Lightnews — Scholar-powered news

Peter Škrinjar @peterskrinjar.bsky.social · Mar 21

Yep, that’s a cool idea. Should be quite easy to try actually

Peter Škrinjar @peterskrinjar.bsky.social · Mar 21

Afaik, not really. Recycling is done before diffusion module and the final output coords are not used for a refinement. Someone correct me if I am wrong though. But seems top method in this challenge could done smth like that: polarishub.io/competitions...

Polaris

The benchmarking platform for drug discovery

polarishub.io

1 2

Reposted by Peter Škrinjar

Torsten Schwede @torstenschwede.bsky.social · Feb 26

🧪🧬🔬 Postdoc in Computational Structural Biology at the @biozentrum.unibas.ch & @sib.swiss in Basel, Switzerland.

www.biozentrum.unibas.ch/open-positio...

The position is initially funded for 3 years, possibility to start immediately.

#StructurePrediction, #Bioinformatics, #Uniprot3D, #AI

Open Postdoc Position at the schwede lab, Biozentrum Basel.

1 18 20

Peter Škrinjar @peterskrinjar.bsky.social · Feb 24

In case you missed it!

Polaris @polarishub.io · Feb 24

Thanks to everyone who joined us last week! You can now watch the recording of the webinar: youtu.be/_ykP_tEVrOk

3 12

Reposted by Peter Škrinjar

Janani Durairaj (Jay) @ninjani.bsky.social · Feb 21

Some very cool stuff from the group now in paper form:

“ModelArchive: a deposition database for computational macromolecular structural models”
doi.org/10.1016/j.jm...

And

“Comparing macromolecular complexes - a fully automated benchmarking suite” doi.org/10.21203/rs.... (under review)

Redirecting

doi.org

10 29

Reposted by Peter Škrinjar

Janani Durairaj (Jay) @ninjani.bsky.social · Feb 20

Join us tomorrow to hear @peterskrinjar.bsky.social talk about 🌹 Runs N' Poses 🌹! Curious to hear your thoughts on how to keep up with benchmarking in this field.

Polaris @polarishub.io · Feb 13

We’re hosting @peterskrinjar.bsky.social for a webinar to discuss his latest paper “Have protein co-folding methods moved beyond memorization?”!

Join live on Feb 21st at 11 AM ET. Register here: us06web.zoom.us/webinar/regi...

Explore Runs N’ Poses on Polaris today: polarishub.io/benchmarks/p...

4 13

Reposted by Peter Škrinjar

Gonzalo Parra @gonzaparra.bsky.social · Feb 13

RT please: Just 1.5 months away from our joint 3DBioinfo/3DSIG amazing conference!
These are our two great keynote speakers!
@rolanddunbrack.bsky.social & @silviaosuna.bsky.social!

Check our preliminary programme here:
www.iscb.org/3dbioinfo202...
Registration link:
www.iscb.org/3dbioinfo202...

1 10 18

Reposted by Peter Škrinjar

Polaris @polarishub.io · Feb 13

We’re hosting @peterskrinjar.bsky.social for a webinar to discuss his latest paper “Have protein co-folding methods moved beyond memorization?”!

Join live on Feb 21st at 11 AM ET. Register here: us06web.zoom.us/webinar/regi...

Explore Runs N’ Poses on Polaris today: polarishub.io/benchmarks/p...

8 16

Reposted by Peter Škrinjar

Radoslav Krivak @rdkbio.bsky.social · Feb 10

How long does it take to predict ligand binding sites for all 220k proteins in the PDB? P2Rank 2.5 does it in 3 hours on a single CPU (16-core amd 5950x)—2x faster than the previous version 🚀.

(1/n)

1 8 39

Peter Škrinjar @peterskrinjar.bsky.social · Feb 12

That being said, today I looked at the templates that AF3 used and actually none of the predictions used ground-truth template or templates after the training cutoff, so was a false alarm luckily. Would still be interesting to check why some predictions improved in the lowest bins.

1 2

Peter Škrinjar @peterskrinjar.bsky.social · Feb 12

It's not expected AF3 models would have perfect fidelity when run with ground-truth templates, especially cause it uses multiple templates and also MSA features. Additionally, only protein part is used from the templates, not the ligand information.

1 3

Peter Škrinjar @peterskrinjar.bsky.social · Feb 11

We re-ran AF3 without templates, since we noticed it could use any template in the PDB, including the ground truth. We see the performance drops slightly in the lowest bins, but the gap to other methods still exists. We will update the preprint shortly!

1 5 20

Reposted by Peter Škrinjar

Polaris @polarishub.io · Feb 10

We’re extremely excited to host Runs N’ Poses on Polaris!

The authors show how current co-folding methods struggle to generalize beyond ligand poses in their training data.

Explore the benchmark: polarishub.io/benchmarks/p...

Explore the dataset: polarishub.io/datasets/pli...

3 7

Reposted by Peter Škrinjar

Cas @caswognum.nl · Feb 8

New leaderboard on @polarishub.io for Runs 'N Poses! 🎸

Anyone has any protein-ligand co-folding methods laying around they would like to put to the test?

polarishub.io/benchmarks/p...

Great work @peterskrinjar.bsky.social @jeeberhardt.bsky.social @torstenschwede.bsky.social @ninjani.bsky.social

1 4 9

Reposted by Peter Škrinjar

Torsten Schwede @torstenschwede.bsky.social · Feb 8

Have #AI methods for protein-ligand co-folding moved beyond memorisation and predict really novel leads for #drugdiscovery? Or do we find “more of the same”? This new benchmark lets you find out… ⬇️⬇️⬇️

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

Excited to share our latest preprint evaluating AlphaFold3, Boltz-1, Chai-1 and Protenix for predicting protein-ligand interactions, featuring our newly introduced benchmark dataset 🌹Runs N’ Poses🌹!

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

🧵👇 (1/n)

Have protein-ligand co-folding methods moved beyond memorisation?

Deep learning has driven major breakthroughs in protein structure prediction, however the next critical advance is accurately predicting how proteins interact with other molecules, especially small mo...

www.biorxiv.org

3 19

Reposted by Peter Škrinjar

Jerome Eberhardt @jeeberhardt.bsky.social · Feb 8

Basically, novelty is not defined by time in the PDB.

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

Excited to share our latest preprint evaluating AlphaFold3, Boltz-1, Chai-1 and Protenix for predicting protein-ligand interactions, featuring our newly introduced benchmark dataset 🌹Runs N’ Poses🌹!

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

🧵👇 (1/n)

Have protein-ligand co-folding methods moved beyond memorisation?

Deep learning has driven major breakthroughs in protein structure prediction, however the next critical advance is accurately predicting how proteins interact with other molecules, especially small mo...

www.biorxiv.org

3 11

Reposted by Peter Škrinjar

Janani Durairaj (Jay) @ninjani.bsky.social · Feb 8

We’ve been benchmarking deep learning co-folding methods for protein-ligand complex prediction, leading to the creation of 🌹Runs N’ Poses🌹. Great effort by @peterskrinjar.bsky.social and @jeeberhardt.bsky.social putting this together so quickly. Please have a look, excited for community feedback!

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

Excited to share our latest preprint evaluating AlphaFold3, Boltz-1, Chai-1 and Protenix for predicting protein-ligand interactions, featuring our newly introduced benchmark dataset 🌹Runs N’ Poses🌹!

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

🧵👇 (1/n)

Have protein-ligand co-folding methods moved beyond memorisation?

Deep learning has driven major breakthroughs in protein structure prediction, however the next critical advance is accurately predicting how proteins interact with other molecules, especially small mo...

www.biorxiv.org

10 25

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

I want to thank my co-authors @jeeberhardt.bsky.social, @torstenschwede.bsky.social, @ninjani.bsky.social and all of our collaborators! RunsN’ Poses builds on PLINDER and OpenStructure—this work wouldn’t be possible without them!
Also thanks to @rokbreznikar.bsky.social for this amazing logo! 9/9

2 1 14

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

We’re pre-printing this early to get feedback from the community. We teamed up with @caswognum.nl at @polarishub.io to make the dataset and benchmark ML-ready polarishub.io/datasets/pli.... See also our github.com/plinder-org/... for more. Would love to hear your thoughts! (8/n

runs-n-poses-dataset

plinder-org

polarishub.io

1 2 15

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

Our findings highlight the need for specialised benchmarks for deep learning in protein-ligand interaction prediction. As tasks like co-folding grow more complex, new metrics are needed to assess leakage and difficulty-best defined by interaction similarity, not protein or ligand similarity. (7/n)

1 10

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

To separate pose prediction from ranking, we analyzed all 25 models per method. Top-ranked models (black) outperform the worst (red) but still lag behind the best possible (blue). Success rates still correlate with training similarity, highlighting a key limitation of these methods. (6/n)

1 8

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

The largest cluster has 171 SARS-CoV-2 MPro X-ray structures with different small molecules. Despite explicit chirality in the input, all methods misoriented the 5-chlorobenzofuran-3-aminomethyl group, perhaps mimicking the 5-chloro-2-methoxyphenyl group from the closest training system. (5/n)

1 8

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

To ensure the observed trend is not driven by an over-representation of certain protein families, we also analyzed only cluster representatives to account for potential biases. The observed correlation doesn’t change. (4/n)

1 8

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

Interestingly, we found that ligands with abundant data in PDB, like cofactors, show better prediction performance. When excluding them to focus on drug-like ligands, the correlation between success rate and training similarity becomes even more linear. (3/n)

1 10

Peter Škrinjar @peterskrinjar.bsky.social · Feb 8

In this work, we explored how training data similarity impacts protein-ligand prediction accuracy—an overlooked aspect in recent benchmarks. Our analysis shows that the current co-folding methods struggle to generalize beyond ligand poses in their training data.(2/n)

1 4 21