US EditionOpen-Source Protein Structure AI Aims to Match AlphaFold
OpenFold3 offers commercial users an open-source AI for protein-drug interaction predictions, with five companies collaborating to enhance the model using proprietary data.
- On October 28, the OpenFold Consortium launched OpenFold3 as an open-source reconstruction of Google DeepMind’s AlphaFold3, enabling commercial use by researchers and pharmaceutical companies.
- After DeepMind kept AlphaFold3 closed-source, a petition by hundreds of scientists criticized restricted access, prompting Mohammed AlQuraishi and the open-source research community to reverse-engineer its capabilities.
- The OpenFold3 preview release shares its code for testing after US$ 17 million in development and training on more than 300,000 molecular structures plus over 40 million synthetic structures.
- Five companies formed the Federated OpenFold3 Initiative, announced October 1 by Apheris, Berlin-based company, to train models on about 4,000 to 8,000 protein-drug pairs each and aggregate knowledge on about 20,000 proteins and drugs.
- The consortium says it is still working on technical improvements and plans a full release in the coming months while soliciting researcher feedback from the preview release.
23 Articles
23 Articles
PR Newswire
SandboxAQ Contributes to OpenFold Consortium's Newly Released OpenFold3 Model
/PRNewswire/ -- SandboxAQ today announced key contributions to OpenFold3, the open source biomolecular co-folding model released today by the OpenFold...
Open-source protein structure AI aims to match AlphaFold
The developers of OpenFold3 have released an early version of the tool, which they hope will one day perform on par with DeepMind’s protein-structure model. The developers of OpenFold3 have released an early version of the tool, which they hope will one day perform on par with DeepMind’s protein-structure model.
OpenFold3 preview: Apache-licensed co-folding model
The OpenFold Consortium released a preview of OpenFold3, an Apache 2.0-licensed foundation model that predicts 3D protein structures and co-folded complexes with small-molecule ligands, nucleic acids and other proteins, positioning it for industry use at scale. Simultaneously, Apheris, a Berlin-based provider of enterprise-grade AI applications for drug discovery, launched ApherisFold to run, benchmark, and fine-tune… The post OpenFold3 debuts a…
The OpenFold3 model, launched on 28 October, is a reconstruction of Google DeepMind's AlphaFold3. A large consortium of researchers led by Mohammed Al-Quraishi of the University of California, the OpenFold3 model is a new model of the main artificial intelligence tool for predicting how proteins will interact with small molecules, such as drugs.
Coverage Details
Bias Distribution
- 89% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
