Plant cells are the fundamental building blocks of growth and development. For each cell type, the size, shape, and mechanical properties of the cell wall are customized for particular physiological functions (Szymanski and Cosgrove, 2009; Winship et al., 2011). The morphogenesis of highly polarized cell types such as trichoblasts and pollen tubes is internally programmed and occurs largely in the absence of a neighbor. Most cell types differentiate in the context of a tissue. Therefore, their growth and shape change can operate at larger spatial scales to influence tissue- and organ-level processes. Because plant cells grow symplastically and are mechanically coupled to their neighboring cells, growth properties and information flow within and between tissues can feed back on and influence cell behaviors. Plant cells are also metabolically specialized. Within a single tissue or organ, cell types can differ greatly in terms of how central metabolism is fueled, the types of metabolites that accumulate, and where in the cell they are stored.

Despite the structural and biochemical diversity of different cell types, their cell biology and development can be considered as a similar set of integrated systems-level processes. For example, the metabolic activity and energy status of a cell varies as a function of light levels or developmental stage. The biosynthesis and transport activities of the cytosol and endomembrane systems are integrated with metabolism over time. Cellular systems are also integrated across wide spatial scales. Proteins and protein complexes at the approximately 10- to 100-nm scale can use the cytoskeleton to position organelles and organize the cytoplasm at the approximately 1- to 10-μm spatial scale, to influence cell behaviors. Discovering and unraveling the complexity of these multiscale systems level interactions is a grand challenge in plant research. In recent years, progress has been rapid and is being driven in large part by the widespread use of multichannel quantitative time-lapse imaging. Using this approach, it is possible to create a spatial and temporal coordinate system in which multiple parameters can be measured and cross-correlated, and the effects of mutations or other experimental manipulations can be more deeply analyzed.