Liwei Chang

That’s an easy one to reach agreement on;) If I can add more, they also trained on distributions simulated at melting temperatures but try to predict room temperature conformations.

Alanine dipeptide should be updated to fast folding proteins at this point in time.

Sharing here as well — consider adding this to your or your fav llm’s reading list if you want a dive into 1) protein folding, 2) limitation of AlphaFold and what it learned. You could be one step closer to solve the protein folding problem. www.nature.com/articles/s41...

also faster to hit cluster center of bowls with extended similarity..

was standing on the other side of fig. 1:)

Similarly in predicted folding pathways, the C-termini hairpin of protein G doesn’t change structurally, but the confidence scores vary when surroundings fold closer to native.

I think big part of what AF learned is mimicking the distance distribution of arbitrary pairwise residues/tokens in input from PDB. It changes with environment, adding more sequences affects structure predictions, and scores, which correlate well with the variance in predicted distograms.

It was a surprise to see AF2 can predict folding pathways, but the scores keep increasing from unfolded to folded, unlike (real!) free energy map.

That would be great! But I haven’t found strong evidence for that.. At least the confidence score doesn’t look like a good estimation for free energy from this work. www.biorxiv.org/content/10.1...

How does ramen assemble in warming soup — the ultimate scientific challenge 🍜

- Entropy-driven noodle arrangement

- Hydration kinetics

- Diffusion and buoyancy activated topping sorting
…

An Evaluation of Biomolecular Energetics Learned by AlphaFold
by Lyu, Herschlag et al
doi.org/10.1101/2025...

Interesting comparison of AF2/3 structures with the PDB (AF2 appears better)

Reminder: The PDB itself is probably biased relative to solution conformational ensembles and energetics

LEGOLAS: a Machine Learning method for rapid and accurate predictions of protein NMR chemical shifts. This work introduces LEGOLAS, a fully open source TorchANI-based neural network model designed to predict NMR chemical shifts for protein backbone atoms. LEGOLAS has been designed to be fast, and with...

Ever wanted to predict protein NMR chemical shifts accurately and very very fast ? Great new paper from MIkailya Darrows @mdarrows.bsky.social. chemrxiv.org/engage/chemr...

Comments are welcome !

For those following the @prof-ajay-jain.bsky.social /Cleaves/ @wpwalters.bsky.social preprint re: DiffDock @gcorso.bsky.social

I have read some of the back-&-forth between the author groups

As a practitioner in the field for > 20 yrs (academic side), here is my take: 🧵

Super excited to preprint our work on developing a Biomolecular Emulator (BioEmu): Scalable emulation of protein equilibrium ensembles with generative deep learning from @msftresearch.bsky.social ch AI for Science.

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