Automated identification of stratifying signatures in cellular subpopulations

Significance Single-cell measurement technologies such as flow cytometry permit the investigation of specific cellular subpopulations. Mass cytometry currently measures >40 parameters per cell and produces phenotypically rich datasets that may be retrospectively interrogated for relevant biological signal. There are few methods that identify experimentally relevant subpopulations within these datasets, and most do not scale well to higher-dimensional measurements. To address this bottleneck, we present a data-driven method termed Citrus that identifies cell subsets associated with an experimental endpoint of interest. Citrus can test diverse experimental hypotheses and is demonstrated through the systematic identification of (i) blood cells that signal in response to experimental stimuli and (ii) T-cell subsets whose abundance is predictive of AIDS-free survival risk in patients with HIV. Elucidation and examination of cellular subpopulations that display condition-specific behavior can play a critical contributory role in understanding disease mechanism, as well as provide a focal point for development of diagnostic criteria linking such a mechanism to clinical prognosis. Despite recent advancements in single-cell measurement technologies, the identification of relevant cell subsets through manual efforts remains standard practice. As new technologies such as mass cytometry increase the parameterization of single-cell measurements, the scalability and subjectivity inherent in manual analyses slows both analysis and progress. We therefore developed Citrus (cluster identification, characterization, and regression), a data-driven approach for the identification of stratifying subpopulations in multidimensional cytometry datasets. The methodology of Citrus is demonstrated through the identification of known and unexpected pathway responses in a dataset of stimulated peripheral blood mononuclear cells measured by mass cytometry. Additionally, the performance of Citrus is compared with that of existing methods through the analysis of several publicly available datasets. As the complexity of flow cytometry datasets continues to increase, methods such as Citrus will be needed to aid investigators in the performance of unbiased—and potentially more thorough—correlation-based mining and inspection of cell subsets nested within high-dimensional datasets.