|
Abstract:
|
Single-cell RNA (scRNA) sequencing is a high throughput analysis that enables researchers to understand the gene expression within a single cell, in what quantities they are expressed, and how they differ across thousands of cells within a heterogeneous sample (eg.early embryo development). Single-cell RNA datasets are mostly characterized by a high percentage of missing values due to technical limitations and stochastic gene expression. This could pose a major problem as the missing data can introduce bias and affect downstream analyses. Over the years, a number of methods have been developed to address the issue of missingness in single-cell RNA datasets and test the differential expression between group of cells. In this presentation, I will briefly outline the main differences between the traditional RNA sequencing and single-cell RNA, do a literature review of some scRNA sequencing imputation methods (addressing their advantages and limitations), and conduct an analysis using publicly available scRNA seq data to compare the various imputation methods. Lastly, I will conduct a differential expression test on the sample data using different methods and compare my findings.
|
