Thomas Guiziou
@cen0ten.bsky.social
PhD Student in the Vorobieva Lab in de novo protein design
HalluDesign: Protein Optimization and de novo Design via Iterative Structure Hallucination and Sequence design
It is the same approach as ProteinHunter if I understood both papers. www.biorxiv.org/content/10.1...
1-
It is the same approach as ProteinHunter if I understood both papers. www.biorxiv.org/content/10.1...
1-
HalluDesign: Protein Optimization and de novo Design via Iterative Structure Hallucination and Sequence design
Deep learning has revolutionized biomolecular modeling, enabling the prediction of diverse structures with atomic accuracy. However, leveraging the atomic-level precision of the structure prediction m...
www.biorxiv.org
December 2, 2025 at 11:04 AM
HalluDesign: Protein Optimization and de novo Design via Iterative Structure Hallucination and Sequence design
It is the same approach as ProteinHunter if I understood both papers. www.biorxiv.org/content/10.1...
1-
It is the same approach as ProteinHunter if I understood both papers. www.biorxiv.org/content/10.1...
1-
www.biorxiv.org/content/10.1...
Energy-guided combinatorial co-optimization of antibody affinity and stability
First use of ddG with Ala on the CDRs to map the mutation tolerance of the CDRs. Then apply Funclib to the CDRs to find mutation that should improve affinity.
1/2
Energy-guided combinatorial co-optimization of antibody affinity and stability
First use of ddG with Ala on the CDRs to map the mutation tolerance of the CDRs. Then apply Funclib to the CDRs to find mutation that should improve affinity.
1/2
Energy-guided combinatorial co-optimization of antibody affinity and stability
Affinity maturation is an essential process in antibody engineering. Although powerful, it is iterative and time-consuming and can result in trade-offs, where affinity is gained at the cost of other e...
www.biorxiv.org
November 30, 2025 at 10:08 PM
www.biorxiv.org/content/10.1...
Energy-guided combinatorial co-optimization of antibody affinity and stability
First use of ddG with Ala on the CDRs to map the mutation tolerance of the CDRs. Then apply Funclib to the CDRs to find mutation that should improve affinity.
1/2
Energy-guided combinatorial co-optimization of antibody affinity and stability
First use of ddG with Ala on the CDRs to map the mutation tolerance of the CDRs. Then apply Funclib to the CDRs to find mutation that should improve affinity.
1/2
Exciting paper from colleagues:
www.nature.com/articles/s41...
www.nature.com/articles/s41...
MISO: microfluidic protein isolation enables single-particle cryo-EM structure determination from a single cell colony - Nature Methods
The micro isolation (MISO) method combines microfluidics-based protein purification with cryo-EM grid preparation, enabling cryo-EM structure determination of soluble and membrane proteins starting fr...
www.nature.com
November 16, 2025 at 3:28 PM
Exciting paper from colleagues:
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted by Thomas Guiziou
If you’re interested in learning more about protein folding and misfolding, I’ve created a convenient reading list with a few essential papers:
scholar.google.com/citations?us...
scholar.google.com/citations?us...
scholar.google.com/citations?us...
scholar.google.com/citations?us...
August 6, 2025 at 7:01 PM
If you’re interested in learning more about protein folding and misfolding, I’ve created a convenient reading list with a few essential papers:
scholar.google.com/citations?us...
scholar.google.com/citations?us...
scholar.google.com/citations?us...
scholar.google.com/citations?us...
Statistical significance does not always equate biological significance, very clear paper arguing against the use of p-values and null hypothesis test. Instead agues that confidence interval should be used. 1/2
rupress.org/jcb/article/...
rupress.org/jcb/article/...
Alternative to the statistical mass confusion of testing for “no effect” | Journal of Cell Biology | Rockefeller University Press
Morgan discusses how progress in cell biology is hindered by significance testing and the need for a shift to effect size estimation.
rupress.org
August 2, 2025 at 10:14 PM
Statistical significance does not always equate biological significance, very clear paper arguing against the use of p-values and null hypothesis test. Instead agues that confidence interval should be used. 1/2
rupress.org/jcb/article/...
rupress.org/jcb/article/...
Reposted by Thomas Guiziou
In 1890, the X and Y chromosomes were discovered. It was found that the men who were tested had 46 chromosomes, including an X and a Y, while women who were tested also had 46 chromosomes, including 2 X chromosomes.
So obviously the conclusion was that the Y chromosome defined masculinity.
So obviously the conclusion was that the Y chromosome defined masculinity.
November 8, 2024 at 11:28 PM
In 1890, the X and Y chromosomes were discovered. It was found that the men who were tested had 46 chromosomes, including an X and a Y, while women who were tested also had 46 chromosomes, including 2 X chromosomes.
So obviously the conclusion was that the Y chromosome defined masculinity.
So obviously the conclusion was that the Y chromosome defined masculinity.
Reposted by Thomas Guiziou
The "reproducibility crisis" in science constantly makes headlines. Repro efforts are often limited. What if you could assess reproducibility of an entire field?
That's what @brunolemaitre.bsky.social et al. have done. Fly immunity is highly replicable & offers lessons for #metascience
A 🧵 1/n
That's what @brunolemaitre.bsky.social et al. have done. Fly immunity is highly replicable & offers lessons for #metascience
A 🧵 1/n
July 10, 2025 at 8:23 AM
The "reproducibility crisis" in science constantly makes headlines. Repro efforts are often limited. What if you could assess reproducibility of an entire field?
That's what @brunolemaitre.bsky.social et al. have done. Fly immunity is highly replicable & offers lessons for #metascience
A 🧵 1/n
That's what @brunolemaitre.bsky.social et al. have done. Fly immunity is highly replicable & offers lessons for #metascience
A 🧵 1/n
I am not particularly surprised for those results. it again highlight the need for better benchmarks for deep learning models.
The timing of the paper is also quite interesting, after the sequence on the absence of the LLM Siri at the WWDC.
machinelearning.apple.com/research/ill...
The timing of the paper is also quite interesting, after the sequence on the absence of the LLM Siri at the WWDC.
machinelearning.apple.com/research/ill...
The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
Recent generations of frontier language models have introduced Large Reasoning Models (LRMs) that generate detailed thinking processes…
machinelearning.apple.com
June 12, 2025 at 7:57 AM
I am not particularly surprised for those results. it again highlight the need for better benchmarks for deep learning models.
The timing of the paper is also quite interesting, after the sequence on the absence of the LLM Siri at the WWDC.
machinelearning.apple.com/research/ill...
The timing of the paper is also quite interesting, after the sequence on the absence of the LLM Siri at the WWDC.
machinelearning.apple.com/research/ill...
Reposted by Thomas Guiziou
New paper from the lab from Sriram Garg in my group. We introduce a general substitution matrix for structural phylogenetics. I think this is a big deal, so read on below if you think deep history is important. academic.oup.com/mbe/advance-...
A general substitution matrix for structural phylogenetics.
Abstract. Sequence-based maximum likelihood (ML) phylogenetics is a widely used method for inferring evolutionary relationships, which has illuminated the
academic.oup.com
June 11, 2025 at 2:01 PM
New paper from the lab from Sriram Garg in my group. We introduce a general substitution matrix for structural phylogenetics. I think this is a big deal, so read on below if you think deep history is important. academic.oup.com/mbe/advance-...
Reposted by Thomas Guiziou
Going to the hospital because I broke my wrist smashing the endorse button:
www.understandingai.org/p/i-got-fool...
www.understandingai.org/p/i-got-fool...
I got fooled by AI-for-science hype—here's what it taught me
I used AI in my plasma physics research and it didn’t go the way I expected.
www.understandingai.org
May 19, 2025 at 6:04 PM
Going to the hospital because I broke my wrist smashing the endorse button:
www.understandingai.org/p/i-got-fool...
www.understandingai.org/p/i-got-fool...
Eye opening talk, I could not recommend it enough.
If you missed out on Sarel Fleishman's extraordinary talk last night, have no fear - you can check out the recording 👇
youtu.be/2aaWh8yVq8U
youtu.be/2aaWh8yVq8U
Designing new and improved functions in natural protein folds
YouTube video by Boston Protein Design and Modeling Club
youtu.be
May 17, 2025 at 10:48 AM
Eye opening talk, I could not recommend it enough.
If physicists are bad philosophers, what does that make biologists? 🙃
www.nature.com/articles/d41...
www.nature.com/articles/d41...
Why bad philosophy is stopping progress in physics
Theoretical physicists are in thrall to a misguided mindset that allows viable ideas to be advanced only by overturning what already exists.
www.nature.com
May 15, 2025 at 9:14 AM
If physicists are bad philosophers, what does that make biologists? 🙃
www.nature.com/articles/d41...
www.nature.com/articles/d41...
Reposted by Thomas Guiziou
Here's a command that converts a DOI to bibtex:
May 6, 2025 at 10:45 PM
Here's a command that converts a DOI to bibtex:
Very interesting and clear paper proposing other avenues to improve models than model size or novel architectures. In addition to a method of probing which pLM is more suited to a given dataset.
Quantifying uncertainty in Protein Representations Across Models and Task [new]
Quantifies the reliability of protein embeddings from language models. It identifies failure modes and enhances the robustness of downstream tasks.
Quantifies the reliability of protein embeddings from language models. It identifies failure modes and enhances the robustness of downstream tasks.
May 6, 2025 at 8:48 PM
Very interesting and clear paper proposing other avenues to improve models than model size or novel architectures. In addition to a method of probing which pLM is more suited to a given dataset.
Reposted by Thomas Guiziou
In silico discovery of nanobody binders to a G-protein coupled receptor using AlphaFold-Multimer https://www.biorxiv.org/content/10.1101/2025.03.05.640882v1
March 10, 2025 at 5:45 PM
In silico discovery of nanobody binders to a G-protein coupled receptor using AlphaFold-Multimer https://www.biorxiv.org/content/10.1101/2025.03.05.640882v1
Reposted by Thomas Guiziou
📢📢 "Proteina: Scaling Flow-based Protein Structure Generative Models"
#ICLR2025 (Oral Presentation)
🔥 Project page: research.nvidia.com/labs/genair/...
📜 Paper: arxiv.org/abs/2503.00710
🛠️ Code and weights: github.com/NVIDIA-Digit...
🧵Details in thread...
(1/n)
#ICLR2025 (Oral Presentation)
🔥 Project page: research.nvidia.com/labs/genair/...
📜 Paper: arxiv.org/abs/2503.00710
🛠️ Code and weights: github.com/NVIDIA-Digit...
🧵Details in thread...
(1/n)
March 4, 2025 at 5:09 PM
📢📢 "Proteina: Scaling Flow-based Protein Structure Generative Models"
#ICLR2025 (Oral Presentation)
🔥 Project page: research.nvidia.com/labs/genair/...
📜 Paper: arxiv.org/abs/2503.00710
🛠️ Code and weights: github.com/NVIDIA-Digit...
🧵Details in thread...
(1/n)
#ICLR2025 (Oral Presentation)
🔥 Project page: research.nvidia.com/labs/genair/...
📜 Paper: arxiv.org/abs/2503.00710
🛠️ Code and weights: github.com/NVIDIA-Digit...
🧵Details in thread...
(1/n)
The feel good event from this weekend in France.
A chartity event from the gaming community raised more than 2 million € for Medecins du Monde.
It was also a very inclusive event with close to gender parity and LGBTQ+ inclusion, which is definitely not usually the case for gaming events.
A chartity event from the gaming community raised more than 2 million € for Medecins du Monde.
It was also a very inclusive event with close to gender parity and LGBTQ+ inclusion, which is definitely not usually the case for gaming events.
2 231 146 €
2 231 146 mercis, 2 231 146 raisons de soutenir les actions de @mdmfrance.bsky.social.
#SpeeDons
2 231 146 mercis, 2 231 146 raisons de soutenir les actions de @mdmfrance.bsky.social.
#SpeeDons
March 4, 2025 at 12:38 PM
The feel good event from this weekend in France.
A chartity event from the gaming community raised more than 2 million € for Medecins du Monde.
It was also a very inclusive event with close to gender parity and LGBTQ+ inclusion, which is definitely not usually the case for gaming events.
A chartity event from the gaming community raised more than 2 million € for Medecins du Monde.
It was also a very inclusive event with close to gender parity and LGBTQ+ inclusion, which is definitely not usually the case for gaming events.
Reposted by Thomas Guiziou
I am very happy to announce that we solved the in-virus #cryoEM structure of the #HIV matrix protein. The structure helped us to resolve a long mystery in HIV biology. A🧵1/5
Read more in Nature: www.nature.com/articles/s41... 🧪
Read more in Nature: www.nature.com/articles/s41... 🧪
March 3, 2025 at 6:57 PM
I am very happy to announce that we solved the in-virus #cryoEM structure of the #HIV matrix protein. The structure helped us to resolve a long mystery in HIV biology. A🧵1/5
Read more in Nature: www.nature.com/articles/s41... 🧪
Read more in Nature: www.nature.com/articles/s41... 🧪
Reposted by Thomas Guiziou
Large scale investigation of GPCR molecular dynamics data uncovers allosteric sites and lateral gateways
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Large scale investigation of GPCR molecular dynamics data uncovers allosteric sites and lateral gateways - Nature Communications
GPCR dynamics are critical for signal transmission into the cell. Here, the authors probe receptor plasticity using molecular dynamics simulations, enabling the detection of allosteric sites and drug ...
www.nature.com
March 2, 2025 at 9:18 AM
Large scale investigation of GPCR molecular dynamics data uncovers allosteric sites and lateral gateways
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted by Thomas Guiziou
We wrote a review article on modelling and design of transcriptional enhancers using sequence-to-function models.
From conventional machine learning methods to CNNs and using models as oracles/generative AI for synthetic enhancer design!
@natrevbioeng.bsky.social
www.nature.com/articles/s44...
From conventional machine learning methods to CNNs and using models as oracles/generative AI for synthetic enhancer design!
@natrevbioeng.bsky.social
www.nature.com/articles/s44...
Modelling and design of transcriptional enhancers - Nature Reviews Bioengineering
Enhancers are genomic elements critical for regulating gene expression. In this Review, the authors discuss how sequence-to-function models can be used to unravel the rules underlying enhancer activit...
www.nature.com
February 28, 2025 at 2:45 PM
We wrote a review article on modelling and design of transcriptional enhancers using sequence-to-function models.
From conventional machine learning methods to CNNs and using models as oracles/generative AI for synthetic enhancer design!
@natrevbioeng.bsky.social
www.nature.com/articles/s44...
From conventional machine learning methods to CNNs and using models as oracles/generative AI for synthetic enhancer design!
@natrevbioeng.bsky.social
www.nature.com/articles/s44...
Reposted by Thomas Guiziou
To pack or not to pack: revisiting protein side-chain packing in the post-AlphaFold era [new]
PSCP struggles to generalize AlphaFold repacking, even w/ confidence. Highlights challenges.
PSCP struggles to generalize AlphaFold repacking, even w/ confidence. Highlights challenges.
February 28, 2025 at 3:23 AM
To pack or not to pack: revisiting protein side-chain packing in the post-AlphaFold era [new]
PSCP struggles to generalize AlphaFold repacking, even w/ confidence. Highlights challenges.
PSCP struggles to generalize AlphaFold repacking, even w/ confidence. Highlights challenges.
Reposted by Thomas Guiziou
Conditional Diffusion with Locality-Aware Modal Alignment for Generating Diverse Protein Conformational Ensembles [new]
Generates diverse protein conformations via diffusion, aligning sequence with geometry.
Generates diverse protein conformations via diffusion, aligning sequence with geometry.
February 28, 2025 at 3:42 AM
Conditional Diffusion with Locality-Aware Modal Alignment for Generating Diverse Protein Conformational Ensembles [new]
Generates diverse protein conformations via diffusion, aligning sequence with geometry.
Generates diverse protein conformations via diffusion, aligning sequence with geometry.
Reposted by Thomas Guiziou
We wrote a review on the "Physics of droplet regulation in biological cells": arxiv.org/abs/2501.13639 Beside the basic #physics of phase separation, we discuss three aspects that separate cellular from traditional droplets:
January 24, 2025 at 7:55 AM
We wrote a review on the "Physics of droplet regulation in biological cells": arxiv.org/abs/2501.13639 Beside the basic #physics of phase separation, we discuss three aspects that separate cellular from traditional droplets:
Reposted by Thomas Guiziou
Register Now for the Webinar on "Streamlining Access to RCSB PDB APIs with Python" on Monday March 24
pdb101.rcsb.org/news...
pdb101.rcsb.org/news...
PDB101: Register Now for Webinar on API Access with Python
PDB-101: Training, Outreach, and Education portal of RCSB PDB
pdb101.rcsb.org
February 18, 2025 at 9:55 PM
Register Now for the Webinar on "Streamlining Access to RCSB PDB APIs with Python" on Monday March 24
pdb101.rcsb.org/news...
pdb101.rcsb.org/news...