Lasse Middendorf
@LasseMiddendorf
PhD Student at @CRGenomica | Synthetic Biology, Proteins & Evolution 🧬 | @ferruz_noelia lab | @igem_muenster 🐝
What structures do de novo proteins have? What are their functions? How does their sequence space localization aid in understanding their properties? Find out in the 2nd preprint from my master’s thesis together with @lacholt & @bharatravi654 🧵1/10 biorxiv.org/content/10.110…
Our PICNIC tool predicts condensate-proteins regardless of their structural disorder across organisms @NatureComms. Congratulations to @AHadarovich for developing the algorithm and to Hari @HymanLab for the experiments! @mpicbg @csbdresden @PoLDresden nature.com/articles/s4146…
🌟 Proteins are the unsung heroes of life! 🌟 Excited to announce my new blog, Decoding Proteins! From enzymes breaking down plastics to breakthroughs in medicine, discover how proteins are shaping our world. 🔗 Check it out: shorturl.at/fMJ38 #Proteins #Biotechnology
After a long sleep, environmental remediation with live biologics is back to the mainstream R&D agenda at this side of the pond. Wanna join an EU-wide on line conversation with bioremediation stakeholders this Nov 10? Express interest and register here 👉 forms.cloud.microsoft/e/b7xVz7dJcu
Wohooo excited to be participating, last time this tournament literally changed the way we think about pLMs- I encourage everyone to participate, we have nothing to lose and loads to learn!!
1/4 🚀 Announcing the 2025 Protein Engineering Tournament. This year’s challenge: design PETase enzymes, which degrade the type of plastic in bottles. Can AI-guided protein design help solve the climate crisis? Let’s find out! ⬇️ #AIforBiology #ClimateTech #ProteinEngineering…
PCR, restriction cloning, Golden Gate. So many core DNA techniques started with a natural enzyme then figured out a use for it Has anyone tried to reverse engineer this? Is there a wishlist of plausible DNA enzyme activities that, if they existed, would be the next CRISPR?
I'm offering free copies of the @AsimovPress book to iGEM students, or college students enrolled in biotechnology courses, in the U.S. If you'd like copies for your team/class, DM me. I'll send out 1 book per student & 1 DNA capsule per shipment. While (limited) supplies last.
Welcoming @LasseMiddendorf as Mayor of iGEM’s Biomanufacturing Village. This Village explores how we can grow the medicines, materials and commodities of the future using synthetic biology- powered by shared tools, scalable ideas and practical roadmaps. villages.igem.org
SYNEVO: Towards Synthetic Evolution of Biomolecules via Aligning Protein Language Models to Biological Hardware 1.SYNEVO is a closed-loop AI-driven framework that combines protein language models (pLMs), high-throughput experimental feedback, and reinforcement learning (RL) to…
DeNoFo: a file format and toolkit for standardised, comparable de novo gene annotation biorxiv.org/content/10.110… #biorxiv_bioinfo
It's long seemed that molecular biology is a natural home for ML interpretability research, given the maturity of human-constructed models of biological mechanisms--permitting direct comparison with their ML-derived counterparts--unlike vision and NLP. Our first foray below👇.
Can we learn protein biology from a language model? In new work led by @liambai21 and me, we explore how sparse autoencoders can help us understand biology—going from mechanistic interpretability to mechanistic biology.
Paper from Matt and Dana (thanks to all other authors) published today nature.com/articles/s4225… Has been available via preprint - a few key points (I'm gonna run out of space...) in next tweet
How can we improve protein representation learning by leveraging evolutionary information from ancestral sequences? @ANUmedia @NatMachIntell "Leveraging ancestral sequence reconstruction for protein representation learning" • Protein language models (PLMs) that convert amino…
Protein language models excel at generating natural-like proteins but struggle to sample from rare, high-value regions, like highly active enzymes. We introduce DPO_pLM, an RL framework to iteratively optimize protein properties! See Filippo’s thread and arxiv.org/abs/2412.12979
Watch your protein parameters go brrr 🚀 Check out our preprint on applying reinforcement learning (RL) to optimize Protein Language Models: arxiv.org/abs/2412.12979 With @mariartigues, Andi, @talaldotpdb, @marcguellc and @ferruz_noelia
Are you at the NeurIPS? Come see my talk tomorrow at the MLSB workshop at 8:30! I'll talk about how we're using reinforcement learning to guide protein language models 🚦🛤️📈 Our preprint seems stuck in the screening process; see our repo for now!: github.com/AI4PDLab/DPO_p…
#Protein People 🧬✨: What is your favourite protein and what makes it so special to you? Bonus points for designed proteins! I’ve got an exciting project brewing that I can’t wait to reveal soon! Drop your answers below and stay tuned 👀🎉
Next up is Andrea Barrera! Presenting a fascinating talk which leaves us wondering, can metabolic intervention affect genome stability? #CRGPhDSymp2024
Exciting talk by Nikola Todorov on how diet-induced sperm RNAs shape offspring metabolism! Fascinating insights into epigenetics and inheritance. #CRGPhDSymp2024
A highly relevant talk from Nico Lorenzon describing novel diagnostic tools to assess treatment resistance in Mayor Depressive Disorder #CRGPhDSymp2024
Highlighting the amazing research at the #CRGPhDSymp2024 Poster Session! Check out some snapshots from today’s sessions


Kicking us off today is @TPerteghella from @GuigoLab with her talk on the latest release of #GENCODE, dramatically increasing the number of annotated lncRNAs. Well done to everyone involved in the consortium! #CRGPhDSymp2024