Simon Mathis
@SimMat20
PhD @Cambridge_Uni | AI in protein design & engineering | biotech & environmental applications | enzymes | geometric DL | prev. @ETH_physics
Introducing Latent-X — our all-atom frontier AI model for protein binder design. State-of-the-art lab performance, widely accessible via the Latent Labs Platform. Free tier: platform.latentlabs.com Blog: latentlabs.com/latent-x/ Technical report: tinyurl.com/latent-X
Check out La-Proteina, our new model for all-atom structure generation at scale! Was a very fun project to work on with @tomasgeffner and the rest of the team. Two things I am particularly excited about in 🧵 1/n
Presenting La-Proteina! A new model for scalable, all-atom protein design 🧬 Backbone + sequence + side-chains, indexed and unindexed atomistic motif scaffolding, scalable up to 800 residues, and more… A thread 🧵
Many top academic AI researchers are GPU poor. The Sovereign AI Unit wants to change this story. We are seeking ambitious AI research proposals for UK academics aiming to take their research to the next level. Read below to find out more and apply 🧵👇
😍
Really pleased to share what I have been working on for 2 months: 🇬🇧 UK SovAI are today announcing our £8m seed investment into OpenBind - A consortium that will actually make AI for drug discovery great by generating 500k experiment protein-ligand complexes!! Explainer 🧵 (1/n)
Exciting! 🙌
We're backing the 'OpenBind' project, a new AI initiative announced at #LondonTechWeek that will build the world's biggest protein dataset right here in the UK. That means faster drug discoveries, new treatments for diseases like cancer, and cutting the cost of getting…
Sharing a surprising result on AI molecule generation with Transformers: All-atom DiT with minimal molecular inductive bias is on par with SOTA equivariant diffusion (SemlaFlow, a very strong and optimized model) Scaling pure Transformers can alleviate explicit bond prediction!
Following our successful SF event, we (compmotifs.com) are running another hackathon! When? 7-8 June Where? London Who? Builders & researchers from academia & industry What? Develop computational tools to advance the natural sciences Join us: lu.ma/apsqlxlj?utm_s….
Equivariance is dead - long live equivariance? Scaling pure Transformers is the future of molecular generative models? I think its a lot more nuanced -- I tried to give some perspectives in my spotlight talk at #ICLR2025 AI4Mat Workshop last week
RFdiffusion2 generates new enzyme structures just from the most basic descriptions of their geometry. See @woodyahern's thread to see the next generation of de novo enzyme design:
New enzymes can unlock chemistry we never had access to before. Here, we introduce RFdiffusion2 (RFD2), a generative model that makes significant strides in de novo enzyme design. Preprint: biorxiv.org/content/10.110… Code: coming soon Animation credit: x.com/ichaydon (1/n)
Highly recommend a read - @woodyahern is one of the most competent people I had the pleasure to work with, he’s expertise not only in machine learning but also protein design and biochemistry - amazing work, Woody + team!
New enzymes can unlock chemistry we never had access to before. Here, we introduce RFdiffusion2 (RFD2), a generative model that makes significant strides in de novo enzyme design. Preprint: biorxiv.org/content/10.110… Code: coming soon Animation credit: x.com/ichaydon (1/n)
Very Exciting -- Congratulations!
I am delighted to announce that @georg_e_winter will join Aithyra as the Life Science Director. Winter is coming! oeaw.ac.at/en/news/oeaw-g…
Our first attempts at mechanistic interpretability of Transformers from the perspective of network science and graph theory! A wonderful collaboration with superstars @elb4tu, Deepro Choudhury, @pl219_Cambridge as part of the Geometric Deep Learning class at @Cambridge_CL!
We will be hosting the first mzmine Community Meeting at @IOCBPrague. Please register if you wish to learn about the latest developments in mzmine. More info: docs.google.com/document/d/12E…
@Latent_Labs comes out of stealth today with $50M funding. Our goal? To push the frontiers of generative biology, giving partners instant access to tools capable of accelerating drug design. Every biotech or pharma company searching for the best therapeutic molecules understands…
I’m really excited to be attending NeurIPS and presenting our work on efficient fine-tuning of pre-trained diffusion models for SOTA conditional generation. Come chat with us on 12th Dec (Thursday) at 11am! Thread below (🧵) -
We’re looking forward to chatting with folks at NeurIPS this year about this work! Work done with excellent collaborators @DenkerAlexander, Francisco Vargas, @DidiKieran, @SimMat20, @vdutor, @BarbanoRiccardo, @MathieuEmile, @julia_tweeting_, and @pl219_Cambridge!
Very happy news for the lab indeed!
Great news! @tomas_pluskal from @IOCBPrague @CzechAcademy has received an ERC Consolidator Grant #ERCCoG from @ERC_Research for his TerpenCode project, addressing challenges in enzyme development with innovative machine learning applications in biochemistry. Congratulations!…
Thank you @Buttenschoen for featuring pymol-remote on the @OPIGlets blog! blopig.com/blog/2024/11/c…
Excited to announce our new #proteindesign strategy for allosterically controlling the kinetics of protein-protein interactions! Read on for cool applications in cytokine signaling, biosensing, and protein circuits. biorxiv.org/content/10.110…
Also decided to take the plunge and now 🦋ing at harrisbio