PhD Defense by Alexander V. Nielsen

Computational tools for identifying and relating cell-types in development and stem-cell engineering

Abtract: The ability of stem-cells to differentiate into various cell-types is the cornerstone underpinning the marvellous diversity  of tissues that make up animals. Despite their absolutely integral role in the formation of all multi-cellular life, the mechanisms  that underlie stem-cells ability to orchestrate and coordinate the choices surrounding differentiation are not fully understood.

In  this thesis I describe the work from 3 articles: For the first article, we investigated the changes stem-cells undergo when they first start forming the mammalian gut.

We did this by examining the RNA content of thousands of individual cells obtained from mouse embryos using single-cell RNA sequencing (scRNAseq) throughout the early days of development. To aid in this  investigation,

I developed a computational algorithm to compare and quantify the relationship between noisy high-dimensional data, such as scRNAseq data. Analysing the scRNAseq data, we find and characterise a group of cells previously thought to be confined to the extra-embryonic membranes surrounding the embryo, that seemingly defy their differentiation decision, and end up becoming part of the embryo’s gut.

In the 2 last articles, we investigated the differentiation decisions cells undergo when being chemically reprogrammed into different cell-types. We find a set of genes that are critical to activate for the reprogramming of somatic fibroblast cells to XEN-like cells to be successful.

We further show that a set of chemicals can be used as a starting point for making new reprogramming protocols for multiple types of cells, showing that it works for generating both neuronal and muscle cells from fibroblast.