Rosetta-DL

A bundle of deep-learning packages for biomolecular structure prediction and design contributed to the Rosetta Commons.

The packages listed below are provided by contributors to the Rosetta Commons for free for non-commercial use under the Rosetta-DL Non-Commercial Licensing Agreement or a more permissive license.

For commercial licensing of this bundle, please contact the University of Washington CoMotion ([email protected]), which manages these packages on behalf of the Rosetta Commons member institutions. Licensing revenue supports the Rosetta Commons research community by funding conferences, workshops, summer interns and post-baccalaureate scholars, mini-grants, user support, documentation and code infrastructure and testing.

Rosetta-DL packages

1. RoseTTAFold - Protein folding - Code uses MIT license, data and weights use Rosetta-DL license
2. DeepAb - Antibody structure prediction - Code, data, and weights use Rosetta-DL license
3. MaSIF - Molecular surface interaction fingerprints - MIT license
4. trRosetta2 - Protein folding by Baker lab in CASP14 - MIT license
5. Protein-Seq-Des - Protein sequence design from a learned potential - BSD3 license
6. FvHallucinator - Hallucination of antibody structures - Rosetta-DL license