Boltz-1
a year ago
- #open-source
- #bioinformatics
- #biomolecular-structure
- Boltz-1 is an open-source model for predicting biomolecular structures, including proteins, RNA, DNA, and other molecules, with support for modified residues, covalent ligands, and glycans.
- The model is available under the MIT license, allowing free use for both academic and commercial purposes.
- Installation can be done via PyPI (`pip install boltz -U`) or GitHub for daily updates.
- Boltz-1 accepts input in Fasta files, YAML schema, or directories for batched processing.
- Evaluation scripts and predictions for Boltz-1, Chai-1, and AlphaFold3 are provided for benchmarking.
- The model can be retrained using provided training instructions.
- Community engagement is encouraged via Slack for discussions and collaborations.
- Boltz-1 runs on Tenstorrent hardware through a fork by Moritz Thüning.
- Citations are required for research use, including the Boltz-1 paper and ColabFold for MSA generation.