Umberto Lupo
@umbislupo.bsky.social
Senior AI Scientist at Absci (@abscibio)
Pinned
Origin-1: a generative AI platform for de novo antibody design against novel epitopes [new]
GenAI designs novel Ab targeting. It uses structure, sequence, & co-folding for binding.
GenAI designs novel Ab targeting. It uses structure, sequence, & co-folding for binding.
This was fun work and a remarkable effort across the computational and wet-lab teams!
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
Reposted by Umberto Lupo
Important perspective from Greenland.
January 20, 2026 at 1:31 PM
Important perspective from Greenland.
Reposted by Umberto Lupo
UK Research and Innovation, plus its research councils, quietly quit X.
Via @robinbisson.bsky.social @sophieatrpn.bsky.social
www.researchprofessionalnews.com/rr-news-uk-r...
Via @robinbisson.bsky.social @sophieatrpn.bsky.social
www.researchprofessionalnews.com/rr-news-uk-r...
UK Research and Innovation and its councils quietly quit X - Research Professional News
Funding agency has “been using alternative platforms” amid ongoing controversies over social media company
www.researchprofessionalnews.com
January 19, 2026 at 12:17 PM
UK Research and Innovation, plus its research councils, quietly quit X.
Via @robinbisson.bsky.social @sophieatrpn.bsky.social
www.researchprofessionalnews.com/rr-news-uk-r...
Via @robinbisson.bsky.social @sophieatrpn.bsky.social
www.researchprofessionalnews.com/rr-news-uk-r...
I found the secret Pret at Heathrow Terminal 5
January 19, 2026 at 11:44 AM
I found the secret Pret at Heathrow Terminal 5
This was fun work and a remarkable effort across the computational and wet-lab teams!
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
Origin-1: a generative AI platform for de novo antibody design against novel epitopes [new]
GenAI designs novel Ab targeting. It uses structure, sequence, & co-folding for binding.
GenAI designs novel Ab targeting. It uses structure, sequence, & co-folding for binding.
January 15, 2026 at 11:19 AM
This was fun work and a remarkable effort across the computational and wet-lab teams!
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]
At Absci, we performed de novo antibody design campaigns against "zero-prior" epitopes—lacking structural data from antibody-antigen or protein-protein complexes.
Model architectures, training data curation, and scoring protocols are fully described.
Preprint: www.absci.com/wp-content/u...
Model architectures, training data curation, and scoring protocols are fully described.
Preprint: www.absci.com/wp-content/u...
January 14, 2026 at 10:59 PM
At Absci, we performed de novo antibody design campaigns against "zero-prior" epitopes—lacking structural data from antibody-antigen or protein-protein complexes.
Model architectures, training data curation, and scoring protocols are fully described.
Preprint: www.absci.com/wp-content/u...
Model architectures, training data curation, and scoring protocols are fully described.
Preprint: www.absci.com/wp-content/u...
Reposted by Umberto Lupo
GERMAN PRESIDENT STEINMEIER:
“.. the United States has broken with the values that it helped to establish ..
“.. we have now moved beyond the stage where we can lament the lack of respect for international law or the erosion of the international order; we are far beyond that, I believe.”
“.. the United States has broken with the values that it helped to establish ..
“.. we have now moved beyond the stage where we can lament the lack of respect for international law or the erosion of the international order; we are far beyond that, I believe.”
January 9, 2026 at 9:15 AM
GERMAN PRESIDENT STEINMEIER:
“.. the United States has broken with the values that it helped to establish ..
“.. we have now moved beyond the stage where we can lament the lack of respect for international law or the erosion of the international order; we are far beyond that, I believe.”
“.. the United States has broken with the values that it helped to establish ..
“.. we have now moved beyond the stage where we can lament the lack of respect for international law or the erosion of the international order; we are far beyond that, I believe.”
Recently found out about @jvkersch.bsky.social's `tmtools` github.com/jvkersch/tmt.... It works nicely! Being able to pass user-defined sequence alignments is a nice (if simple) feature that is missing from OpenStructure's own `tmtools` @torstenschwede.bsky.social.
GitHub - jvkersch/tmtools: Python bindings for the TM-align algorithm and code for protein structure comparison developed by Zhang et al.
Python bindings for the TM-align algorithm and code for protein structure comparison developed by Zhang et al. - jvkersch/tmtools
github.com
January 7, 2026 at 2:34 PM
Recently found out about @jvkersch.bsky.social's `tmtools` github.com/jvkersch/tmt.... It works nicely! Being able to pass user-defined sequence alignments is a nice (if simple) feature that is missing from OpenStructure's own `tmtools` @torstenschwede.bsky.social.
Reposted by Umberto Lupo
Black swan moment for energy production and storage?
Just seen the new solid state battery unveiled by Donut Lab and it's an industry changer. This is one to watch. The cells exist to buy TODAY. It's not tech trapped in a lab.
Seriously, watch it all the way through.
www.youtube.com/watch?v=Y-aP...
Seriously, watch it all the way through.
www.youtube.com/watch?v=Y-aP...
World's First All-Solid-State Battery in Production Vehicles | Donut Battery
YouTube video by Donut Lab
www.youtube.com
January 6, 2026 at 6:58 PM
Black swan moment for energy production and storage?
🏛️ 🔍 👀
"There are no optical corrections in the Parthenon"
"There are no optical corrections in the Parthenon"
The illusion of illusions
One of the oldest myths in human history is that the Parthenon was cleverly built with curves so that it looks straight, to correct some visual illusions
In this paper I demonstrate that there are NO optical corrections in the Parthenon
arxiv.org/abs/2510.16831. [1/n]
One of the oldest myths in human history is that the Parthenon was cleverly built with curves so that it looks straight, to correct some visual illusions
In this paper I demonstrate that there are NO optical corrections in the Parthenon
arxiv.org/abs/2510.16831. [1/n]
The illusion of illusions: There are no optical corrections in the Parthenon
One of the oldest and most enduring myths in human history is the belief that the Parthenon was cleverly designed with various curved structures and sizes in order to correct optical illusions, and th...
arxiv.org
October 29, 2025 at 12:58 PM
🏛️ 🔍 👀
"There are no optical corrections in the Parthenon"
"There are no optical corrections in the Parthenon"
Reposted by Umberto Lupo
Visiting Ephesus at #embointegmod25 to look for the missing AlphaFold 3 code
October 2, 2025 at 2:32 PM
Visiting Ephesus at #embointegmod25 to look for the missing AlphaFold 3 code
Reposted by Umberto Lupo
Would you like to come to my lab for a postdoctoral position? Apply for the EPFL AI Centre fellowship: epfl.ch/research/fun.... FYI, I have funding (ERC & Simon's collaboration) also for 1-2 top candidates who will not secure the fellowship. Same for
@krzakalaf.bsky.social lab.
@krzakalaf.bsky.social lab.
EPFL AI Center and Swiss AI Initiative Postdoctoral Fellowships
The 2nd call is now open with a deadline for submissions of 3 November (17.00 CET)!Applications are encouraged from researchers at the postdoctoral level with a keen interest in collaborative, interdi...
epfl.ch
September 30, 2025 at 11:03 AM
Would you like to come to my lab for a postdoctoral position? Apply for the EPFL AI Centre fellowship: epfl.ch/research/fun.... FYI, I have funding (ERC & Simon's collaboration) also for 1-2 top candidates who will not secure the fellowship. Same for
@krzakalaf.bsky.social lab.
@krzakalaf.bsky.social lab.
Reposted by Umberto Lupo
💥 We are *also* organizing Machine Learning for Structural Biology @ EurIPS, Copenhagen!!
Topics include anything in the ML+bio intersection.
Submit your ML+bio short papers! Authors can even present in both locations if they have people in both locations 🚀
Topics include anything in the ML+bio intersection.
Submit your ML+bio short papers! Authors can even present in both locations if they have people in both locations 🚀
You asked and we listened... @workshopmlsb.bsky.social is excited to be expanding to Copenhagen, DK at @euripsconf.bsky.social 🎉
Two workshops (San Diego & Copenhagen) will run concurrently to support broader attendance. You can indicate your location preference(s) in the submission portal💫
Two workshops (San Diego & Copenhagen) will run concurrently to support broader attendance. You can indicate your location preference(s) in the submission portal💫
September 12, 2025 at 1:44 PM
💥 We are *also* organizing Machine Learning for Structural Biology @ EurIPS, Copenhagen!!
Topics include anything in the ML+bio intersection.
Submit your ML+bio short papers! Authors can even present in both locations if they have people in both locations 🚀
Topics include anything in the ML+bio intersection.
Submit your ML+bio short papers! Authors can even present in both locations if they have people in both locations 🚀
Cool-looking work by @piompons.bsky.social @albecazzaniga.bsky.social @fra-cutu.bsky.social, going beyond existing works on sampling different conformational ensembles with AF2.
Our study is now published on JCIM🎉
We expanded and refined the preprint thanks to the insightful feedback from reviewers!
paper: pubs.acs.org/doi/10.1021/...
code: github.com/RitAreaScien...
We expanded and refined the preprint thanks to the insightful feedback from reviewers!
paper: pubs.acs.org/doi/10.1021/...
code: github.com/RitAreaScien...
Evolutionary Constraints Guide AlphaFold2 in Predicting Alternative Conformations and Inform Rational Mutation Design
Investigating structural variability is essential for understanding protein biological functions. Although AlphaFold2 accurately predicts static structures, it fails to capture the full spectrum of f...
pubs.acs.org
September 4, 2025 at 8:10 AM
Cool-looking work by @piompons.bsky.social @albecazzaniga.bsky.social @fra-cutu.bsky.social, going beyond existing works on sampling different conformational ensembles with AF2.
Reposted by Umberto Lupo
Recently tested some de-novo minibinders against two targets (thanks Adaptyv!) designed using our open-source design library, `mosaic`; our best method got hit rates of 7/10 and 8/10 and affinities as low as single-digit nanomolar. Wrote up some thoughts here: blog.escalante.bio/minibinder-d...
September 2, 2025 at 2:39 PM
Recently tested some de-novo minibinders against two targets (thanks Adaptyv!) designed using our open-source design library, `mosaic`; our best method got hit rates of 7/10 and 8/10 and affinities as low as single-digit nanomolar. Wrote up some thoughts here: blog.escalante.bio/minibinder-d...
Reposted by Umberto Lupo
Posts about research on Bluesky receive substantially more attention than similar posts on X, formerly called Twitter
go.nature.com/45Ftiw4
go.nature.com/45Ftiw4
Research posts on Bluesky are more original — and get better engagement
Bluesky posts about science garner more likes and reposts than similar ones on X.
go.nature.com
August 29, 2025 at 8:28 AM
Posts about research on Bluesky receive substantially more attention than similar posts on X, formerly called Twitter
go.nature.com/45Ftiw4
go.nature.com/45Ftiw4
Two exciting openings with us! 🤖🧬🆎🧫💉
- AI Scientist 👉 lnkd.in/eDXHH4E8
- AI Scientist, Drug Creation 👉 lnkd.in/eEvGyaTR
You'll work on antibody sequence/structure design, antibody-antigen co-folding, antibody-antigen binding prediction, physics-based methodologies, and more!
DMs welcome!
- AI Scientist 👉 lnkd.in/eDXHH4E8
- AI Scientist, Drug Creation 👉 lnkd.in/eEvGyaTR
You'll work on antibody sequence/structure design, antibody-antigen co-folding, antibody-antigen binding prediction, physics-based methodologies, and more!
DMs welcome!
August 29, 2025 at 8:13 AM
Two exciting openings with us! 🤖🧬🆎🧫💉
- AI Scientist 👉 lnkd.in/eDXHH4E8
- AI Scientist, Drug Creation 👉 lnkd.in/eEvGyaTR
You'll work on antibody sequence/structure design, antibody-antigen co-folding, antibody-antigen binding prediction, physics-based methodologies, and more!
DMs welcome!
- AI Scientist 👉 lnkd.in/eDXHH4E8
- AI Scientist, Drug Creation 👉 lnkd.in/eEvGyaTR
You'll work on antibody sequence/structure design, antibody-antigen co-folding, antibody-antigen binding prediction, physics-based methodologies, and more!
DMs welcome!
Reposted by Umberto Lupo
🎉 Excited to share that the last paper of my PhD is now published in PRX Life!
We introduce RAG-ESM, a retrieval-augmented framework that makes pretrained protein language models (like ESM2) homology-aware with minimal training cost.
📄 Paper: journals.aps.org/prxlife/abst...
We introduce RAG-ESM, a retrieval-augmented framework that makes pretrained protein language models (like ESM2) homology-aware with minimal training cost.
📄 Paper: journals.aps.org/prxlife/abst...
📢 Our new preprint is out on bioRxiv! We introduce RAG-ESM, a retrieval-augmented framework that improves pretrained protein language models like ESM2 by making them homology-aware with minimal additional training costs.
🔗 doi.org/10.1101/2025...
💻 github.com/Bitbol-Lab/r...
1/7
🔗 doi.org/10.1101/2025...
💻 github.com/Bitbol-Lab/r...
1/7
August 21, 2025 at 4:13 PM
🎉 Excited to share that the last paper of my PhD is now published in PRX Life!
We introduce RAG-ESM, a retrieval-augmented framework that makes pretrained protein language models (like ESM2) homology-aware with minimal training cost.
📄 Paper: journals.aps.org/prxlife/abst...
We introduce RAG-ESM, a retrieval-augmented framework that makes pretrained protein language models (like ESM2) homology-aware with minimal training cost.
📄 Paper: journals.aps.org/prxlife/abst...
@cyrilmalbranke.bsky.social is 🦋!
"ProteomeLM significantly outperforms DCA in recovering experimentally validated interactions. In H. sapiens, ProteomeLM achieves an AUC of 0.83, compared to 0.73 for DCA. Among the top 10 million scored pairs, it recovers 50% of known PPI, versus only 20% for DCA"
"ProteomeLM significantly outperforms DCA in recovering experimentally validated interactions. In H. sapiens, ProteomeLM achieves an AUC of 0.83, compared to 0.73 for DCA. Among the top 10 million scored pairs, it recovers 50% of known PPI, versus only 20% for DCA"
August 21, 2025 at 2:51 PM
@cyrilmalbranke.bsky.social is 🦋!
"ProteomeLM significantly outperforms DCA in recovering experimentally validated interactions. In H. sapiens, ProteomeLM achieves an AUC of 0.83, compared to 0.73 for DCA. Among the top 10 million scored pairs, it recovers 50% of known PPI, versus only 20% for DCA"
"ProteomeLM significantly outperforms DCA in recovering experimentally validated interactions. In H. sapiens, ProteomeLM achieves an AUC of 0.83, compared to 0.73 for DCA. Among the top 10 million scored pairs, it recovers 50% of known PPI, versus only 20% for DCA"
Reposted by Umberto Lupo
A no-miss opportunity to join EPFL. We are a vibrant (had to use it 😉) community working in all branches of basic sciences and engineering, and interdisciplinary studies are strongly encouraged.
If you want more information, contact the committee heads, or feel free to DM me.
tinyurl.com/bd27ky7p
If you want more information, contact the committee heads, or feel free to DM me.
tinyurl.com/bd27ky7p
Assistant/associate Professor of AI-assisted Biophysics at the Ecole polytechnique fédérale de Lausanne (EPFL) and Group Leader at the Paul Scherrer Institute (PSI)
The School of Basic Sciences (Physics, Chemistry and Mathematics) at EPFL and the Paul Scherrer Institute (PSI) jointly seek to appoint a Tenure Track Assistant or tenured Associate Professor in AI-As...
tinyurl.com
August 21, 2025 at 6:35 AM
A no-miss opportunity to join EPFL. We are a vibrant (had to use it 😉) community working in all branches of basic sciences and engineering, and interdisciplinary studies are strongly encouraged.
If you want more information, contact the committee heads, or feel free to DM me.
tinyurl.com/bd27ky7p
If you want more information, contact the committee heads, or feel free to DM me.
tinyurl.com/bd27ky7p
"When even the most benign activities are subject to capricious disruption and political interference, the luxury of disengagement is no longer a viable option."
I wrote an op-ed on the world-class STEM research ecosystem in the United States, and how this ecosystem is now under attack on multiple fronts by the current administration: newsletter.ofthebrave.org/p/im-an-awar...
I’m an award-winning mathematician. Trump just cut my funding.
The “Mozart of Math” tried to stay out of politics. Then it came for his research.
newsletter.ofthebrave.org
August 19, 2025 at 1:02 PM
"When even the most benign activities are subject to capricious disruption and political interference, the luxury of disengagement is no longer a viable option."
Reposted by Umberto Lupo
AtomWorks is out! Building upon @biotite_python, we built a toolkit for all things biomolecules and trained RF3 with it. All open-source, test it via `pip install atomworks`!
AtomWorks: github.com/RosettaCommo...
RF3: github.com/RosettaCommo...
Paper: tinyurl.com/y2w4z65b
1/6
AtomWorks: github.com/RosettaCommo...
RF3: github.com/RosettaCommo...
Paper: tinyurl.com/y2w4z65b
1/6
August 15, 2025 at 5:44 PM
AtomWorks is out! Building upon @biotite_python, we built a toolkit for all things biomolecules and trained RF3 with it. All open-source, test it via `pip install atomworks`!
AtomWorks: github.com/RosettaCommo...
RF3: github.com/RosettaCommo...
Paper: tinyurl.com/y2w4z65b
1/6
AtomWorks: github.com/RosettaCommo...
RF3: github.com/RosettaCommo...
Paper: tinyurl.com/y2w4z65b
1/6
Reposted by Umberto Lupo
From social media apps to bank loans and hiring/firing decisions to search engines and shopping sites, are you tired of living a life ruled by opaque algorithms?
Pre-order now to Take Back Control on August 5.
bit.ly/4idpeGr
Pre-order now to Take Back Control on August 5.
bit.ly/4idpeGr
June 4, 2025 at 4:28 PM
From social media apps to bank loans and hiring/firing decisions to search engines and shopping sites, are you tired of living a life ruled by opaque algorithms?
Pre-order now to Take Back Control on August 5.
bit.ly/4idpeGr
Pre-order now to Take Back Control on August 5.
bit.ly/4idpeGr
Reposted by Umberto Lupo
Excited to share work with
Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org
biorxiv.org/content/10.1...
TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵
Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org
biorxiv.org/content/10.1...
TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵
Scaling down protein language modeling with MSA Pairformer
Recent efforts in protein language modeling have focused on scaling single-sequence models and their training data, requiring vast compute resources that limit accessibility. Although models that use ...
biorxiv.org
August 5, 2025 at 6:31 AM
Excited to share work with
Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org
biorxiv.org/content/10.1...
TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵
Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org
biorxiv.org/content/10.1...
TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵
Reposted by Umberto Lupo
Our development of machine-learned transferable coarse-grained models in now on Nat Chem! doi.org/10.1038/s415...
I am so proud of my group for this work! Particularly first authors Nick Charron, Klara Bonneau, Aldo Pasos-Trejo, Andrea Guljas.
I am so proud of my group for this work! Particularly first authors Nick Charron, Klara Bonneau, Aldo Pasos-Trejo, Andrea Guljas.
Navigating protein landscapes with a machine-learned transferable coarse-grained model - Nature Chemistry
The development of a universal protein coarse-grained model has been a long-standing challenge. A coarse-grained model with chemical transferability has now been developed by combining deep-learning m...
doi.org
July 18, 2025 at 10:45 AM
Our development of machine-learned transferable coarse-grained models in now on Nat Chem! doi.org/10.1038/s415...
I am so proud of my group for this work! Particularly first authors Nick Charron, Klara Bonneau, Aldo Pasos-Trejo, Andrea Guljas.
I am so proud of my group for this work! Particularly first authors Nick Charron, Klara Bonneau, Aldo Pasos-Trejo, Andrea Guljas.