Abstract
Currently, identifying novel biomarkers remains a crucial need for cancer immunotherapy. By leveraging single-cell cytometry data, Greene et al. developed an interpretable machine learning method, FAUST, to discover cell populations associated with clinical outcomes.