OpenFold is a non-profit consortium building free, open-source AI tools for structural biology and drug discovery. It offers models and code under the Apache 2.0 license and a prediction server.
Web appFree
Step 1: Open the OpenFold site and choose Explore Models and Code.
Step 2: Download the models and code you want to run.
Step 3: Or click Launch Prediction Server to run predictions from the site.
Step 4: Join the community through the CoFold Summit or the SEPIQ challenge.
Step 5: Contribute by joining the consortium, collaborating or donating.
Yes. It is a non-profit effort and the tools are released under the Apache 2.0 license.
A consortium of more than 40 member organizations, hosted by the Open Molecular Software Foundation, a 501(c)3 non-profit.
Members contribute financially or technically and take part in roadmap decisions. Membership is mainly for organizations and uses a participation agreement.
Researchers can collaborate on benchmarks, engineering and data, and individuals can donate through the foundation with OpenFold in the memo.
M3GNetGraph neural-network materials model for energies, forces, stresses, relaxation, and molecular dynamics
BorzoiDeep-learning model for predicting RNA-seq coverage and gene-expression effects from genomic DNA sequence
ImmuneBuilderImmuneBuilder is a set of deep learning models that predict the structures of antibodies, nanobodies and T-cell receptors from sequence. It is installed with pip and used from Python or the command line.