US EditionAlphaFold is running out of data — so drug firms are building their own version
- The AI Structural Biology Consortium, powered by Apheris, is initiating a collaboration to evaluate and refine OpenFold3, a protein structure prediction algorithm developed by Mohammed AlQuraishi at Columbia University, for predicting 3D structures of molecule complexes relevant to drug discovery, particularly small molecule-protein and antibody-antigen interactions.
- AlphaFold, which revolutionized biological research by providing access to virtually all of the 200 million protein structures identified by researchers and relies on the PDB, a public repository housing over 200,000 entries of experimentally-determined protein and nucleic acid structure data collected over 50 years, is facing a data shortage, especially regarding data relevant to drug discovery.
- OpenFold3 will be fine-tuned using proprietary data from AbbVie and Johnson & Johnson within a confidentiality-preserving environment, and access to the fine-tuned model will be limited to participants contributing their own proprietary data.
- Robin Roehm, CEO and co-founder of Apheris, stated that leveraging industrial data is a 'huge value add' and that 'the real breakthroughs can only happen through increased amounts of data and tapping into industrial data,' while John Karanicolas, head of computational drug discovery at AbbVie, indicated the consortium allows them to share data and explore the hypothesis that their internal datasets will be highly complementary when training AI models.
Insights by Ground AI
Does this summary seem wrong?
4 Articles
4 Articles
All
Left
Center
2
Right
AlphaFold is running out of data — so drug firms are building their own version
Thousands of 3D protein structures locked up in big-pharma vaults will be used to create a new AI tool that won’t be open to academics. Thousands of 3D protein structures locked up in big-pharma vaults will be used to create a new AI tool that won’t be open to academics.
·United Kingdom
Read Full Article
Coverage Details
Total News Sources4
Leaning Left0Leaning Right0Center2Last UpdatedBias Distribution100% Center
Bias Distribution
- 100% of the sources are Center
100% Center
C 100%
Factuality
To view factuality data please Upgrade to Premium
Ownership
To view ownership data please Upgrade to Vantage
