Abstract
Single cell RNA-sequencing is an increasingly popular tool for investigating the variability in gene expression between individual cells. Compared to the previously wide spread methods such as bulk RNA-sequencing, the single cell approach gives the advantage of a much higher cellular resolution, but it also provides us with much noisier data. In the recent years a large number of bioinformatics tools have been developed to analyze scRNA-seq data. There is an abundance of methods, for example more than 50 methods for trajectory inference have been developed since 2014. Many of the tools previously developed for bulk RNA-seq can also be applied to single cell data, but there are some crucial differences in the inherent characteristics of the data that differentiates scRNA-seq data from its bulk counterpart, among others in the statistical characteristics of the data. In order to use the large amounts of data generated by scRNA-seq to produce new biological insights, we need to integrate the relevant tools into an integrated coherent framework. This thesis presents a pipeline that I developed, called SingleFlow, to perform large scale analysis in such an integrated framework. The pipeline’s usefulness was validated by applying it in the context of natural killer (NK) cell biology. There are a number of questions unanswered in the field of NK cell biology. By applying the pipeline to a unique scRNA-seq data set of NK cells from two different donors, we identified a temporal transcriptional map of human NK cell differentiation. By mapping gene expression trends to pseudotime, we identified distinct transcriptional checkpoints that represent changes during NK cell differentiation. We also identified previously undescribed subsets within the CD56bright subset of NK cells. The combination of the pipeline’s analysis and the potential of the novel data set proved useful in identifying important gene programs that are associated with NK cell differentiation. This knowledge holds potential to guide the development of new strategies for NK cell-based cancer immunotherapy.