Abstract

Single-cell computational/system immunology

Jinmiao Chen, Project Leader Research Scientist, Singapore Immunology Network (SIgN), A*STAR

Human immune system comprises highly heterogeneous immune cell subsets. Integrative high-dimensional single cell analysis using CyTOF and single-cell RNA-sequencing offers unprecedented opportunities in studying immune cell diversity, cell lineage, cell movement, and cell-cell interaction. However, analysis and interpretation of high-dimensional single-cell data remains challenging with existing methods. My lab (single-cell computational/system immunology lab) has developed a series of novel methods including DensVM, clusterX and ModelGate to automatically identify immune cell subsets. These methods provide unbiased data-driven delineation of cellular diversity and composition. In addition to cellular diversity and subset identification, we have invented cell lineage algorithms including SIDAP, NBOR, uSPIN and Mpath algorithms to map linear or non-linear multi-branching cellular trajectories in which individual cells or cell subsets are assigned pseudo-temporal order of cellular development and differentiation. By integrating unsupervised learning strategies, these algorithms are able to infer the relationship between different cell populations during development, differentiation, activation and migration. This presentation will discuss how single-cell analysis can benefit immunology research and what the challenges in data analysis are and how my lab has addressed these challenges by developing novel computational algorithms.