Reconstructing Single-Cell Trajectories via Stochastic Tree Search

• Activity Number: 55 - Statistical methods for data from single cell technologies
• Type: Contributed
• Date/Time: Sunday, August 8, 2021 : 3:30 PM to 5:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Jingyi Zhai* and Hui Jiang
• Companies: University of Michigan and University of Michigan
• Keywords: Single-cell; Trajectory estimation; Stochastic tree searching

Abstract

The single-cell RNA-sequencing (scRNA-seq) technology is a recent advancement that enables the measurement of gene expression at single cell level, so it prompts the understanding in the dynamic cellular process. Reconstructing a cell trajectory from the gene expression for a sample of cells is introduced as a new research area by this technology. The high-dimensional gene expression data space and the associated high-level noise pose difficulties in modeling the trajectory from the original expression data. We develop a new TR method to estimate a tree-structured cell trajectory from scRNA-seq data. We derive a penalized likelihood framework and a stochastic optimization algorithm to search through the non-convex tree space to obtain the global solution. We compare our proposed approach with other existing methods using simulated and real scRNA-seq data sets. As the simulation study and the real data example show, our algorithm is more accurate and less sensitive to outliers than other compared methods in terms of cell ordering estimation.