Very cool work! Really pinning down some important hypotheses about virus emergence!
got myself a comment in the nyt article (although @davidlrobertson.bsky.social had the best quote! 👀)
Archived article link here: archive.md/UjxJ0
There's a common misconception that zoonotic viruses require significant adaptation to jump from animals to cause human epidemics.
Not so 👇.
Further, we see clear signs of 1977 flu experiencing cell passage, prior to epidemic.
SARS-CoV-2? Business as usual.
www.sciencedirect.com/science/arti...
In new study led by Brendan Larsen, we map functional constraint across the Nipah virus F protein to define constrained epitopes for antibody targeting and identify mutations that stabilize the prefusion conformation for vaccine immunogens.
www.pnas.org/doi/10.1073/...
We have an open Assistant Professor position (tenure track) @ethz.ch in Computational Immunology - ethz.ch/en/the-eth-z... Please apply / share!
🤩 🤩 🤩
It is almost five years since I first applied AlphaFold to a handful of viral proteins.
Today, hundreds of thousands of predictions and many discoveries later, we are delighted to announce the integration of our Viro3D dataset into the @ebi.embl.org and Google DeepMind AlphaFold Database.
Plus got to do some advertising for the new lab 👀 👀
#strphy26
What a great conference!! Leaving Brisbane inspired about the future of structural phylogenetics!
Thanks again to the organisers @cpuentelelievre.bsky.social @proteinmechanic.bsky.social and Jordan Douglas for an amazing work putting it all together!
Despite the rainy weather, the seafood is still top notch! 🦪 🦪
Very excited to be in (slightly rainy until tomorrow ☔) Brisbane attending the first Australasian Protein Structural Phylogenetics Meeting #strphy26 @strphy.bsky.social
Catch me at my talk tomorrow afternoon or at welcome drinks tonight! 🥂
First human #H5N1 case of 2026 in #Cambodia: a 30-year-old man from Kampot province. Sequencing, virus isolation & animal investigations ongoing. Dead chicken at home before onset. Patient has since recovered. 🦠💉😷🐓🥼🔬🧪
Sequencing, contact tracing, and poultry investigations ongoing.
This work will hopefully form the basis for a lot of the exciting work we will be doing at the (soon to be!) Antigen Evolution & Design laboratory at
@pasteur.fr
🇫🇷 (coming in June 2026👀👀)
2️⃣A set of bespoke versions of the popular ESM2 and ProtT5 pLMs finetuned on influenza A virus HA sequences.
zenodo.org/records/1489...
Our paper comes with:
1️⃣The plm_entropy python package for seamlessly applying our method to your own sequences
github.com/spyros-lytra...
We apply this to the influenza a virus HA protein and test how our method performs between different HA subtypes.
From a sequence alignment one can infer which sites are on average more or less likely to change in a group of related proteins
What we tried here is using pLMs to infer variable or conserved sites for a single sequence, or sequence context.
Our paper on inferring context dependent entropy using protein language models is officially out in NAR Genomics & Bioinformatics! 🧬🤖
with Adam Strange, Jumpei Ito, and @systemsvirology.bsky.social
academic.oup.com/nargab/artic...
details below...
#NARGAB
Congrats!! 🎉🎉
FoldMason is out now in @science.org. It generates accurate multiple structure alignments for thousands of protein structures in seconds. Great work by Cameron L. M. Gilchrist and @milot.bsky.social.
📄 www.science.org/doi/10.1126/...
🌐 search.foldseek.com/foldmason
💾 github.com/steineggerla...
Remember? RNA virus journal club is already this Thursday! 🤩
Join us and join our (RdRp summit) slack (link in the comment):
calendar.app.google/K2fwcsL78K1E...
RNA Virus Journal ClubJoin Zoom Meeting us02web.zoom.us/j/8109697228...
Meeting ID: 810 9697 2281
Passcode: 174610
Great opportunity to learn how to make very cool looking phylogenies by a real tree wizard @evogytis.bsky.social ! 🌲🧙♂️
Don't miss out!
🎉 Congratulations to Dr Daniel Weir for recently passing his Viva!
Daniels research explored how influenza A virus infection moves directly between cells, independently of virus particles, focusing on how infection induces the formation of tunnelling nanotube-like structures.
Congrats to first author Wenye Li and all collaborators, especially the team that collected all the valuable samples in Vietnam! 🇻🇳 🎉
Bats are thought to have little to no disease from many of their viruses. Our work may shift this paradigm, at least when it comes to SARSr-CoVs and horseshoe bats, since positive selection in the virus receptor would require an evolutionary pressure to lower virus infectivity.
Taken together, this data is the first direct evidence of SARSr-CoV infection acting as a selective pressure on the receptor of the viruses’ natural reservoir host!
We further present a cryoEM structure showing how the R. affinis ACE2 site 24 directly interacts with the Ra22QT77 spike receptor-binding domain, explaining its effect on virus infectivity (read more about the evolution of the Ra22QT77 spike and its relatives in the preprint!)
...while there's geographic separation of the bats with each ACE2 residue, consistent with the R. affinis ACE2 tree.
ACE2 site 24 in particular has evidence of positive selection and has had at least two infectivity-changing substitutions specific to the north bat ACE2 lineage.
Since we know which amino acids differ between the ACE2 proteins, we can model the effect of each amino acid change to the infectivity of each virus spike.
We find that certain ACE2 sites known to be on the ACE2-spike interacting surface clearly virus affect infectivity...
We performed pseudovirus assays using spikes from 36 SARSr-CoV against cells expressing all R. affinis ACE2 genotypes. What we see is surprising variation in how each virus spike can infect each ACE2, even though they all come from the same bat species!
Interestingly, despite all bats being from the same species, there were 15 unique ACE2 genotypes in our samples, suggesting that there's substantial intraspecies diversity in the ACE2 proteins.
So we asked, could this diversity affect how SARSr-CoVs infect these bats? 🤔
