Michael Jendrusch
@mjendrusch.bsky.social
(he/him) Former PhD student / Postdoc @ Korbel group, EMBL.
Protein ML person, mathematics & science enthusiast.
developer of salad
preprint: https://www.biorxiv.org/content/10.1101/2025.01.31.635780v1
code: https://github.com/mjendrusch/salad
Protein ML person, mathematics & science enthusiast.
developer of salad
preprint: https://www.biorxiv.org/content/10.1101/2025.01.31.635780v1
code: https://github.com/mjendrusch/salad
Pinned
New protein ML preprint from my PhD project.
We describe salad (sparse all-atom denoising), a family of efficient protein structure diffusion models and show that it works well on a bunch of protein design task previously described in the literature.
Preprint: www.biorxiv.org/content/10.1...
(1/N)
We describe salad (sparse all-atom denoising), a family of efficient protein structure diffusion models and show that it works well on a bunch of protein design task previously described in the literature.
Preprint: www.biorxiv.org/content/10.1...
(1/N)
Reposted by Michael Jendrusch
Brand new preprint from my lab, showing that TnpB, the ancestor of Cas12, acts as a gene drive in plasmids! And it turns out in conjugative plasmids that it acts as a primitive anti-self defense system, providing a potential link between its transposon effect and becoming CRISPR!
What is the best strategy to win any contest?
Eliminate your opponents of course.
Recently, my friend @fernpizza.bsky.social showed how plasmids compete intracellularly (check out his paper published in Science today!). With @baym.lol, we now know they can fight.
www.biorxiv.org/content/10.1...
Eliminate your opponents of course.
Recently, my friend @fernpizza.bsky.social showed how plasmids compete intracellularly (check out his paper published in Science today!). With @baym.lol, we now know they can fight.
www.biorxiv.org/content/10.1...
November 20, 2025 at 10:17 PM
Brand new preprint from my lab, showing that TnpB, the ancestor of Cas12, acts as a gene drive in plasmids! And it turns out in conjugative plasmids that it acts as a primitive anti-self defense system, providing a potential link between its transposon effect and becoming CRISPR!
Reposted by Michael Jendrusch
New preprint! We measured temperature- and pH-induced aggregation for over 18,000 natural and de novo designed protein domains!
November 19, 2025 at 9:16 PM
New preprint! We measured temperature- and pH-induced aggregation for over 18,000 natural and de novo designed protein domains!
Reposted by Michael Jendrusch
Tired of knowing what you are looking for in your tomograms?
Why not try bluesky-less Frosina's new self-supervised algorithm for semantic segmentation and particle picking for #teamtomo (includes a new denoiser without the need for odd/even tomos)
Code is on GitHub if you want to try it out
Why not try bluesky-less Frosina's new self-supervised algorithm for semantic segmentation and particle picking for #teamtomo (includes a new denoiser without the need for odd/even tomos)
Code is on GitHub if you want to try it out
November 19, 2025 at 2:04 PM
Tired of knowing what you are looking for in your tomograms?
Why not try bluesky-less Frosina's new self-supervised algorithm for semantic segmentation and particle picking for #teamtomo (includes a new denoiser without the need for odd/even tomos)
Code is on GitHub if you want to try it out
Why not try bluesky-less Frosina's new self-supervised algorithm for semantic segmentation and particle picking for #teamtomo (includes a new denoiser without the need for odd/even tomos)
Code is on GitHub if you want to try it out
Reposted by Michael Jendrusch
A few py2Dmol updates 🧬
py2dmol.solab.org
Integration with AlphaFoldDB (will auto fetch results). Drag and drop results from AF3-server or ColabFold for interactive experience! (1/4)
py2dmol.solab.org
Integration with AlphaFoldDB (will auto fetch results). Drag and drop results from AF3-server or ColabFold for interactive experience! (1/4)
November 19, 2025 at 8:15 AM
A few py2Dmol updates 🧬
py2dmol.solab.org
Integration with AlphaFoldDB (will auto fetch results). Drag and drop results from AF3-server or ColabFold for interactive experience! (1/4)
py2dmol.solab.org
Integration with AlphaFoldDB (will auto fetch results). Drag and drop results from AF3-server or ColabFold for interactive experience! (1/4)
Reposted by Michael Jendrusch
#1 lesson I give to protein designers for wet lab purposes : Please design with a Trp and/or multiple Tyr. Please do not design with 10 cysteines.
November 18, 2025 at 1:48 PM
#1 lesson I give to protein designers for wet lab purposes : Please design with a Trp and/or multiple Tyr. Please do not design with 10 cysteines.
Reposted by Michael Jendrusch
Has this idea of sampling neighbors from an inverse cubic distribution been used in protein structure GNNs at all since the idea was first presented in the Chroma paper ~3 years ago? I thought it was incredibly clever and expected it to catch on, but haven't seen it since
November 16, 2025 at 9:01 PM
Has this idea of sampling neighbors from an inverse cubic distribution been used in protein structure GNNs at all since the idea was first presented in the Chroma paper ~3 years ago? I thought it was incredibly clever and expected it to catch on, but haven't seen it since
Reposted by Michael Jendrusch
hey bsky! excited to share what i've been working on for a few years now.
cryoJAX: A Cryo-EM Image Simulation Library In JAX
www.biorxiv.org/content/10.1...
cryoJAX is a modular framework for implementing forward models of cryo-EM image formation. these can be used to build powerful data analyses!
cryoJAX: A Cryo-EM Image Simulation Library In JAX
www.biorxiv.org/content/10.1...
cryoJAX is a modular framework for implementing forward models of cryo-EM image formation. these can be used to build powerful data analyses!
www.biorxiv.org
November 14, 2025 at 5:33 PM
hey bsky! excited to share what i've been working on for a few years now.
cryoJAX: A Cryo-EM Image Simulation Library In JAX
www.biorxiv.org/content/10.1...
cryoJAX is a modular framework for implementing forward models of cryo-EM image formation. these can be used to build powerful data analyses!
cryoJAX: A Cryo-EM Image Simulation Library In JAX
www.biorxiv.org/content/10.1...
cryoJAX is a modular framework for implementing forward models of cryo-EM image formation. these can be used to build powerful data analyses!
Reposted by Michael Jendrusch
Global Analysis of Aggregation Determinants in Small Protein Domains https://www.biorxiv.org/content/10.1101/2025.11.11.687847v1
November 13, 2025 at 1:47 AM
Global Analysis of Aggregation Determinants in Small Protein Domains https://www.biorxiv.org/content/10.1101/2025.11.11.687847v1
Reposted by Michael Jendrusch
Are you an early-stage graduate student (2nd or 3rd year) or early-stage postdoc based in the US or Canada, working primarily in Drosophila? Would you like to help improve the experience of all trainees working in Drosophila research? If so, read on.
(Please repost to reach a broad audience.)
(Please repost to reach a broad audience.)
November 12, 2025 at 4:49 AM
Are you an early-stage graduate student (2nd or 3rd year) or early-stage postdoc based in the US or Canada, working primarily in Drosophila? Would you like to help improve the experience of all trainees working in Drosophila research? If so, read on.
(Please repost to reach a broad audience.)
(Please repost to reach a broad audience.)
Reposted by Michael Jendrusch
#proteindesign #proteinbinder
100 designs
⬇️
34 got through the screening (ipTM, PAE, RMSD)
⬇️
4 ordered (gene synthesis + codon optimization)
⬇️
3 expressed and soluble
⬇️
1 interacts with its target !!!!
Now the story begins !
100 designs
⬇️
34 got through the screening (ipTM, PAE, RMSD)
⬇️
4 ordered (gene synthesis + codon optimization)
⬇️
3 expressed and soluble
⬇️
1 interacts with its target !!!!
Now the story begins !
a man with glasses and a mustache is wearing a white shirt and says yes .
ALT: a man with glasses and a mustache is wearing a white shirt and says yes .
media.tenor.com
November 7, 2025 at 12:03 PM
#proteindesign #proteinbinder
100 designs
⬇️
34 got through the screening (ipTM, PAE, RMSD)
⬇️
4 ordered (gene synthesis + codon optimization)
⬇️
3 expressed and soluble
⬇️
1 interacts with its target !!!!
Now the story begins !
100 designs
⬇️
34 got through the screening (ipTM, PAE, RMSD)
⬇️
4 ordered (gene synthesis + codon optimization)
⬇️
3 expressed and soluble
⬇️
1 interacts with its target !!!!
Now the story begins !
Reposted by Michael Jendrusch
Can an AI tool help us better understand the origins of cancer?
Researchers from EMBL's Korbel Group have developed a new AI method – MAGIC – which, through a game of molecular laser tag, is shedding light on how chromosomal abnormalities form in cells.
www.embl.org/news/science...
Researchers from EMBL's Korbel Group have developed a new AI method – MAGIC – which, through a game of molecular laser tag, is shedding light on how chromosomal abnormalities form in cells.
www.embl.org/news/science...
October 29, 2025 at 4:11 PM
Can an AI tool help us better understand the origins of cancer?
Researchers from EMBL's Korbel Group have developed a new AI method – MAGIC – which, through a game of molecular laser tag, is shedding light on how chromosomal abnormalities form in cells.
www.embl.org/news/science...
Researchers from EMBL's Korbel Group have developed a new AI method – MAGIC – which, through a game of molecular laser tag, is shedding light on how chromosomal abnormalities form in cells.
www.embl.org/news/science...
Reposted by Michael Jendrusch
Is 3D dragging you down? Wish you could instead use the 2D ColabFold representation for all your work? 🤓
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
October 29, 2025 at 1:39 AM
Is 3D dragging you down? Wish you could instead use the 2D ColabFold representation for all your work? 🤓
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
Reposted by Michael Jendrusch
New review on experimental datasets that can be used to benchmark protein force fields. And if that doesn’t tickle your fancy, the data can also be used to benchmark machine learning models for biomolecular structure and dynamics.
In the latest @livecomsjournal.bsky.social
perpetual review, Cavender et al overview NMR and crystallographic experimental datasets that can be used to benchmark protein force fields, including best practices for setup and analysis of simulations!:
livecomsjournal.org/index.php/li...
#compchem
perpetual review, Cavender et al overview NMR and crystallographic experimental datasets that can be used to benchmark protein force fields, including best practices for setup and analysis of simulations!:
livecomsjournal.org/index.php/li...
#compchem
October 28, 2025 at 3:51 PM
New review on experimental datasets that can be used to benchmark protein force fields. And if that doesn’t tickle your fancy, the data can also be used to benchmark machine learning models for biomolecular structure and dynamics.
Reposted by Michael Jendrusch
Our colleagues Vinith Kishore and Valentin Debarnot from the @ivandokmanic.bsky.social lab have come up with an amazing deep learning tool for denoising and filling the missing wedge in #cryoET data. I'm pleased to introduce Icecream🍧
Icecream: High-Fidelity Equivariant Cryo-Electron Tomography
Cryo-electron tomography (cryo-ET) visualizes 3D cellular architecture in near-native states. Recent deep-learning methods (CryoCARE, IsoNet, DeepDeWedge, CryoLithe) improve denoising and artifact cor...
www.biorxiv.org
October 23, 2025 at 11:13 AM
Our colleagues Vinith Kishore and Valentin Debarnot from the @ivandokmanic.bsky.social lab have come up with an amazing deep learning tool for denoising and filling the missing wedge in #cryoET data. I'm pleased to introduce Icecream🍧
Reposted by Michael Jendrusch
**Job Alert** Exciting postdoc position available: theoretical and experimental cryo-EM studies of flexible biomolecules. Competitive salary, collaborative environment at NYSBC and Flatiron Institute. Please share!! and contact: [email protected]
October 22, 2025 at 2:55 PM
**Job Alert** Exciting postdoc position available: theoretical and experimental cryo-EM studies of flexible biomolecules. Competitive salary, collaborative environment at NYSBC and Flatiron Institute. Please share!! and contact: [email protected]
Reposted by Michael Jendrusch
Reposted by Michael Jendrusch
Opportunity in computational structural biology at ISTA (Vienna Austria) duration: 5 years with possibility of permanent contract #StructBiol #cryoEM #cryoET #chemsky 🧪
ista.ac.at/en/job/tenur...
ista.ac.at/en/job/tenur...
Tenure-based Research Staff Scientist Position in Computational Structural Biology
ISTA (www.ista.ac.at) is seeking a highly skilled Staff Scientist in Computational Structural Biology to support and advance the Institute’s structural biology research through the development and imp...
ista.ac.at
October 16, 2025 at 6:30 AM
Opportunity in computational structural biology at ISTA (Vienna Austria) duration: 5 years with possibility of permanent contract #StructBiol #cryoEM #cryoET #chemsky 🧪
ista.ac.at/en/job/tenur...
ista.ac.at/en/job/tenur...
Reposted by Michael Jendrusch
Check out our preprint! With new molecular mechanisms, 140 subtomogram averages, and ~600 annotated cells under different conditions, we @embl.org were able to describe bacterial populations with in-cell #cryoET. And there’s a surprise at the end 🕵️
www.biorxiv.org/content/10.1...
#teamtomo
www.biorxiv.org/content/10.1...
#teamtomo
October 15, 2025 at 6:26 AM
Check out our preprint! With new molecular mechanisms, 140 subtomogram averages, and ~600 annotated cells under different conditions, we @embl.org were able to describe bacterial populations with in-cell #cryoET. And there’s a surprise at the end 🕵️
www.biorxiv.org/content/10.1...
#teamtomo
www.biorxiv.org/content/10.1...
#teamtomo
Reposted by Michael Jendrusch
I am thrilled to release ProteinDJ: a high-performance and modular protein design pipeline. Our open-source workflow incorporates #RFdiffusion, #ProteinMPNN, #FAMPNN, #AlphaFold2 and #Boltz-2. It is a fast, free, and fun way to design proteins (1/5)
doi.org/10.1101/2025.09.24.678028 #proteindesign
doi.org/10.1101/2025.09.24.678028 #proteindesign
September 28, 2025 at 9:16 PM
I am thrilled to release ProteinDJ: a high-performance and modular protein design pipeline. Our open-source workflow incorporates #RFdiffusion, #ProteinMPNN, #FAMPNN, #AlphaFold2 and #Boltz-2. It is a fast, free, and fun way to design proteins (1/5)
doi.org/10.1101/2025.09.24.678028 #proteindesign
doi.org/10.1101/2025.09.24.678028 #proteindesign
Reposted by Michael Jendrusch
🚨 New Web Resource Alert! 🚨 We're delighted to share Viro3D a database of >85000 viral protein structure predictions from >4400 human & animal viruses.
🔗 viro3d.cvr.gla.ac.uk
📄 www.embopress.org/doi/full/10....
@molsystbiol.org @cvrinfo.bsky.social @uofgmvls.bsky.social #Virology #AlphaFold 🧪 🦠
🔗 viro3d.cvr.gla.ac.uk
📄 www.embopress.org/doi/full/10....
@molsystbiol.org @cvrinfo.bsky.social @uofgmvls.bsky.social #Virology #AlphaFold 🧪 🦠
Viro3D
viro3d.cvr.gla.ac.uk
September 26, 2025 at 11:36 AM
🚨 New Web Resource Alert! 🚨 We're delighted to share Viro3D a database of >85000 viral protein structure predictions from >4400 human & animal viruses.
🔗 viro3d.cvr.gla.ac.uk
📄 www.embopress.org/doi/full/10....
@molsystbiol.org @cvrinfo.bsky.social @uofgmvls.bsky.social #Virology #AlphaFold 🧪 🦠
🔗 viro3d.cvr.gla.ac.uk
📄 www.embopress.org/doi/full/10....
@molsystbiol.org @cvrinfo.bsky.social @uofgmvls.bsky.social #Virology #AlphaFold 🧪 🦠
Reposted by Michael Jendrusch
Viro3D paper is out! We predicted 85,000 protein structures from human & animal viruses. 1/5 🧵
📑 Paper doi.org/10.1038/s443...
🔭 Explore virosphere viro3d.cvr.gla.ac.uk
📑 Paper doi.org/10.1038/s443...
🔭 Explore virosphere viro3d.cvr.gla.ac.uk
Viro3D: a comprehensive database of virus protein structure predictions | Molecular Systems Biology
imageimageViro3D provides proteome-level, high confidence AI-protein structure predictions for
>4,400 viruses, allowing mapping of form and function across the human and animal
virosphere. Viro3D i...
doi.org
September 26, 2025 at 2:06 PM
Viro3D paper is out! We predicted 85,000 protein structures from human & animal viruses. 1/5 🧵
📑 Paper doi.org/10.1038/s443...
🔭 Explore virosphere viro3d.cvr.gla.ac.uk
📑 Paper doi.org/10.1038/s443...
🔭 Explore virosphere viro3d.cvr.gla.ac.uk
With this, the last bit of my PhD at @embl.org is finally out!
We developed salad (sparse all-atom denoising), a family of blazing fast protein structure diffusion models.
Paper: nature.com/articles/s42256-…
Code: github.com/mjendrusch/salad
Data: zenodo.org/records/14711580
1/🧵
We developed salad (sparse all-atom denoising), a family of blazing fast protein structure diffusion models.
Paper: nature.com/articles/s42256-…
Code: github.com/mjendrusch/salad
Data: zenodo.org/records/14711580
1/🧵
‘Salad’ – a new AI model from EMBL scientists – offers major improvements in synthetic protein design.
Salad is significantly faster than comparable methods, and designing proteins that don't exist in nature can have applications in many scientific fields.
www.nature.com/articles/s42...
Salad is significantly faster than comparable methods, and designing proteins that don't exist in nature can have applications in many scientific fields.
www.nature.com/articles/s42...
September 24, 2025 at 12:27 PM
With this, the last bit of my PhD at @embl.org is finally out!
We developed salad (sparse all-atom denoising), a family of blazing fast protein structure diffusion models.
Paper: nature.com/articles/s42256-…
Code: github.com/mjendrusch/salad
Data: zenodo.org/records/14711580
1/🧵
We developed salad (sparse all-atom denoising), a family of blazing fast protein structure diffusion models.
Paper: nature.com/articles/s42256-…
Code: github.com/mjendrusch/salad
Data: zenodo.org/records/14711580
1/🧵
Reposted by Michael Jendrusch
Part of something that we have been working on for a while-A 3.1 Å map of a flexibly tethered 38 kDa domain. Hopefully preprinted soon along with a few other structures. #CryoEM
September 21, 2025 at 3:02 PM
Part of something that we have been working on for a while-A 3.1 Å map of a flexibly tethered 38 kDa domain. Hopefully preprinted soon along with a few other structures. #CryoEM
Reposted by Michael Jendrusch
A molecular-resolution look into the near-native architecture of the spinach chloroplast🌱. This one was a long time in the oven, but we're happy to finally share our "version of record". What long-standing debates did we settle? Check back for a short thread🧵 on Monday. #TeamTomo #PlantScience 🧪🧶🧬🔬🌾
🌱 Using ‘compelling’ methods, including #CryoET, researchers mapped spinach thylakoid membranes at single-molecule precision, revealing how photosynthetic complexes are organised and settling long-standing debates on chloroplast architecture.
buff.ly/j3TSIkn
buff.ly/j3TSIkn
September 20, 2025 at 7:24 PM
A molecular-resolution look into the near-native architecture of the spinach chloroplast🌱. This one was a long time in the oven, but we're happy to finally share our "version of record". What long-standing debates did we settle? Check back for a short thread🧵 on Monday. #TeamTomo #PlantScience 🧪🧶🧬🔬🌾
Reposted by Michael Jendrusch
MMseqs2-GPU sets new standards in single query search speed, allows near instant search of big databases, scales to multiple GPUs and is fast beyond VRAM. It enables ColabFold MSA generation in seconds and sub-second Foldseek search against AFDB50. 1/n
📄 www.nature.com/articles/s41...
💿 mmseqs.com
📄 www.nature.com/articles/s41...
💿 mmseqs.com
GPU-accelerated homology search with MMseqs2 - Nature Methods
Graphics processing unit-accelerated MMseqs2 offers tremendous speedups for homology retrieval from metagenomic databases, query-centered multiple sequence alignment generation for structure predictio...
www.nature.com
September 21, 2025 at 8:06 AM
MMseqs2-GPU sets new standards in single query search speed, allows near instant search of big databases, scales to multiple GPUs and is fast beyond VRAM. It enables ColabFold MSA generation in seconds and sub-second Foldseek search against AFDB50. 1/n
📄 www.nature.com/articles/s41...
💿 mmseqs.com
📄 www.nature.com/articles/s41...
💿 mmseqs.com