Lars Dornfeld

Scatter plot of LDDT as a function of sequence identity for high coverage homology models. The horizontal red line at 40% sequence identity highlights the presence high quality models in the low sequence identity region.

Filtering out homologous structures from the PDB at 40% sequence identity is not enough to create a robust test set. Significant leakage persists at this level, and comparative modeling can still produce high quality models.

🙌

Scaling down protein language modeling with MSA Pairformer Recent efforts in protein language modeling have focused on scaling single-sequence models and their training data, requiring vast compute resources that limit accessibility. Although models that use ...

Excited to share work with
Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org
biorxiv.org/content/10.1...
TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵

Rational engineering of allosteric protein switches by in silico prediction of domain insertion sites - Nature Methods ProDomino is a machine leaning-based method, trained on a semisynthetic domain insertion dataset, to guide the engineering of protein domain recombination.

Our work on rationally engineering allosteric protein switches is now out in Nature Methods: www.nature.com/articles/s41... Thanks a lot to @grunewald.bsky.social and @noahholzleitner.bsky.social for the comprehensive news and views: www.nature.com/articles/s41...

Amazing news! And more than deserved 🙌 looking forward to all your new exciting research - Will be an outstanding lab for PhDs / PostDocs!

Today's #RDKit blog post is a heartfelt plea for clearer communication.
greglandrum.github.io/rdkit-blog/p...

Can an AI model predict perfectly and still have a terrible world model?

What would that even mean?

Our new ICML paper (poster tomorrow!) formalizes these questions.

One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵

A good example of unfolding „cat“alysis

Was an amazing talk @delafuentelab.bsky.social 🙌

BioEmu is out! Grateful to have had the opportunity to work with such an incredible team on this project 🤗

I was involved in this project as an early stage master’s student in protein engineering and it sparked everything that came after 🙌

I’m incredibly grateful I got to work with the amazing supervisor @juliankompa.bsky.social and the rest of the team on such a cool project. Check it out!

Happy to announce that our paper, "ProtMamba: a homology-aware but alignment-free protein state space model", has been published in Bioinformatics! 🎉

doi.org/10.1093/bioi...

We will be hosting a mirror site of the CASP data and workflows at @sib.swiss. Operating a redundant site with independent funding is a good idea for important scientific infrastructure, independent from the current funding situation in the US.

I had the honor of receiving both the Otto Hahn Medal and Award of the May Planck Society. I am looking forward to transforming this chance into an own research group pushing the boundaries of interdisciplinary phage research!
Huge thanks to all colleagues, collaborators, family and friends!

🧪 New preprint: "Generative Modeling of Full-Atom Protein Conformations using Latent Diffusion on Graph Embeddings"

We generate complete, all-atom protein conformational ensembles, capturing both backbone and side-chain dynamics of, e.g., the Dopamine D2 receptor.

arxiv.org/abs/2506.17064

Joe Greener

Excellent opportunity for a postdoc to join @jgreener64.bsky.social’s group at the LMB.
You’ll help develop novel molecular simulation methods to optimise sequences for diverse design tasks in computational protein design.
See: www.nature.com/naturecareer...
Apply by 13 JUL
#PostdocJobs #ScienceJobs

Excited to share my main PhD work is finally out! 🥳

We built a phage-assisted evolution platform to evolve allosteric protein switches (POGO-PANCE) and introduce cumulative, targeted mutations/Indels (RAMPhaGE)!

See how we evolved ultra-strong, light-switchable AraC variants 🌚🌞

For me as well 🥲

An improved model for prediction of de novo designed proteins with diverse geometries https://www.biorxiv.org/content/10.1101/2025.06.02.657515v1

🚀 Excited to release BoltzDesign1!

✨ Now with LogMD-based trajectory visualization.
🔗 Demo: rcsb.ai/ff9c2b1ee8
Feedback & collabs welcome! 🙌

🔗: GitHub: github.com/yehlincho/Bo...
🔗: Colab: colab.research.google.com/github/yehli...
@sokrypton.org @martinpacesa.bsky.social

Guiding Generative Protein Language Models with Reinforcement Learning Autoregressive protein language models (pLMs) have emerged as powerful tools to efficiently design functional proteins with extraordinary diversity, as evidenced by the successful generation of divers...

Two more🙃:
1) arxiv.org/abs/2412.12979
2) arxiv.org/abs/2505.15093

Protein design and optimization for synthetic cells Nature Reviews Bioengineering, Published online: 19 May 2025; doi:10.1038/s44222-025-00318-1Bottom-up synthetic biology might greatly benefit from custom-made proteins as components of new biological systems. This Review discusses various protein optimization and design approaches, and explores how these can contribute to the generation of controllable synthetic cells.

Protein design and optimization for synthetic cells

Enjoy @julianvangerwen.bsky.social!

Adapting ProteinMPNN for antibody design without retraining The neural network ProteinMPNN designs protein sequences capable of folding into predefined tertiary structures and quaternary assemblies. It has become widely used due to its high success rates when ...

"Adapting ProteinMPNN for antibody design without retraining"

Our several-years-old fix to ProteinMPNN's tendency to make weird antibody CDR sequences is finally preprinted

Inverse folding NNs are better at predicting changes to a mutated protein's equilibrium dynamics, relative to WT, than protein folding NNs (BioEmu & AF-Cluster). This was shown on an enzyme that is more enzymatically promiscuous in the open state.

Reading group tomorrow: @json_yim and @woodyahern present "Atom level enzyme active site scaffolding using RFdiffusion2" www.biorxiv.org/content/10.1...

Join on Zoom at 9am PT / 12pm ET / 6pm CEST: portal.valencelabs.com/starklyspeak...

TFW you realize that you built a *bunch* of models using sklearn's RandomForestRegressor using default params, which, of course, means that you didn't actually build random forests. 🤦

Ah well, they don't take that long to build and the real random forests build even faster.

Preprint from the lab🚨
Have you ever engineered proteins to be more stable and were unhappy about your predictor's success rate? We got you covered with BoostMut!
Great work led by @kerlenkorbeld.bsky.social now online at www.biorxiv.org/content/10.1...
A thread 🧵

Modular Engineering of Thermo-Responsive Allosteric Proteins Thermogenetics enables non-invasive spatiotemporal control over protein activity in living cells and tissues, yet its applications have largely been restricted to transcriptional regulation and membra...

www.biorxiv.org/content/10.1...