Skip to main content
Open Access Publications from the University of California

UC Davis

UC Davis Electronic Theses and Dissertations bannerUC Davis

Assessing the Current, Future Landscape of scRNA-seq Data Exploration and Cell Type Modeling


Droplet-based microfluidic techniques in combination with continued sequencing advancement has revolutionized many areas of cell based research. In particular, microfluidics have allowed research in genomics to examine cellular structures such as developing embryos at a resolution previously unobtainable. The introduction of single-cell RNA sequencing (scRNA-seq) technologies and bioinformatics pipelines has enabled many breakthrough discoveries such as targets for immune system regulation, the Human Cell Atlas, and cell development trajectories and their associated gene expressions. scRNA-seq is a rapidly growing approach for answering questions related to cell populations and their behaviors, responses, and peculiarities in important areas such as cancer, autoimmune deficiencies, and many other diseases. The answers that scRNA-seq can provide guide new therapeutics and the many new emerging technologies in single-cell genomics such as VDJ analysis and CITE-Seq have much more to offer. VDJ single-cell analysis targets the transcriptomic and immune receptor qualities and clonotypes of T and B cells. Cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) allows for the quantification of proteins with the usage of DNA-barcoded antibodies. These techniques along with many others are applied to bioinformatics processes and create additional data and necessary techniques for downstream analysis. These many emerging technologies come with a variety of challenges in computational biology and bioinformatics. Data from single-cell analysis are often complex, containing many forms of metadata such as clonotype, cell types or subtypes, and one or more treatments. Translational bioinformatics is a crucial step for allowing specialized researchers to easily interact with the data and ask directed questions. In addition, tissue and cell samples can vary widely in cell type contents, expression patterns, treatments and extraction. Though many methods exist for cell identification, these often are not one size fits all. In the following chapters two particular applications and methods will be discussed. The first is a method that uses generalized linear models and custom gene signature sets to identify cell types and shows improvement on existing methods. The second is an application that uses the R programming language package called Shiny that is popular for interactive data visualization. The application presented enables more complex visualizations than other available tools based on categorical and continuous variables in the dataset.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View