←
Yunguan Fu 🧬540 posts
Yunguan Fu 🧬 avatar

Yunguan Fu 🧬

@mathpluscode

MTS @Boltz_bio; Ex. Staff RE @instadeepai; PhD @UCLmedphys; MSc CSML @uclcs; MEng @Polytechnique.

◷ Joined February 2012▧ 5 Photos and videos
Hannes Stark avatar
⟳ Yunguan Fu 🧬 retweeted
Hannes Stark@HannesStaerk·5h

More money for better models for more "50-fold growth in token usage" for more money for better models for more ... 🙃

We’ve raised $110M to bring AI-powered biomolecular design to every scientist! Boltz’s Series A is led by @BessemerVP, following 50-fold growth in molecular token usage from our users since June and partnerships with 6 of the world’s 20 largest pharma companies.

0:16
Fang Wu avatar
⟳ Yunguan Fu 🧬 retweeted
Fang Wu@WUFang40615703·Nov 17

“Wet-lab validated.” AI binder design papers keep reporting impressive hit rates. But does it bind in a second lab? One independent retest: only 29/107 published AI-designed nanobodies showed binding—and 6 of those had high aggregation. Germinal stands out for the wrong reason: just 1 of 12 published binders showed detectable binding on retest—a result that warrants an explanation. How much trust should we put in headline hit rates? Read more on my blog 👇 smiles724.github.io/blog/ai-…

Post attachment
Gabriele Corso avatar
⟳ Yunguan Fu 🧬 retweeted
Gabriele Corso@GabriCorso·Oct 30

Great to see teams using the Boltz API at the Ebolathon successfully designing novel molecules! 140 molecules tested, 11 selective hits in Ebola pseudovirus assays. Exciting to see hackathon ideas turn into experimental results! 🧪 muni.bio/research/hits-from-…

Charlie Harris avatar
⟳ Yunguan Fu 🧬 retweeted
Charlie Harris@charlieharris01·Sep 21

The UK’s most powerful government-built AI supercomputer, Isambard-AI, cost £225m The Botley Road bridge cost roughly the same Apparently our frontier AI capacity and one bridge are comparable national investments

Cars and buses go from Oxford Station down Botley Road for the first time in three years! The project took longer than the construction of The Titanic and saw costs balloon to £237m. Looks good though.

0:16
Martin Pacesa avatar
⟳ Yunguan Fu 🧬 retweeted
Martin Pacesa@MartinPacesa·Sep 21

Target cross reactivity in the Adaptyv competition is a perfect use case for BindCraft2 natively. Try it out here: Overview + modalities: pacesalab.com/bindcraft Code: github.com/PacesaLab/BindCra… BindCraft1 variant by Lennart Nickel and Paul Kittner github.com/martinpacesa/Bind…

We're hosting a protein design competition with @adaptyvbio. We'll be providing up to $1M in Claude credits, and additional funding to experimentally validate 5k designs submitted by the community. All results will be published openly and participants keep ownership of their designs. We've selected five problems that push the boundaries of what is possible with today's capabilities: species cross-reactivity, pH-sensitivity, and peptide-MHC specificity, and difficult targets such as GPCRs. @modal will provide compute credits and @TwistBioscience will provide DNA. Apply here: docs.google.com/forms/d/e/1F…

Edoardo Contente avatar
⟳ Yunguan Fu 🧬 retweeted
Edoardo Contente@edoardocontente·Sep 9

You get the option to select "Do not train on my content" (which should be active already!?) (5/9)

Post attachment
Noelia Ferruz avatar
⟳ Yunguan Fu 🧬 retweeted
Noelia Ferruz@ferruz_noelia·Aug 3

A database of (>300k) BindCraft designs to 8k human targets

The Human Bindome: A Proteome-scale Atlas of Designed Binder Candidates 1. The paper introduces the Human Bindome: 306,146 designed, genetically encodable protein binder candidates (fixed length 90 aa) spanning 8,296 human proteins, each with an explicit sequence, a predicted binder–target complex structure, and in silico confidence metrics. 2. Core motivation: conventional antibody reagents are often expensive, variably characterized, and frequently lack disclosed sequences, contributing to reproducibility issues; the Bindome aims to provide openly specified alternatives at proteome scale. 3. Methodologically, they scale the experimentally benchmarked BindCraft pipeline (reported prior success rates 10–100% depending on target) by adding automated domain-level target selection and an accelerated parallel implementation suitable for proteome-scale execution. 4. Target selection is driven by AlphaFold DB full-length models plus PAE matrices: proteins are segmented into domains via PAE-based clustering, and “targetable domains” are filtered for confident folding, compactness, contact density, acceptable length, and exclusion of membrane-embedded regions. 5. A key engineering detail: designing against cropped domains makes computation cheaper and enables multiple distinct sites per protein, but can create clashes with other parts of the full-length protein; they address this by superimposing each complex onto the full-length model and rejecting sterically clashing designs unless the clash is in low-confidence regions (high PAE / low pLDDT). 6. All candidates pass unchanged BindCraft stringency thresholds, including binder monomer fold confidence (pLDDT > 0.8) and interaction placement confidence (normalized iPAE < 0.35); distributions are not piled up at cutoffs (medians pLDDT 0.92, iPAE 0.23), enabling within-target prioritization. 7. Coverage numbers: the “targetable proteome” is 10,664 of 20,279 proteins; Bindome covers 77.8% of that targetable set and 40.9% of the full human proteome, with ~37 candidates per target protein on average. 8. Structural breadth: mapping domains to CATH superfamilies suggests 78.2% of superfamilies represented in the human target-domain set have at least one binder candidate; uncovered regions are largely outside the current scope (notably integral membrane proteins). 9. Epitope biology: binder epitopes show convergence within domain families (e.g., grooves/pockets in WD40 and bromodomains; broad kinase-domain surfaces), and epitope residues are significantly more evolutionarily conserved than matched surface controls—more consensus across binders corresponds to higher conservation. 10. Functional relevance and perturbation potential: epitopes often overlap annotated functional sites (PTMs, protein–protein interfaces, ligand-binding sites, DNA-binding regions), and 26% of ASD-annotated allosteric targets have at least one binder overlapping an allosteric site; examples include candidates positioned to occlude TP53 DNA-binding or disrupt SOD1 interaction interfaces. 11. Interface properties vs native PPIs: binder interfaces are native-like in mean area and slightly more polar, but show a strong bias toward helix-mediated contacts on the designed binder side; binder folds are often not globally novel, yet interfaces appear largely novel by interface similarity comparisons. 12. Accessibility is treated as a first-class output: a free web interface supports ranking/inspection and bulk downloads; a standardized 3D-Beacons API enables programmatic retrieval; and an MCP server allows natural-language “agentic” querying via LLM tools. 13. ML utility: the dataset is released with leakage-controlled train/val/test splits based on interface clustering (205,553 train; 21,205 val; 21,955 test), designed to support development/evaluation of next-generation binder design models. 💻Code: github.com/wejulius/Bindome 📜Paper: biorxiv.org/content/10.64898… #ProteinDesign #ComputationalBiology #StructuralBioinformatics #AlphaFold #ProteinBinders #Reproducibility #FunctionalGenomics #DrugDiscovery #MachineLearning

Post attachment
Kimi.ai avatar
⟳ Yunguan Fu 🧬 retweeted
Kimi.ai@Kimi_Moonshot·Jul 27

Releasing the model weights and technical report of Kimi K3. Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window. New model architecture: 2.5x the intelligence per unit of compute, not just more params. Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale. Model weights: huggingface.co/moonshotai/Ki… Tech report: github.com/MoonshotAI/Kimi-K… Tech blog: kimi.com/blog/kimi-k3

Post attachment
Charlie Harris avatar
⟳ Yunguan Fu 🧬 retweeted
Charlie Harris@charlieharris01·Jul 2

Excited to announce a major partnership with GSK based around our latest frontier models!!! Two UK-based companies, one landmark AI x Bio agreement. Proud of what this says about the UK’s frontier ecosystem 🇬🇧

Today, we are excited to announce a major partnership with @GSK to deploy the latest Boltz models across GSK’s research organization!

0:15
Zach Ocean avatar
⟳ Yunguan Fu 🧬 retweeted
Zach Ocean@zachocean_·Jun 30

This was a really cool product to work on. Using an agent to set up and launch molecular design campaigns feels like magic I'm especially excited about what AI science agents mean for education. I don't come from a science background, so joining Boltz to work on tools for scientists has meant diving into an environment of new concepts and jargon. Without LLMs, that barrier might have been impenetrable. I'm hopeful that young budding scientists and old guys like me making a career shift will use Claude Science + Boltz to learn about molecular design and experiment with vibe biology. Grateful for the opportunity to work on ambitious initiatives like this one with the great scientists and engineers on the Boltz team.

Excited to share that Boltz is accessible via Claude products, including Claude Science, Anthropic’s new AI workbench for scientists!

Post attachment
Charlie Harris avatar
⟳ Yunguan Fu 🧬 retweeted
Charlie Harris@charlieharris01·Jun 30

Super excited that all the latest Boltz models are integrated into Claude Science!!! Make sure to enable the connector when you install

Post attachment

Excited to share that Boltz is accessible via Claude products, including Claude Science, Anthropic’s new AI workbench for scientists!

Post attachment
Kexin Huang avatar
⟳ Yunguan Fu 🧬 retweeted
Kexin Huang@KexinHuang5·Jun 25

Excited to launch Biomni everywhere: web, desktop, mobile, and MCP! Our vision is simple: biological research should happen wherever inspiration strikes. Whether you’re at your bench or riding the subway, Biomni is there when the idea hits. Start a new binder from your phone, then continue seamlessly across devices. We're also launching Biomni MCP, so you can access Biomni’s biology capabilities wherever you work:

Biology doesn't happen in one place. We're bringing Biomni Lab from your browser to your phone, your desktop, and your agent of choice via MCP. Biomni comes with you wherever your work happens. Sign up for the closed beta (Mobile and Desktop): forms.gle/JbB4vV4GdaZcaLU19 Use Biomni MCP today: mcp.phylo.bio/mcp Blog: phylo.bio/blog/biomni-everyw…

0:22