Michele Invernizzi

Hope you'll like it, we tried to meticulously include all the details that can be useful to practitioners

Sampling IDPs is tough, but can be worth the effort!

OPES multithermal made it a little easier for us, hopefully you'll find it useful as well

Third preprint of the year is from @julianstreit.bsky.social who, with our collaborators at Peptone, show that multithermal On-the-fly Probability Enhanced Sampling (OPES) enables efficient generation of atomistic ensembles for disordered peptides and proteins 🍝

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

We (@sobuelow.bsky.social) developed AF-CALVADOS to integrate AlphaFold and CALVADOS to simulate flexible multidomain proteins at scale

See preprint for:
— Ensembles of >12000 full-length human proteins
— Analysis of IDRs in >1500 TFs

📜 doi.org/10.1101/2025...
💾 github.com/KULL-Centre/...

Machine Learning Research Engineer at Peptone Our Computational Sciences group is a team of machine-learning engineers, computational chemists, and drug-discovery scientists. We build in-silico tools that let our Medicinal-Chemistry and Biology c...

jobs.peptone.io/p/c2b34ca3dd...

Overlayed configurations of a small molecule binding (?) to an unstructured protein

We have a job opening at Peptone.io for an ML researcher.

Come help us find new ways to understand and drug intrinsically disordered proteins (IDPs), it's a very interesting and important problem!

Link in the reply ↓

Everything everywhere all at once: a probability-based enhanced sampling approach to rare events - Nature Computational Science A single semi-automatic enhanced sampling method for rare events, based on machine-learned committor functions, allows simultaneous sampling of reactive events, calculation of free energy and understa...

Very interesting and powerful method!

www.nature.com/articles/s43...

The plan at FutureHouse has been to build scientific agents for discoveries. We’ve spent the last year researching the best way to make agents. We’ve made a ton of progress and now we’ve engineered them to be used at scale, by anyone. Free and on API.

Presenting one of my favorite manuscripts I've ever worked on:

"Characterizing structural and kinetic ensembles of intrinsically disordered proteins using writhe"

www.biorxiv.org/content/10.1...

by Tommy Sisk, with a generative modeling component done in collaboration with @smnlssn.bsky.social

Thanks for sharing!

"De novo prediction of protein structural dynamics"

I'll be presenting an overview of the field tomorrow at a workshop. Link to a PDF copy of the presentation: delalamo.xyz/assets/post_...

Presenting our work on minimum energy path generation between two states for physical systems at the FPI Workshop at @ICLR tomorrow! Scaling up to solvated BPTI and observing the same conformational changes as long reference MD with six orders fewer force field evals! Drop by!

Kinetic rates calculation via non-equilibrium dynamics This study introduces a novel computational approach based on ratchet-and-pawl molecular dynamics (rMD) for accurately estimating ligand dissociation kinetics in protein-ligand complexes. By integrati...

New preprint on arXiv! We propose a new technique to compute kinetic rates using multiple independent non-equilibrium (ratchet&pawl MD) simulations!
We focused here on ligand unbinding kinetics, but this method can be applied to any situation where a reaction coordinate can be defined!

AlphaFold is amazing but gives you static structures 🧊

In a fantastic teamwork, @mcagiada.bsky.social and @emilthomasen.bsky.social developed AF2χ to generate conformational ensembles representing side-chain dynamics using AF2 💃

Code: github.com/KULL-Centre/...
Colab: github.com/matteo-cagia...

BoltzDesign1: Inverting All-Atom Structure Prediction Model for Generalized Biomolecular Binder Design by @yehlincho.bsky.social @martinpacesa.bsky.social @sokrypton.org 🧶🧬

www.biorxiv.org/content/10.1...

Wondering how to predict protein flexibility in a sec? No time to run MD simulations but want to go beyond pLDDT? Check out BBFlow
arxiv.org/html/2503.05...

Useful in particular for de novo designs.

Led by Nico Wolf & Leif Seute, w Seva, Simon, and Jan. @mpip-mainz.mpg.de @hitsters.bsky.social

Small proteins can be more complex than they look!

We know proteins fluctuate between different conformations- but by how much? How does it vary from protein to protein? Can highly stable domains have low stability segments? @ajrferrari.bsky.social experimentally tested >5,000 domains to find out!

Protein function often depends on protein dynamics. To design proteins that function like natural ones, how do we predict their dynamics?

@hkws.bsky.social and I are thrilled to share the first big, experimental datasets on protein dynamics and our new model: Dyna-1!

🧵

Straight to the reading list:

Training a machine learning model based on residues with missing NMR assignments as a proxy for protein motion

Very excited for my first BSKY post. We present a new method, Loxodynamics, for exploring chemical and catalytic reaction space!

Figure from the paper illustrating sequence–ensemble–function relationships for disordered proteins. ML prediction (black) and design (orange) approaches are highlighted on the connecting arrows. Prediction of properties/functions from sequence (or vice versa, design) can include biophysics approaches via structural ensembles, or bioinformatics approaches via other hetero- geneous sources. The lower panels show examples of properties and functions of IDRs for predictions or design targets. ML, machine learning; IDRs, intrinsically disordered proteins and regions.

Our review on machine learning methods to study sequence–ensemble–function relationships in disordered proteins is now out in COSB

authors.elsevier.com/sd/article/S...
Led by @sobuelow.bsky.social and Giulio Tesei

PLUMED Tutorials: A collaborative, community-driven learning ecosystem In computational physics, chemistry, and biology, the implementation of new techniques in shared and open-source software lowers barriers to entry and promotes

The paper describing our community effort to collect and organize #plumed tutorials has been published in the Journal of Chemical Physics, as part of the Michele Parrinello Festschrift! doi.org/10.1063/5.02...

Illustration showing how in 2025, CADD scientists are forced to use the same published force field model week after week in a manner than cannot learn from new experimental data that contradicts it. In the future (2027?), CADD scientists will be able to make good general predictions with a foundation simulation model, but will be able to fine-tune that model after every new batch of data to deliver systematically more accurate predictions week after week.

As a peek toward where we're headed:

Right now, CADD scientists are forced to use the same model week after week, even if new experimental data says the model is inaccurate.

If we can fine- models, we can exploit that data to systematically improve our predictions week by week!

The BioEmu-1 model and inference code are now public under MIT license!!!

Please go ahead, play with it and let us know if there are issues.

github.com/microsoft/bi...

I am hiring a postdoctoral scholar with a start date summer or fall 2025. Projects will be focused on thermodynamically consistent generative models, broadly defined. If you’re interested, please send a CV and one paragraph about why you think you’d be a good fit to rotskoff@stanford.edu

Mapping Long-Range Interactions in α-Synuclein using Spin-Label NMR and Ensemble Molecular Dynamics Simulations The intrinsically disordered protein α-synuclein plays a key role in the pathogenesis of Parkinson's disease (PD). We show here that the native state of α-synuclein consists of a broad distribution of...

It’s been 20 years today since my first paper on intrinsically disordered proteins

Mapping Long-Range Interactions in α-Synuclein using Spin-Label NMR and Ensemble Molecular Dynamics Simulations
doi.org/10.1021/ja04...

and I thought I would tell the somewhat random path that led to this paper. 1/n

It was a very interesting talk, looking forward for the preprint!

Confronting risks of mirror life Broad discussion is needed to chart a path forward.

With today's report outlining risks on mirror life www.science.org/doi/full/10.... many have asked:

Could mirror life survive in the wild?

Yes. While mirror life in the wild could have some significant disadvantages (like finding food it can digest), they do not appear to be insurmountable: 🧵

In his book “The Nature of Statistical Learning” V. Vapnik wrote:
“When solving a given problem, try to avoid a more general problem as an intermediate step”