AI + physics-informed protein engineering for manufacturable bispecifics. In silico design resolves light-chain pairing. 7/10 targets, 24/55 BsAbs functional; manufacturability met. With @HormelInst @UMNews.
Preprint: biorxiv.org/content/10.1101/…
Accurate inverse folding models are crucial for antibody CDR design, especially with the rise of LLMs in protein engineering. This benchmark evaluates these models, aiming to improve antibody therapeutics.
biorxiv.org/content/10.1101/…
New high-throughput computational method to profile epitopes using N-linked glycans! This epic collaboration is important for drug design, as it allows for rapid epitope identification and unlocks new innovative therapeutics. Exciting times! 🧬💊
Link: doi.org/10.1002/pro.4726
Per Greisen retweeted
Computational design of N-linked glycans for high throughput epitope profiling biorxiv.org/cgi/content/shor… #bioRxiv
Per Greisen retweeted
For the first time, completely designed — or “de novo,” meaning from the beginning — proteins are being tested in clinical trials, perhaps most notably as a cancer immunotherapy invented in David Baker’s lab and developed by a spinout.
endpts.com/scientists-are-fi…
Per Greisen retweeted
Computational design of mechanically coupled axle-rotor protein assemblies science.org/doi/10.1126/scie…
Per Greisen retweeted
How can we help address #ClimateChange? Our new study in @NatureBiotech with @lanzatech @MichaelKoepke80 @leangching @NUSynBio @smashty @ORNL and @ENERGY demonstrates a new framework for carbon-negative #biomanufacturing , enabled by #cellfree #synbio
nature.com/articles/s41587-0…
Interested in developing and applying next generation #machinelearning to #peptide design and prediction - we have multiple permeant positions at Novo Nordisk #machinelearning as well as summer intern
lnkd.in/grTS8HdD
More on the intern position:
lnkd.in/gNNUNYFB
Fun little project on enzyme specificity and product inhibition - as always fascinating how few mutations can radically change the kinetics of an enzyme
nature.com/articles/s41467-0…
Per Greisen retweeted
"The AlphaFold parameters are made available under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) license" 🙂
github.com/deepmind/alphafol…
(thanks to @BrianWeitzner for alerting me)
Per Greisen retweeted
You can now register for our @proteinsociety symposium in memory of Dan Tawfik, March 9 2022, at 10am EST. Participation is free, but registration is required. us06web.zoom.us/webinar/regi…
Super fun data-driven protein engineering project with great #teamwork
A factor VIIIa–mimetic bispecific antibody, Mim8, ameliorates bleeding upon severe vascular challenge in hemophilia A mice | Blood | American Society of Hematology ashpublications.org/blood/ar…
Per Greisen retweeted
Here's how mutations in #SARSCoV2 Nu variant (B.1.1.529) will affect polyclonal and monoclonal antibodies targeting RBD. These assessments based on deep-mutational scanning experiments; underlying data can be explored interactively at jbloomlab.github.io/SARS2_RB… (1/n)
Per Greisen retweeted
Using coevolution and deep-learning methods (RoseTTAFold & AlphaFold), researchers led by @UWproteindesign and @UTSWNews identified and built accurate models of eukaryotic protein complexes.“...our results herald a new era of structural biology," they say. science.org/doi/10.1126/scie…
Per Greisen retweeted
Congratulations to our scientific team members at Synlogic and @Zymergen for their latest publication in Nature Communications prn.to/3bm3eJb
Per Greisen retweeted
Now out on the cover of @ScienceMagazine:
Our work with @RDasLab on predicting 3D RNA structures!
science.sciencemag.org/conte…
Check out the official press release:
news.stanford.edu/2021/08/26…
and our earlier work on protein complexes:
doi.org/10.1002/prot.26033
Super cool de novo protein from Koga lab - and there is a knot in one of the protein - really interesting!
biorxiv.org/content/10.1101/…
Pretty impressive time scale - great tool for a lot of applications within diagnostics !
An engineered protein-phosphorylation toggle network with implications for endogenous network discovery science.sciencemag.org/conte…
Per Greisen retweeted
New interesting work from DeepMind.
Proteins are not static bricks! Feasibility study to infer a continuous distribution of all states using an end-to-end model from Cryo-EM images to atom coordinates: arxiv.org/abs/2106.14108.
@danrsm, @GarneloMarta, @MichaelZielins, @JonasAAdler, @arkitus, @CarlDoersch, @pushmeet