Technology Dissemination

The Dean lab has worked to bring non-commercially available yet potentially transformative imaging technologies to biologists throughout UT Southwestern. Depending upon their mode of operation, these microscopes routinely deliver best-in-class or near best-in-class performance. For example, in collaboration with Dr. Reto Fiolka, we developed Field Synthesis, which reproduces many of the advantageous properties of Lattice Light-Sheet Microscopy, albeit with substantially improved light-throughput, ease of alignment, and the ability to simultaneously image multiple fluorophores. Given the demand for imaging chemically cleared tissue at the highest resolution, we also developed two variants of Axially Swept Light-Sheet Microscopy that provide an unparalleled combination of field-of-view and isotropic resolution.

The first is optimized for whole-organ imaging, and provides ~600 nm isotropic resolution, a field-of-view of 870 x 870-micron, and with tiling and image fusion, is capable of imaging cm-scale objects. The second, which provides an isotropic resolution of ~300 nm and a field-of-view of 327 x 327 microns, is ideal for sub-cellular biology within intact tissues. To our knowledge, this is the highest z-resolution ever achieved in light-sheet microscopy in the absence of super-resolution techniques. We also developed an Oblique Plane Microscope with ~200 and ~450 nm lateral and axial resolution, respectively, a high-speed spectral total internal reflection microscope capable of imaging ~8 intracellular labels, and a 3-photon laser-scanning microscope with adaptive optics for intravital imaging at depths beyond 1 mm.

Computer Vision

The large increase in data flux afforded by modern imaging systems necessitates scalable and quantitative computer vision analyses. As such, we work closely with the Danuser lab in an effort to develop cutting-edge computer vision software, as well as the high-performance computing team at UT Southwestern (BioHPC) to optimize workflows and accelerate image processing tasks. Software developed includes u-Track 3D, which places globally optimal spatiotemporal tracking solutions in human interpretable dynamic regions of interest.

Software also automatically provides trackability metrics that quantitatively inform the user on the accuracy of their results. Thus, u-Track 3D solves a fundamental challenge in 3D particle tracking; the challenges in visualizing, optimizing, and measuring trajectory information in dense particle fields. Separately, we also helped develop software tools that represent the cell surface as a mesh, thereby enabling an accurate description of cell morphology, the automatic detection of morphological motifs, biosensor translocation, and the subcellular spatiotemporal dynamics of key oncogenes.