Navigate: an open-source platform for smart light-sheet microscopy

Abstract

Navigate enables biologists and technology developers alike to establish and reuse smart microscopy pipelines on diverse sets of hardware from within a single framework. While generalizable Python-based frameworks for smart microscopy have been built, they were designed for stimulated emission depletion⁵ or single-molecule localization microscopy⁶ and do not yet address LSFM’s specific acquisition challenges, including decoupled illumination and detection optomechanics and a lack of an optical substrate for focus maintenance. While GUI-based frameworks for image postprocessing exist⁶, to the authors’ knowledge, navigate is the only software that enables decision-based acquisition routines to be generated in a code-free format (Supplementary Table 1).

A schematic of navigate’s software architecture is presented in Supplementary Fig. 1. The plug-in architecture of navigate facilitates the addition of new hardware, enabling users to integrate otherwise unsupported devices. For image-based feedback, custom analysis routines can also be loaded within navigate’s environment to evaluate images stored as NumPy arrays in memory. Navigate supports the addition of REST-API interfaces for two-way communication with image analysis programs running outside of Python or in different Python environments, such as Ilastik⁷, enabling developers to make calls to state-of-the-art software while avoiding dependency conflicts. Image-based feedback can be leveraged to perform diverse tasks, such as sensorless adaptive optics in optically complex specimens (Fig. 1c). We believe this flexibility is necessary for the software to accommodate the diverse modalities of LSFM and to integrate feedback mechanisms.