Template-based mapping of dynamic motifs in tissue morphogenesis

摘要

Much like many ballet dances that are choreographed by combining the same movements in different ways, tissue morphogenesis relies on combinations of highly conserved sub- to multi-cellular behaviors, or motifs, such as cell division, death and intercalation. Algorithms for systematic mapping of morphogenetic motifs are essential for making the most of rapidly accumulating live imaging datasets. Here, we present a template-based strategy for motif mapping using an integrated time series data mining and graph theoretic approach. Rather than relying on a set of manually-crafted criteria, our algorithm identifies instances of the motif based on similarity to a user-provided example or template. This strategy allows our algorithm to accurately identify the complete duration of each instance and automatically label different stages throughout its progress, such as cell cycle phases during cell division. As an illustration, we mapped intercalary cell behaviors during convergence-extension movements in wild-type and mutant Drosophila embryos. Our results revealed statistical properties in the dynamics of T1-cell rearrangements and multicellular rosettes and identified a new mode of iterative cell intercalation. Our time series approach opens new avenues for systematic decomposition and analysis of tissue morphogenesis.