Felix Koulischer
@felixkoulischer.bsky.social
PhD working on Diffusion Models and all their applications, from images to proteins! Passions include theoretical physics, birds and orgiamis
Reposted by Felix Koulischer
Thrilled to announce our new preprint, “Protein Hunter: Exploiting Structure Hallucination within Diffusion for Protein Design,” in collaboration with @Griffin, @GBhardwaj8 and @sokrypton.org
🧬Code and notebooks will be released by the end of this week.
🎧Golden- Kpop Demon Hunters
🧬Code and notebooks will be released by the end of this week.
🎧Golden- Kpop Demon Hunters
October 13, 2025 at 3:45 PM
Thrilled to announce our new preprint, “Protein Hunter: Exploiting Structure Hallucination within Diffusion for Protein Design,” in collaboration with @Griffin, @GBhardwaj8 and @sokrypton.org
🧬Code and notebooks will be released by the end of this week.
🎧Golden- Kpop Demon Hunters
🧬Code and notebooks will be released by the end of this week.
🎧Golden- Kpop Demon Hunters
Reposted by Felix Koulischer
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
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!
Reposted by Felix Koulischer
The developers of OpenFold3 have released an early version of the tool, which they hope will one day perform on par with DeepMind’s protein-structure model
go.nature.com/4qvCQly
go.nature.com/4qvCQly
Open-source protein structure AI aims to match AlphaFold
The developers of OpenFold3 have released an early version of the tool, which they hope will one day perform on par with DeepMind’s protein-structure model.
go.nature.com
October 29, 2025 at 10:01 AM
The developers of OpenFold3 have released an early version of the tool, which they hope will one day perform on par with DeepMind’s protein-structure model
go.nature.com/4qvCQly
go.nature.com/4qvCQly
🍾 Celebration time!! 🍾
Last week, I obtained an FWO grant on the topic of: "Physics-Guided Protein Diffusion Models", on which I'll work until November 2029!! 🤗
Looking forward to all collaborations, projects and discoveries that are waiting ahead!🐳
Last week, I obtained an FWO grant on the topic of: "Physics-Guided Protein Diffusion Models", on which I'll work until November 2029!! 🤗
Looking forward to all collaborations, projects and discoveries that are waiting ahead!🐳
October 8, 2025 at 7:11 AM
🍾 Celebration time!! 🍾
Last week, I obtained an FWO grant on the topic of: "Physics-Guided Protein Diffusion Models", on which I'll work until November 2029!! 🤗
Looking forward to all collaborations, projects and discoveries that are waiting ahead!🐳
Last week, I obtained an FWO grant on the topic of: "Physics-Guided Protein Diffusion Models", on which I'll work until November 2029!! 🤗
Looking forward to all collaborations, projects and discoveries that are waiting ahead!🐳
Many people are complaining on the review process at NeurIPS this year. Getting good scores and a reject is definitely frustrating, however the score/decision distribution for the main track looks normal; ACs should have the power to (marginally) question reviewers claims.
October 1, 2025 at 1:26 PM
Many people are complaining on the review process at NeurIPS this year. Getting good scores and a reject is definitely frustrating, however the score/decision distribution for the main track looks normal; ACs should have the power to (marginally) question reviewers claims.
🌟 Our paper regarding "Feedback Guidance of Diffusion Models" has been accepted at NeurIPS25!!! 🌟
Very grateful to everyone who contributed! 🙌
link: arxiv.org/abs/2506.06085
Very grateful to everyone who contributed! 🙌
link: arxiv.org/abs/2506.06085
October 1, 2025 at 8:33 AM
🌟 Our paper regarding "Feedback Guidance of Diffusion Models" has been accepted at NeurIPS25!!! 🌟
Very grateful to everyone who contributed! 🙌
link: arxiv.org/abs/2506.06085
Very grateful to everyone who contributed! 🙌
link: arxiv.org/abs/2506.06085
Reposted by Felix Koulischer
La-Proteina from NVidia to do co-design of fully atomistic protein structures (sequence + side-chains + backbone) for up to 800 residues is now open source with permissive licensing
🖥️ github.com/NVIDIA-Digit...
🗎 arxiv.org/abs/2507.09466
🖥️ github.com/NVIDIA-Digit...
🗎 arxiv.org/abs/2507.09466
September 11, 2025 at 8:23 AM
La-Proteina from NVidia to do co-design of fully atomistic protein structures (sequence + side-chains + backbone) for up to 800 residues is now open source with permissive licensing
🖥️ github.com/NVIDIA-Digit...
🗎 arxiv.org/abs/2507.09466
🖥️ github.com/NVIDIA-Digit...
🗎 arxiv.org/abs/2507.09466
Reposted by Felix Koulischer
Nature research paper: One-shot design of functional protein binders with BindCraft
go.nature.com/46bhOAv
go.nature.com/46bhOAv
One-shot design of functional protein binders with BindCraft - Nature
BindCraft, an open-source, automated pipeline for de novo protein binder design with experimental success rates of 10–100%, leverages AlphaFold2 weights to generate binders with nanomolar affinity without the need for high-throughput screening.
go.nature.com
September 1, 2025 at 5:37 PM
Nature research paper: One-shot design of functional protein binders with BindCraft
go.nature.com/46bhOAv
go.nature.com/46bhOAv
🌟 Very happy to share our most recent work titled: 🌟
"Feedback Guidance of Diffusion Models"
Before applying guidance check whether you're on the right track, and only if you're not increase the guidance scale! ✔️
It seems DMs are not so different from us: Self-reflection is key! 🙄
"Feedback Guidance of Diffusion Models"
Before applying guidance check whether you're on the right track, and only if you're not increase the guidance scale! ✔️
It seems DMs are not so different from us: Self-reflection is key! 🙄
July 31, 2025 at 1:43 PM
🌟 Very happy to share our most recent work titled: 🌟
"Feedback Guidance of Diffusion Models"
Before applying guidance check whether you're on the right track, and only if you're not increase the guidance scale! ✔️
It seems DMs are not so different from us: Self-reflection is key! 🙄
"Feedback Guidance of Diffusion Models"
Before applying guidance check whether you're on the right track, and only if you're not increase the guidance scale! ✔️
It seems DMs are not so different from us: Self-reflection is key! 🙄