Stanford Medicine Researchers Develop Advanced AI Models for Cell Biology, Enabling Cross-Species Comparison

Researchers at Stanford Medicine have developed two new artificial intelligence models, universal cell embedding and TranscriptFormer, which significantly advance cell biology research. TranscriptFormer, trained on data from 112 million cells across 12 species, facilitates dramatically easier comparisons of cells among diverse species, offering new avenues for insights into diseases and potential cell treatments. The work was published in Nature and Science.