Multi-Sample Differential Expression Analysis of RNA-Seq Single-Cell Data

• Activity Number: 318 - Advances on the Analysis of Single-Cell Sequencing Data
• Type: Invited
• Date/Time: Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
• Sponsor: WNAR
• Author(s): George Tseng* and Li Zhu and An-Shun Tai and Wei Chen
• Companies: University of Pittsburgh and University of Pittsburgh and National Tsing Hua University and University of Pittsburgh

Abstract

Single-cell RNA sequencing provides transcriptomic profiles at unprecedented high-resolution but with a cost of high technical noise. With the rapid technology development, it is now economically practical to sequence single cells from multiple subjects with different biological conditions to better reveal the heterogeneity at the cell level. Existing tools aiming to identify differentially expressed genes often fail to consider technical variation and subject variation, leading to inaccurate findings. We propose a generalization of zero-inflated negative binomial model which corrects both confounding variations. Applying our model to two real data sets and a synthesized data showed improved performance than existing methods. ZINBsubj is available at gihub https://github.com/lizhu06/ZINBsubj.