Nonparametric Bayesian Two-Level Clustering for Subject-Level Single-Cell Expression Data

The advent of single-cell sequencing opens new avenues for personalized treatment. In this paper, we address a two-level clustering problem simultaneous subject subgroup discovery (subject level) and cell type detection (cell expression data from multiple subjects. However, current statistical approaches either cluster cells without considering the heterogeneity or group subjects using information. To bridge gap between grouping, develop nonparametric Bayesian model, Subject Cell Single-Cell (SCSC) to achieve grouping simultaneously. SCSC does not need prespecify number number. It automatically induces structures matches types across Moreover, it directly models raw count by deliberately data's dropouts, library sizes, over-dispersion. A blocked Gibbs sampler is proposed posterior inference. Simulation studies application multi-subject iPSC scRNA-seq dataset validate ability simultaneously cells.