Bioinformatics Seminar
Time:
Venue: Na
5 June 2018
NaImproving differential expression analysis for single-cell RNA-seq data: method and application
Chengzhong YeWEHI Bioinformatics
Single-cell RNA sequencing (scRNA-seq) technology enables high-throughput transcriptome profiling at single-cell resolution ; providing researchers with an unprecedented opportunity for dissecting heterogeneous biological systems. Yet distinct features in the scRNA-seq data present a variety of analytical challenges. For example ; extensive loss of molecules occurs at all stages of the experimental process ; which gives rise to the characteristic dropout events in data. This talk consists of two parts ; both focusing on bioinformatics approaches for scRNA-seq experiments. The first part describes the design of a novel statistical method for differential gene expression (DE) analysis. Through explicit modelling of the molecule capturing process ; we are able infer and perform DE analysis on the unobserved pre-dropout molecule counts. Benchmarking using simulated and real data showed improved performance compared with existing methods. The second part discusses a data analysis workflow of a breast cancer tumour infiltrating T cell dataset. Leveraging several new dedicated scRNA-seq methods ; including our DE model ; we have identified and characterized a novel cell population ; tissue-resident memory T cells ; in the tumour infiltrating T cells.
;