38 PM
Reposted by Gabriele Corso
Very happy @roman-bushuiev.bsky.social and I joined the amazing team led by @hannes-stark.bsky.social to work on BoltzGen, a generative model for binder design based on Boltz-2. Excited what it will enable!
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project is collaborating with a broad network of leading wetlabs that test BoltzGen at an unprecedented scale, showing success on many novel targets and pushing the model to its limits!
October 27, 2025 at 12:35 PM
Reposted by Gabriele Corso
Thrilled to contribute to this exciting work on protein binder design together w/ @hannes-stark.bsky.social
, Felix Faltings, Regina Barzilay, Tommi Jaakkola, and co. We applied BoltzGen to design novel antimicrobial peptides targeting DNA gyrase based on inhibitory protein fragments. (1/n)
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project is collaborating with a broad network of leading wetlabs that test BoltzGen at an unprecedented scale, showing success on many novel targets and pushing the model to its limits!
October 27, 2025 at 12:39 PM
Reposted by Gabriele Corso
I’ve been testing BoltzGen a bit recently and while I haven’t done any experimental testing yet, the quality of the software is very clear. It installs, runs, logs everything, has tons of options. Very excited to test out the designs irl!
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project is collaborating with a broad network of leading wetlabs that test BoltzGen at an unprecedented scale, showing success on many novel targets and pushing the model to its limits!
October 27, 2025 at 1:57 PM
Reposted by Gabriele Corso
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project is collaborating with a broad network of leading wetlabs that test BoltzGen at an unprecedented scale, showing success on many novel targets and pushing the model to its limits!
October 26, 2025 at 10:40 PM
Pretty surreal when out of the blue you see your model up on an ad on highway billboards and airport screens 🤯 (pictures from San Diego Airport and a highway in SF)
September 17, 2025 at 9:59 AM
Boltz v2.2.1 out. A few improvements including support for .pdb templates, better treatment of stereochemistry in guidance potentials, and improved documentation. As always a great thank you to all those in the community who contributed via PR, raising issues or directly reporting issues to us.
September 9, 2025 at 1:31 PM
Sitting at the #AITHYRA symposium hearing about incredible new high-throughput datasets 🤯! If you have developed new data that you think could improve Boltz, e.g. protein small-molecule affinity, protein protein affinity, binding site (eg via proteomics), let's work together! 🤗
September 9, 2025 at 9:36 AM
Reposted by Gabriele Corso
Here's how to predict binding of tens or hundreds of small molecules to protein assemblies using Boltz 2 in ChimeraX.
Accuracy depends on how similar the ligands and binding pockets are to existing experimental structures.
www.rbvi.ucsf.edu/chimerax/dat...
August 20, 2025 at 9:01 PM
Reposted by Gabriele Corso
August 21, 2025 at 8:44 AM
Reposted by Gabriele Corso
If you use Boltz1/2, BioEmu, Chai1, or other MSA-dependent models, you’re likely using our ColabFold server. Please be considerate! Avoid large submissions across many IPs instead generate the MSA locally. Our server is an old-timer from 2014 and can’t handle that load.
August 15, 2025 at 5:48 PM
Reposted by Gabriele Corso
ChimeraX daily builds can predict binding affinity of small molecules using Boltz 2 on your Mac, Windows or Linux computer. www.rbvi.ucsf.edu/chimerax/dat...
July 24, 2025 at 2:44 AM
Reposted by Gabriele Corso
Excited to share our work on modeling the human protein interactome in 3D. This currently includes 1,394 interaction structures predicted with Boltz-2. Grateful for the open-access resources making this possible!

Preprint: doi.org/10.1101/2025...
July 4, 2025 at 11:38 AM
Reposted by Gabriele Corso
Boltz-2 : Towards accurate and efficient binding affinity prediction by @gcorso.bsky.social Saro Passaro and Jeremy Wohlwend

youtu.be/iHDauMATkr0?...
Boltz-2: Towards Accurate and Efficient Binding Affinity Prediction
YouTube video by Valence Labs
youtu.be
June 19, 2025 at 7:25 AM
Nice interview of Phil Sharp on the controversial birth of biotech and recombinant DNA with some great analogies to the current discourse on AI and gene editing!

- YouTube www.youtube.com/watch?v=LLij...
- Spotify spotifycreators-web.app.link/e/NNmzqnTjgUb
- Apple podcasts.apple.com/us/podcast/h...
The Nobel that Sparked Biotech - Prof Phil Sharp
YouTube video by Health Horizons
www.youtube.com
June 17, 2025 at 7:32 PM
Thank you everyone for attending the Boltz-2 Boston, San Francisco and Paris events this week! Given the success of the in-person seminars and the many requests, we are organizing a virtual seminar on Tuesday at 12pm ET / 6pm CET! Sign up here: lu.ma/4bpuwbsr
June 14, 2025 at 7:46 PM
Reposted by Gabriele Corso
A new AI model jointly predicts protein structure and binding affinity, enabling rapid, accurate screening of small molecules and accelerating early-stage drug discovery. Boltz-2 is now open source.
Protein-binding affinity model expands role of AI in drug discovery
Understanding how molecules interact is central to biology: from decoding how living organisms function to uncovering disease mechanisms and developing life-saving drugs.
phys.org
June 10, 2025 at 1:47 AM
Reposted by Gabriele Corso
Open science, activated.

Since the release of Boltz-2 last Friday – the new open-source protein structure and protein binding affinity model from MIT & Recursion – we’ve been introducing the model to the broader community and the reception has been terrific. 🧵
June 13, 2025 at 1:13 PM
Reposted by Gabriele Corso
On June 18 at the Molecular Machine Learning Conference (MoML) at MILA, Gabriele Corso, one of the lead researchers behind Boltz-2, will present more on this latest breakthrough in AI drug discovery alongside researchers from Valence.

👉 Learn more: portal.ml4dd.com #MOML2025
Home | MoML 2025
portal.ml4dd.com
June 13, 2025 at 1:13 PM
Reposted by Gabriele Corso
Boltz-2 just dropped: open-source AI that predicts both protein complex folds ✚ binding affinities in one shot 🚀

This is a win for protein AI, but let's not forget MSAs, the bioinformatics backbone many structure models lean on.
June 13, 2025 at 2:04 PM
Reposted by Gabriele Corso
We added Boltz-2 to our protein design package! Under the hood this relies on a JAX translation, which, thanks to @jeremywohlwend.bsky.social and @gcorso.bsky.social ’s clean code, was fairly easy to write. We’ve been getting great results -- and we have yet to explore the most exciting new features
June 12, 2025 at 8:52 PM
Reposted by Gabriele Corso
It's exciting to see our software used as a benchmark for cutting-edge AI methods!
On the standard FEP+ affinity benchmark, whose targets were held out of training, Boltz-2 achieves an average Pearson of 0.62—comparable to OpenFE, a widely adopted open-source FEP pipeline, while being over 1000x faster!
June 13, 2025 at 1:01 AM
Reposted by Gabriele Corso
A team led by Regina Barzilay, a computer science professor at MIT, has launched Boltz-2, an algorithm that unites protein folding and prediction of small-molecule binding affinity in one package. cen.acs.org/pharmaceutic... #chemsky 🧪
MIT and Recursion partner for AI drug discovery tool
Freely available Boltz-2 algorithm can predict small-molecule binding affinities
cen.acs.org
June 10, 2025 at 3:58 PM