Alberto Borsatto

How can we identify cryptic binding sites that are not visible in the structure of a protein in its ground state? 🧐 In our new Sci Adv paper, we used ML and fine-tuned a PLMs to identify and predict these cryptic sites directly from sequences. science.org/doi/10.1126/sc…

New OpenFold3 preview out! (OF3p2) It closes the gap to AlphaFold3 for most modalities. Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9

Absolute Binding Free Energies with OneOPES 1/ This paper introduces OneOPES, a transferable enhanced sampling strategy for accurately calculating absolute binding free energies (ABFEs) in protein-ligand complexes. 2/ OneOPES is designed to address the challenges of ABFE calculations, achieving a mean unsigned error within 1 kcal/mol of experimental free energies, tested on BRD4 and Hsp90 protein targets with 17 diverse ligands. 3/ OneOPES uses a simple set of geometric collective variables (CVs) that are adaptable to various systems without needing tailored adjustments, making it highly versatile for drug discovery applications. 4/ A key innovation is its ability to sample different ligand binding modes and accurately predict protein-ligand interactions regardless of the initial configuration, showing promise for lead optimization in drug discovery. 5/ The OneOPES strategy combines multiple enhanced sampling techniques (e.g., OPES Explore and replica exchange) to reduce dependence on optimal CVs, significantly accelerating the convergence of free energy calculations. 6/ The method successfully integrates hydration and torsional CVs, improving sampling efficiency for complex ligand binding/unbinding mechanisms and providing reliable free energy estimates in less than 700 ns per replica. 7/ This strategy is particularly useful for exploring systems with multiple ligand binding poses and cryptic binding sites, offering insights into ligand binding free energy profiles along physical association pathways. @M_Karrenbrock @albertobors @ValerioRizzi33 📜Paper: doi.org/10.1021/acs.jp…

Any feedback is welcome! If you'd like to try it out, you can find example input files here: zenodo.org/records/136200… 5/🧵

Not trivial: if we run 3 independent simulations of the same system, all estimates independently converge to the same ABFE value! 4/🧵

Looking for an accurate and transferable strategy to calculate protein-ligand absolute binding free energies? Check out our latest paper now published in @JPhysChem with @M_Karrenbrock @ValerioRizzi33 @D_Lukauskis Simone Aureli @gervasiolab doi.org/10.1021/acs.jp… 1/🧵

Excited to see our work on accelerating DeepLNE++ published in @JChemPhys at doi.org/10.1063/5.0226… The updated link for tutorials: github.com/ThorbenF/DeepL… Many thanks to @ValerioRizzi33, Simone Aureli, and the entire @gervasiolab. Summary below: x.com/TFroehlking/st…

Our paper with @TFroehlking @ValerioRizzi33 & @LuigiBonati on learning path-like collective variables with autoencoders is now out in @JChemPhys ! This is the latest step in a journey that began with "From A to B in Free Energy Space". pubs.aip.org/aip/jcp/articl…

@gervasiolab @GianquintoE @ValerioRizzi33 GitHub tutorial at: github.com/Gervasiolab/Ge…

SWISH-X recovers crystal-like configurations of the various cryptic sites in significantly shorter simulation times compared to our SWISH and mixed solvent simulations. Check out the tutorial : github.com/Gervasiolab/Ge… by @albertobors, @GianquintoE, @ValerioRizzi33 3 /🧵

Our new strategy for cryptic pocket detection, SWISH-X is out in @JCIM_JCTC! doi.org/10.1021/acs.jc… It combines SWISH's Hamiltonian replica exchange with expanded ensemble sampling with OPES to speed up cryptic pocket exploration. 1 / 🧵

1/n: We are excited to share that our paper on Chroma, a general purpose diffusion model for proteins, is out today in @Nature! nature.com/articles/s4158… A couple of my favorite highlights in the 🧵below 👇

@pedrojuanbj @gervasiolab @cecamEvents Thanks Pedro!

@gervasiolab @albertobors @cecamEvents Congratulations 👏🎉!!!! @albertobors

Congratulations to @albertobors for winning the best poster at the @cecamEvents workshop on Making the invisible protein life visible!

@pedrojuanbj Bravo Pedro! Congrats :)

I did a thing!.....Two weeks ago!...but still hasn't kicked in yet! Passed my viva and celebrated accordingly! 🎓 🎉🥳🎉🥳🥂🍻 Thank you to everyone I met on the way! 🎓🎓🥰🥰 And onto the next thing! 💪🏻💪🏻💪🏻

The group enjoying the SBDD23 conference in beautiful Sestri Levante. Congratulations to @devivo_marco @matteodp and Andrea Cavalli on organising such a successful conference! #SBDD23

New enhanced sampling approach from @gervasiolab combining the On-the-fly Probability Enhanced Sampling (OPES) approach with replica-exchange enhancements. Tested on alaninedipeptide, trypsin/benzamidine binding and chignolin folding. biorxiv.org/content/10.110…

Congratulations to @D_Lukauskis, Simone Aureli and our collaborators at @ucb_news on the publication of the paper on Open BPMD in @JCIM_JCTC. The approach is fast and effective in ranking ligand binding poses. The code is available on @github pubs.acs.org/doi/10.1021/ac…

Our article "Revealing druggable cryptic pockets in the Nsp1 of SARS-CoV-2 and other β-coronaviruses by simulations and crystallography", a successful collaboration with @c3starellas @shozeb_haider @albertobors @igaldadas and F Kozielski is out in @eLife doi.org/10.7554/eLife.…

A team from @unige_en , in collaboration with @ucl and @UniBarcelona, reveals a hidden cavity on a key #COVID19 #protein to which future drugs could bind. @albertobors ow.ly/ef0t50LJj6g