Westlake University and Alibaba Damo Academy Launch AI Model for Cell Reprogramming

Westlake University in China, in collaboration with Alibaba Damo Academy, has announced the development of an AI model called "Guiyuan." This model is designed to predict cell fate and assist in stem cell reprogramming, addressing a long-standing challenge in regulating cellular functions which previously relied heavily on time-consuming experiments and expert judgment.

The "Guiyuan" model employs a dual modal encoding strategy to analyze a vast array of potential combinations—nearly 4 million—known to influence cell development, including small molecule drugs and protein-based growth factors. The AI aims to predict which combinations will lead to desired cellular developmental states, potentially shortening an experimental process that could otherwise take years or decades.

Traditionally, stem cell reprogramming, such as transforming skin cells into induced pluripotent stem cells (iPSCs), has been highly dependent on individual experience and iterative trial-and-error. The "Guiyuan" model offers researchers recommendations for the most effective combinations and helps explain the underlying biological mechanisms, making the process more predictable and efficient.

The research team successfully utilized the model to generate high-quality epiblast-like stem cells, mimicking early embryonic cells. These cells are crucial for understanding embryonic development and may have applications in areas such as cell therapy and tissue engineering. The model is also being applied to other tasks, including the production of precursor cells for dopaminergic neurons, which are relevant for Parkinson's disease treatment.