Key points from article :
A new player in AI-driven drug discovery, Chai Discovery, has emerged with a powerful new tool for molecular structure prediction, known as Chai-1. This young startup, backed by prominent names such as OpenAI and Thrive Global, has raised $30 million in early-stage funding. The team behind it includes experts from top tech companies like OpenAI, Google AI, and Meta.
Chai-1 is a multi-modal AI model designed to predict the structure of key biological molecules, including proteins, DNA, and RNA. The startup is offering free access to Chai-1 through a web interface, encouraging both commercial and research collaboration. Unlike other models like AlphaFold, Chai-1 achieves accurate molecular predictions without needing multiple sequence alignments (MSAs), and it excels at predicting multimer structures.
One standout feature is Chai-1's adaptability. By integrating lab-derived data in real time, it can significantly improve the accuracy of predictions, such as doubling the precision of antibody-antigen structure predictions. Chai Discovery envisions a future where biology is not just a science but an engineering discipline, and Chai-1 is just the beginning of this transformation.
