Daniel Alvarez Salmoral

PhD Candidate at @NKI_nl | Perrakis lab |
#StructuralBiology | #Bioinformatics

[Daniel Alvarez Salmoral @AlvarezSalmoral]

🚀 Excited to share my first preprint on bioRxiv:[https://t.co/6lhGuCFT30…]

AlphaBridge simplifies protein structure predictions for experts and novices by combining complex analysis with visual representations and interactive tools in one user-friendly platform.

[Bjørn Kwee @Bjornformatics]

It's not the models, it's the data!

We show that a substantial proportion of a widely used TCR-pMHC database does not functionally validate, causing underestimation of performance of TCR-pMHC prediction models.

[Daniel Alvarez Salmoral @AlvarezSalmoral]

Thrilled to present Alphabridge at #NWOChains! We aim to facilitate researchers to analyze interacting interfaces in predicted biomolecular structures. Honored to share insights with top chemists in the Netherlands. 🚀🔬
Preprint: https://t.co/SCK9TR5QJd

[Daniel Alvarez Salmoral @AlvarezSalmoral]

@Jiaxing_Tan_ @rborza_ Thanks for letting me know!! It was something I might be afraid it could happen. As soon as i can I will make the code also compatible with locally installed AF3 output.

[Chubby♨️ @kimmonismus]

We were so blind. It was right before our eyes. The path to AGI was always so obvious.

[johnparkhill @j0hnparkhill]

I wanna talk to the people who optimized the AF3 distogram hyperparameters to the fourth digit 😅

[Liam Bai @liambai21]

Ever wondered how a protein language model sees your favorite protein?

Checkout out our SAE visualizer where you can now search any sequence for activating features.

[Liam Bai @liambai21]

Remember Golden Gate Claude?

@etowah0 and I have been working on applying the same mechanistic interpretability techniques to protein language models.

We found lots of features and they’re... pretty weird?

🧵

[Daniel Alvarez Salmoral @AlvarezSalmoral]

Presenting my poster on AlphaBridge at the annual @oncodeinstitute meeting was a fantastic opportunity! We're now working on a web server update and will incorporate all the feedback shared. Any other ideas or suggestions? Let us know!

📜 Article: https://t.co/SCK9TR6oyL

[Daniel Alvarez Salmoral @AlvarezSalmoral]

@CbirtDirector Thanks for sharing our work!!! Here is the link for our web server in case anyone wants to try it out https://t.co/Mi09bDcDrX

[Razvan Borza @rborza_]

🚨Identify confident Protein-Protein/DNA/RNA Interfaces!!

🔧Our tool highlight the most reliable interaction sites, converting AlphaFold3 predictions into intuitive 3D models and clear diagrams

🤔How many interfaces will your prediction show?
📢Try it and share your feedback!

[Razvan Borza @rborza_]

🚀 Excited about our pre-print:
AlphaBridge: A User-Friendly Tool for Interpreting Protein Predictions

Whether you're an expert or just new to the field of Protein Prediction, AlphaBridge is the way to go!

🔁Try it and share your thoughts!

📜 Article: https://t.co/cH0BL8OxpX

[Biology+AI Daily @BiologyAIDaily]

AlphaBridge: tools for the analysis of predicted macromolecular complexes

• AlphaBridge introduces tools to analyze and visualize the interfaces between predicted macromolecular complexes using data from AlphaFold3. It processes the predicted interactions into intuitive diagrams and 3D models for easy interpretation.

• The tool uses a novel “Predicted Merged Confidence” (PMC) matrix, which combines the Predicted Local Distance Difference Test (pLDDT) and Predicted Aligned Error (PAE) metrics to identify multi-component structural modules and binary interfaces with high confidence.

• AlphaBridge uses graph-based community clustering to detect interaction motifs in biomolecular complexes, then visualizes these interactions as chord diagrams (“AlphaBridge diagrams”) to provide a clear and intuitive summary of predicted binding interfaces.

• The tool allows researchers to evaluate the reliability of interactions in predicted protein-protein, protein-DNA, and protein-RNA complexes, offering a scoring and ranking system for predicted interfaces before detailed analysis.

• By varying the cut-off for contact predictions, AlphaBridge can help distinguish between reliable and unreliable predictions, providing insights into physiological relevance. It was particularly useful for evaluating interactions in DNA mismatch repair proteins and RNA binding proteins.

• The AlphaBridge webserver (https://t.co/HMxfguvmnn) offers an interactive platform where users can upload AlphaFold3 prediction files, visualize the predicted 3D structures, and analyze contact interfaces through an integrated web viewer.

• AlphaBridge enhances accessibility for both expert and non-expert users by combining multiple layers of information—contact links, prediction confidence, and visual representations—into one interactive diagram, simplifying the evaluation of complex predictions.

@TassosPerrakis @rborza_
💻Code: https://t.co/pwKfDtQ7MM
📜Paper: https://t.co/EB8po1CCfP

[bioRxiv Bioinfo @biorxiv_bioinfo]

AlphaBridge: tools for the analysis of predicted macromolecular complexes https://t.co/cYkWSlzblK #biorxiv_bioinfo

[Robbie Joosten @Robbie_Joosten]

Our latest resource that we (@AlvarezSalmoral, @rborza_, @TassosPerrakis) created at @NKI_nl is named #AlphaBridge and can help you interpret and validate the interfaces in AlphaFold3 complexes. @instructhub @ccp4_mx

[Daniel Alvarez Salmoral @AlvarezSalmoral]

Explore it on our web server: https://t.co/Mi09bDcDrX!

[Daniel Alvarez Salmoral @AlvarezSalmoral]

🚀 Excited to share my first preprint on bioRxiv:[https://t.co/6lhGuCFT30…]

AlphaBridge simplifies protein structure predictions for experts and novices by combining complex analysis with visual representations and interactive tools in one user-friendly platform.