Towards using organoids in functional genomics – A single cell characterisation of mouse gastruloids

Abstract

Gastrulation is the stage of development during which pluripotent cells differentiate and commit to one of the three main germ layers. While many of the key signalling pathways are well-known, the epigenetic processes underlying this lineage commitment remain poorly understood in mammals. Large-scale perturbation studies are necessary to functionally study the epigenetics of mouse gastrulation. While it has thus far not been feasible to perform such studies at scale in vivo, organoid models of mouse gastrulation are now making them possible. This thesis concerns the characterisation of an organoid model of mouse gastrulation called gastruloids, as a model for such perturbation studies. Gastruloids are grown by aggregating around 300 mESCs for 48hrs, followed by a 24-hr treatment with a WNT agonist which stimulates the gastruloids to produce all three main germ lineages. However, to use gastruloids in perturbation screens, in particular those with single cell readout, it is important to understand the total gene expression in each of a gastruloid’s cells, as well as how much heterogeneity there is between gastruloids. Using single cell RNA-sequencing of individual mouse gastruloids grown in different conditions and by different researchers between day 3 and day 5 of differentiation (roughly corresponding to E6.5-E8.75), this thesis shows substantial inter-organoid and inter-experiment heterogeneity. This heterogeneity in differentiation behaviour is important when using organoids as a model to study the mechanisms regulating gene expression changes in early development, and to that end this thesis introduces a computational method to explicitly model the inter-individual null distribution.