Boltz-1: Democratizing Biomolecular Interaction Modeling
2024

Boltz-1: A New Tool for Biomolecular Interaction Modeling

publication Evidence: high

Author Information

Author(s): Wohlwend Jeremy, Corso Gabriele, Passaro Saro, Reveiz Mateo, Leidal Ken, Swiderski Wojtek, Portnoi Tally, Chinn Itamar, Silterra Jacob, Jaakkola Tommi, Barzilay Regina

Primary Institution: Cold Spring Harbor Laboratory

Hypothesis

Can Boltz-1 achieve AlphaFold 3-level accuracy in predicting biomolecular complex structures?

Conclusion

Boltz-1 matches the performance of top commercial models and sets a new standard for accessible structural biology tools.

Supporting Evidence

  • Boltz-1 achieves AlphaFold 3-level accuracy in predicting 3D structures.
  • The model demonstrates performance on par with state-of-the-art commercial models.
  • Training and inference code, model weights, datasets, and benchmarks are released under the MIT open license.

Takeaway

Boltz-1 is a new computer program that helps scientists understand how molecules interact, making it easier for them to discover new drugs.

Methodology

The study introduces an open-source deep learning model with innovations in architecture, speed optimization, and data processing.

Digital Object Identifier (DOI)

10.1101/2024.11.19.624167

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication