Google Deepmind released a new AI system, for designing protein structures that successfully bind to target molecules.
Dataset Used
Inputs
- Structure of Target Molecule
- A set of Preferred Binding Locations
Outputs
- Candidate protein that binds to the target at those locations
Illustration of a predicted protein binder structure interacting with a target protein. Shown in blue is a predicted protein binder structure generated by AlphaProteo, designed for binding to a target protein. Shown in yellow is the target protein, specifically the SARS-CoV-2 spike receptor-binding domain
Use Cases
- Finding Protein Structures that successfully bind to target molecules
- Disease Understanding
- Advancing Drug Design
Key Features
- It’s the first model to be able to successfully design a binding protein for VEGF-A protein.
Results
Limitations
- Unable to design successful binders against 8th target TNFa (a protein associated with autoimmune diseases)