Building a Systematic, Comprehensive Mammalian Cell Fate Map

项目总结

PROJECT SUMMARY Cell fate maps are immensely powerful. They illuminate the pathways of differentiation and showcase the dynamics of and coordination by cells to achieve complex biological structures. Accordingly, a sufficiently high- resolution fate map of mammalian development would function as a guide to investigate factors that direct these processes, and serve as an invaluable tool to generate and evaluate in vitro models and design cellular therapies. The need for cellular therapies, including agents that might replace or repair damaged tissue or organs, are becoming more urgent as the population ages. Until recently, lineage tracing in mammals relied on techniques with limited precision, scope, and ability for new discovery. It is not surprisingly then that fundamental questions, such as how cells move from pluripotency to more restricted cells types, remain opaque. This project aims to address those questions by building a comprehensive catalogue of mammalian differentiation. Leveraging recent innovations in single-cell RNA-sequencing and mammalian genome editing using CRISPR-Cas9, this project seeks to establish a systematic, evolving lineage tracing platform capable of recording mammalian processes. Simultaneous capture of single cell transcriptomes along with lineage information facilitates the ability to link a cell’s current state with a piece of its history. An essential component to this technology is a complementary computational infrastructure for processing and analyzing data. To find differentiation pathways stemming from pluripotency, the lineage tracing platform will be applied to an in vitro model of mouse development, gastruloids. Gastruloids form from the aggregation of mouse embryonic stem cells and faithfully mimic many features of mammalian embryogenesis. Moreover, gastruloids are a well- defined, tractable model enabling higher throughput studies in a cost-effective manner. To interpret data produced from the lineage tracing experiments, a hidden Markov model is proposed to integrate information across data sets and identify differentiation trajectories. Importantly, multiple trajectories may be discovered for a given cell type showcasing the unbiased, data driven property of this technology. Finally, the modes by which the lineage tracing platform may be used to identify genetic regulators and to inform cell type manipulation in vitro are discussed.