Paulo C. T. Souza

Cutaway and close-up views of a Martini coarse-grained whole-cell model of JCVI-syn3A, showing the densely packed cytoplasm with proteins, RNA, metabolites, and chromosome inside a lipid membrane with embedded membrane proteins

Our paper on [Bentopy](doi.org/10.1002/pro....) is out in Protein Science! We developed Bentopy to make assembling large-scale MD models more accessible, building on what we learned from trying to simulate whole-cell models. Here's our updated Martini JCVI-syn3A cell model👇

New preprint: “When lipids embrace RNA”

www.biorxiv.org/content/10.6...

Using multiscale simulations (🍸 #Martini + constant-pH MD), we show that:

• Local pKa ≠ global pKa
• Endosomal escape is limited by persistent protonation.

#LNP #MolecularDynamics

Very happy to see our recent work featured on the cover of the Journal of Chemical Theory and Computation 🎉
A long journey with many people involved — grateful to everyone who helped make this happen.

Paper: pubs.acs.org/doi/10.1021/...

@pubs.acs.org #MyACSCover #JCTC #Martini3

Fast Parametrization of Martini3 Models for Fragments and Small Molecules Coarse-grained molecular dynamics simulations, such as those performed with the recently parametrized Martini 3 force field, simplify molecular models and enable the study of larger systems over longer time scales. With this new implementation, Martini 3 allows more bead types and sizes, becoming more amenable to studying dynamical phenomena involving small molecules such as protein–ligand interactions and membrane permeation. However, while solutions existed to automatically model small molecules using the previous iteration of the Martini force field, there is no simple way to generate such molecules for Martini 3 yet. Here, we introduce Auto-MartiniM3, an advanced and updated version of the Auto-Martini program designed to automate the coarse-graining of small molecules to be used with the Martini 3 force field. We validated our approach by modeling 81 simple molecules from the Martini Database and comparing their structural and thermodynamic properties with those obtained from models designed by Martini experts. Additionally, we assessed the behavior of Auto-MartiniM3-generated models by calculating solute translocation and free energy across lipid bilayers. We also evaluated more complex molecules such as caffeine by testing its binding to the adenosine A2A receptor. Finally, our results from deploying Auto-MartiniM3 on a large data set of molecular fragments demonstrate that this program can become a tool of choice for fast, high-throughput creation of coarse-grained models of small molecules, offering a good balance between automation and accuracy. Auto-MartiniM3 source code is freely available at https://github.com/Martini-Force-Field-Initiative/Automartini_M3.

Our collective work with @tbereau.bsky.social and @pauloctsouza.bsky.social labs to automatically parametrize CG models of small molecules is now published in JCTC! Nice Christmas present for 1st author @szczukam.bsky.social and all the coauthors!

pubs.acs.org/doi/full/10....

Looking forward to #acsfall2025! Hoping to finally meet in person the colleagues I’ve long admired at the
@acscomp.bsky.social reception Sunday as well. 😄🍸

Thanks to sponsors: Google Cloud & MolCube!

📄 Full program: acs.digitellinc.com/live/35/page...

#ACSFall2025 #ComputationalChemistry #CoarseGrainedModeling

🔬 Coarse-Grained Modeling @ ACS Fall 2025 — Program Out Now! 🚀

📍 Washington DC | Aug 17–21

- Wed, Aug 20 AM: LNPs — nucleic acid delivery → viral entry simulations
- Wed, Aug 20 PM: Proteins & multiscale modeling
- Thu, Aug 21 AM: Methods & industry

The Martini 3 Lipidome: Expanded and Refined Parameters Improve Lipid Phase Behavior Lipid membranes are central to cellular life. Complementing experiments, computational modeling has been essential in unraveling complex lipid-biomolecule interactions, crucial in both academia and industry. The Martini model, a coarse-grained force field for efficient molecular dynamics simulations, is widely used to study membrane phenomena but has faced limitations, particularly in capturing realistic lipid phase behavior. Here, we present refined Martini 3 lipid models with a mapping scheme that distinguishes lipid tails that differ by just two carbon atoms, enhancing the structural resolution and thermodynamic accuracy of model membrane systems including ternary mixtures. The expanded Martini lipid library includes thousands of models, enabling simulations of complex and biologically relevant systems. These advancements establish Martini as a robust platform for lipid-based simulations across diverse fields.

Our collaborative work "The #Martini3 #Lipidome: Expanded and Refined Parameters Improve Lipid Phase Behavior" is now published in #ACSCentralScience! 🎉

📄 Read: pubs.acs.org/doi/10.1021/...

💾 GitHub: github.com/Martini-Forc...

#MolecularDynamics #Biophysics #Simulations #Lipids

🚀 New preprint out!

"Fast Parameterization of #Martini3 Models for Fragments and Small Molecules" is now on #bioRxiv

👉 doi.org/10.1101/2025...

#AutoMartiniM3 – tool for automated #CG modeling. With @matthchavent.bsky.social, @tbereau.bsky.social and others.

#CoarseGraining #MD #DrugDiscovery

Very happy to release AutoMartiniM3 for Martini3 to automatize CG modelling of small molecules and fragments.

Great work from M Szczuka in my lab in collab with @pauloctsouza.bsky.social and @tbereau.bsky.social teams!

Give it a try: github.com/Martini-Forc...

poke @cg-martini.bsky.social

A little teaser for next week 😜
poke @pauloctsouza.bsky.social & @tbereau.bsky.social

GōMartini 3: From large conformational changes in proteins to environmental bias corrections - Nature Communications This work presents GōMartini 3, an improved coarse-grained protein model combining physics- and structure-based approaches. It boosts computational efficiency and accuracy for structured soluble and m...

New paper out in Nature Communications from @natureportfolio.nature.com :
We present #GōMartini 3 — combining structure-based Gōwith #Martini3 to better capture #protein conformational changes and environmental effects.

📄 www.nature.com/articles/s41...

🔬Just published in Bioinformatics Advances: "Challenges in predicting PROTAC-mediated protein-protein interfaces with AlphaFold reveal a general limitation on small interfaces": https://doi.org/10.1093/bioadv/vbaf056

#ProteinDesign #AlphaFold #StructuralBiology

@pauloctsouza.bsky.social

🤝 Join Our Sponsors! We already have great support but welcome more! Interested in sponsoring our ACS Fall 2025 symposium on Coarse-Grained Modeling? Contact us:

📩 Jianing Li: jianing-li@purdue.edu
📩 John Shelley: john.shelley@schrodinger.com
📩 Paulo Souza: paulo.telles_de_souza@ens-lyon.fr

ACS Fall - American Chemical Society Join us at ACS Fall 2025 in Washington, DC or online, August 17-21. Choose from thousands of presentations, network, attend courses and visit the expo.

📢 Call for Speakers! Want to present at ACS Fall 2025 in Washington, DC (Aug 17-21)? Submit an abstract for our Coarse-Grained Modeling symposium!

www.acs.org/events/fall....

#ACSFall2025 #CG #MolecularDesign #FormulationDesign

🎉 Big news! 🎉 Together with Jianing Li (Purdue Univ.) & John Shelley (@schrodingerinc.bsky.social), I’m organizing a symposium on #CG #Modeling for Molecular & Formulation Design at ACS Fall 2025 (Aug 17-21, Washington, DC)! 🔬✨ For my part, I am grateful for the support of @ensdelyon.bsky.social!

From my side, thanks for the support of #LBMC (
@lbmcinlyon.bsky.social), #CBPsmn (Centre Blaise Pascal de Simulation et de Modélisation Numérique), #ENSdeLyon (École normale supérieure de Lyon), CNRS (@cnrs.bsky.social), CNRS Délégation Rhône Auvergne and CNRS Biologie (@cnrsbiologie.bsky.social).

Whether you're new to #ComputationalModeling or an expert, this review explores how #MolecularDynamics & #machinelearning advance #API encapsulation, transport, and release—bridging experimental & computational research for better #DrugDelivery.

Computational Methods for Modeling Lipid-Mediated Active Pharmaceutical Ingredient Delivery Lipid-mediated delivery of active pharmaceutical ingredients (API) opened new possibilities in advanced therapies. By encapsulating an API into a lipid nanocarrier (LNC), one can safely deliver APIs n...

🚀 Check out our latest review, "Computational Methods for Modeling Lipid-Mediated Active Pharmaceutical Ingredient Delivery", now published in Molecular Pharmaceutics (@ACS Journals)! 🧪📖

🔗 Read here: doi.org/10.1021/acs....

Systematic Approach to Parametrization of Disaccharides for the Martini 3 Coarse-Grained Force Field Sugars are ubiquitous in biology; they occur in all kingdoms of life. Despite their prevalence, they have often been somewhat neglected in studies of structure–dynamics–function relationships of macro...

So seems like the group is on a roll this month!
Here’s our latest work on parametrising coarse-grained models of sugars (disaccharides) by Astrid Brandner & in collaboration with Iain Smith, @pauloctsouza.bsky.social and @cg-martini.bsky.social pubs.acs.org/doi/10.1021/... #glycotime

Paulo Cesar Telles de Souza on LinkedIn: #chemrxiv #lipidnanoparticles #lnps #mrna #martini3 #coarsegrained #cbpsmn… | 10 comments I’m thrilled to share the preprint of our latest work, “Martini 3: Building Blocks for Lipid Nanoparticle (LNP) Design”, now available on #ChemRxiv! Read the… | 10 comments on LinkedIn

For those who missed it, here’s the original Linkedin post about our work: www.linkedin.com/feed/update/...

After some work, we’ve now made the #GitHub for our #Martini3 #LNP models and protocols publicly available: github.com/Martini-Forc...

We also have a new #ChemRxiv with some minor updates: chemrxiv.org/engage/chemr...

#LipidNanoparticles #MolecularDynamics #DrugDelivery #mRNA #Simulation

GitHub - Martini-Force-Field-Initiative/M3-Lipid-Parameters: Home of the public Martini 3 lipid parameters Home of the public Martini 3 lipid parameters. Contribute to Martini-Force-Field-Initiative/M3-Lipid-Parameters development by creating an account on GitHub.

and get the parameters here: github.com/Martini-Forc...

From my side, thanks for the support of #LBMC (@lbmcinlyon.bsky.social), #CBPsmn (Centre Blaise Pascal de Simulation et de Modélisation Numérique), #ENSdeLyon (École normale supérieure de Lyon), CNRS (@cnrs.bsky.social), CNRS Délégation Rhône Auvergne and CNRS Biologie (@cnrsbiologie.bsky.social).

Highlight 4) Enhanced Phase Behavior: Refined models address previous limitations, capturing phase transitions with greater accuracy.

Highlight 3) Martini Lipid Benchmark: We introduce a benchmarking dataset for lipid properties, a valuable target for all lipid force fields.

Highlight 2) Comprehensive Lipid Library: Thousands of lipid models, spanning numerous lipid classes, open up exciting possibilities for studying biologically relevant systems.

Highlight 1) Improved Structural Resolution: A refined mapping scheme allows distinction between lipid tails differing by just two carbon atoms, boosting both structural and thermodynamic accuracy.

The #Martini3 #Lipidome represents a significant leap forward in molecular simulation of lipids, providing unprecedented structural resolution for Martini lipid models, refinements on bonded terms, an expanded lipid library, and enhanced phase behavior accuracy.

Excited to announce the release of our preprint, "The Martini 3 Lipidome: Expanded and Refined Parameters Improve Lipid Phase Behavior", now available on @chemrxiv.bsky.social

👉 Read the full preprint here: chemrxiv.org/engage/chemr...