Integrative single cell analysis of haematopoietic stem cell kinetics
EBPOD 2017: Project 3
This is one of 11 joint postdoctoral fellowships offered by EMBL-EBI, the NIHR Cambridge Biomedical Research Centre and the University of Cambridge’s School of the Biological Sciences in 2017.
- David Kent, Cambridge Stem Cell Institute
- Moritz Gerstung, European Bioinformatics Institute (EMBL-EBI)
Background and Summary
A fine-tuned equilibrium between haematopoietic stem cell (HSC) self-renewal and differentiation needs to exist in order to provide the trillions of mature blood cells required daily while not exhausting the stem cell population. The discovery of significant heterogeneity in HSCs has shifted our understanding of how single cells make fate choices and at least four different HSC types have been identified which differ in their self-renewal durability and the numbers and types of cells they can produce. Since HSCs can be isolated at functional purities >50%, single cell RNA sequencing (scRNA-seq) and single cell transplantation outcomes have been robustly combined to allow profiling of individual HSCs and their differentiated progeny at unprecedented detail. However, the volume and detail of these datasets pose substantial challenges for their analysis and a gap currently exists between mechanistic models of cell fate and single cell functional outcomes.
The homeostatic equilibrium of HSC self-renewal and differentiation is altered in leukaemia, often as a result of somatically acquired mutations in a common set of driver genes. The Gerstung group has pioneered molecular classification of AML subtypes and the Kent group recently showed that the order of acquisition of genetic mutations can drive distinct disease evolution. While the molecular characterisation of mutations in haematological malignancies has progressed rapidly, relatively little work has been undertaken to integrate the cell biological properties of HSCs with distinct mutations.
This EBPOD proposal therefore focuses on integrating large single cell datasets including RNA-sequencing, cell culture and transplantation assays of thousands of normal and genetically modified cells to train mechanistic mathematical models of haematopoietic growth and differentiation dynamics in vitro and to predict these kinetics in vivo. These models will uniquely combine genetics, transcriptomics and in vitro and in vivo proliferation and differentiation data from multiple mouse models of hematological malignancies.
1. Train a classifier for single HSC transplantation outcomes from 100s of individual outcomes based on in vitro proliferation and differentiation kinetics and multi-parameter flow cytometric data.
2. Analyse and model the kinetics of single HSC regeneration dynamics based on in vivo flow cytometry data collected at multiple time-points post transplantation.
3. Integrate datasets from pre-leukaemic and leukaemic mutant mouse models to gain insight into the cellular dynamics of leukaemia evolution from a single cell. The data underpinning objectives 1 and 2 will be generated for single wildtype HSCs and objective three will utilise genetically modified mice carrying single and double mutants of TET2 and J AK2 V617F , with the prospect of including data for other known driver mutations in pre-leukaemic and leukaemic mutant strains (e.g., CALR , NPM1 and DNMT3A ). Together this project aims to provide the first comprehensive integration of single cell cellular dynamics and functional outcomes with the well-developed molecular framework of the mutations driving haematological malignancies.