ABSTRACT
Quantification of microtubule (MT) dynamic instability (DI) is essential to mechanistic dissection of MT assembly and the activities of MT binding proteins. Typical methods for quantifying MT dynamics assume that MT behavior consists of growth and shortening phases, with instantaneous transitions (rescues and catastrophes) in between. However, examination of DI data at high temporal and spatial resolution reveals the presence of ambiguous behaviors that cannot easily fit into these categories. Failure to objectively recognize and quantify these behaviors could reduce the reproducibility of DI data and impact attempts to dissect mechanisms. To address these problems, we recently developed STADIA (Statistical Tool for Automated Dynamic Instability Analysis), a MT analysis software package that uses length-history data as input and is (presently) implemented in MATLAB. STADIA uses machine learning methods to objectively analyze and quantify macro-level DI behaviors exhibited by MTs, including variable rates of growth and shortening and a newly quantified DI phase: stutter. Here we overview the process of using STADIA to quantify MT dynamics and provide a set of concrete protocols for using STADIA to process and analyze MT length history data.
