Francesca Cuturello

Francesca Cuturello

@fra-cutu.bsky.social

Computational perspective on molecular evolution & function @areasciencepark

32 Followers 55 Following 1 Posts Joined Jan 2025
6 months ago
Preview
Evolutionary Constraints Guide AlphaFold2 in Predicting Alternative Conformations and Inform Rational Mutation Design Investigating structural variability is essential for understanding protein biological functions. Although AlphaFold2 accurately predicts static structures, it fails to capture the full spectrum of f...

Our study is now published on JCIMπŸŽ‰

We expanded and refined the preprint thanks to the insightful feedback from reviewers!

paper: pubs.acs.org/doi/10.1021/...
code: github.com/RitAreaScien...

10 4 0 1
8 months ago
Preview
AI in Virology: Leveraging AI to Advance Our Understanding of Viruses - Area Science Park Artificial intelligence is opening new perspectives in virus research, providing advanced tools to analyze viral evolution and to understand protein structure and dynamics on a large scale. To explore...

Workshop not to be missed at Area Science Park (Trieste) on the 9th of September.
Special Guests: David Robertson (UK), @grovearmada.bsky.social (UK), Emanuele Andreano (IT)

www.areasciencepark.it/en/events/ai...

2 2 0 0
9 months ago

πŸ”₯ Two PhD positions open @UniTrieste funded by @AreaSciencePark!Β πŸ”₯

Join the Laboratory of Data Engineering to advance research in AI and its scientific applications.

We’re looking for motivated students ready to dive into interdisciplinary research in deep learning and AI.

3 2 2 0
9 months ago

Our integrative approach using #MD and #cryoEM data to construct structural ensembles of #RNA just published on @natcomms.bsky.social doi.org/10.1038/s414... Lead by Elisa Posani, with @magistratolab.bsky.social @bonomimax.bsky.social @pjanos.bsky.social @navtejtoor.bsky.social and Daniel Haack πŸŽ‰

26 8 0 0
9 months ago
Preview
A protein language model for exploring viral fitness landscapes - Nature Communications Ito et al. present CoVFit, an AI model that predicts variant fitness (transmissibility) from spike protein sequences alone. They further demonstrate its utility in forecasting viral evolution via sing...

The CoVFit paper is now published @natcomms.nature.com !! Exciting work led by Jumpei Ito really showing the potential of pLMs for genotype-2-phenotype predictions!!

www.nature.com/articles/s41...

14 5 1 0
10 months ago

🚨 New paper! By combining protein language models, clustering, and DCA, we reveal how MSA subclusters drive AF2 predictions toward alternative conformations, and how statistics from the clustered sequences can inform deisgn of mutations that shift conformer populations.

10 4 0 0
11 months ago
RAG-ESM logo

πŸ“’ Our new preprint is out on bioRxiv! We introduce RAG-ESM, a retrieval-augmented framework that improves pretrained protein language models like ESM2 by making them homology-aware with minimal additional training costs.
πŸ”— doi.org/10.1101/2025...
πŸ’» github.com/Bitbol-Lab/r...

1/7

5 3 1 1
1 year ago
GitHub - microsoft/bioemu-benchmarks: Benchmarking code accompanying the release of `bioemu` Benchmarking code accompanying the release of `bioemu` - microsoft/bioemu-benchmarks

Today we have published BioEmu-Benchmarks (MIT license): a code to evaluate the multi-conformation sampling benchmarks, MD free energy landscape benchmarks, and folding free energy benchmarks shown in the BioEmu-1 paper with BioEmu or your own model. Some details below 🧡

github.com/microsoft/bi...

76 21 4 0
1 year ago
Preview
Enhancing predictions of protein stability changes induced by single mutations using MSA-based language models AbstractMotivation. Protein language models offer a new perspective for addressing challenges in structural biology, while relying solely on sequence infor

I agree. Also, performance metrics in the field can be misleading due to data leakage. Take a look at our work! doi.org/10.1093/bioi...

1 0 0 0
1 year ago
Preview
Rapid and sensitive protein complex alignment with Foldseek-Multimer - Nature Methods Foldseek-Multimer offers a fast strategy for complex-to-complex alignment to quickly identify compatible sets of chain-to-chain alignments by their superpositions. It can compare billions of complex pairs in 11 h.

Rapid and sensitive protein complex alignment with Foldseek-Multimer | Nature Methods https://www.nature.com/articles/s41592-025-02593-7

14 6 0 0
1 year ago
Figs two and three from the paper showing a variety of fine-tuning schemes across various base models, none of which exceed PPI prediction accuracy of 0.65

PPI prediction accuracy doesn't appear to ever exceed ~0.65 when using embeddings from sequence-only protein language models, regardless of model size or fine-tuning scheme www.biorxiv.org/content/10.1...

26 7 2 1
1 year ago
Preview
Multiple Protein Structure Alignment at Scale with FoldMason bioRxiv - the preprint server for biology, operated by Cold Spring Harbor Laboratory, a research and educational institution

FoldMason progressively aligns thousands of protein structures in seconds, enabling remote MSA for distant phylogeny. Highlights: structural flexible MSA, LDDT conservation score, friendly webserver
πŸ’Ύ github.com/steineggerla...
🌐 search.foldseek.com/foldmason
πŸ“„ www.biorxiv.org/content/10.1...

13 6 0 0
1 year ago

Nice day at Area Science Park! It was beautiful to see what previous students of our group are doing @fra-cutu.bsky.social @piompons.bsky.social

1 1 0 0