Project Summary/Abstract Decades of basic research have afforded a general picture for how a viral particle may approach a cell, be internalized by the cell, and hijack the cell to produce more progenitors. The accumulated knowledge, in turn, has allowed one to formulate specific mechanistic questions. For instance, why would a particular mutation in a virus make it more effective in infecting a live cell? Is it because the mutation makes the virus stay on the cell surface longer (on time) but exhibiting the same cell-internalization propensity, or is it because the mutation makes it easier for the virus to invade the cell while the off time remains the same? Where inside a cell and when does a viral particle escape an endosomal enclosure and/or shed its capsid to release its genetic material? By analogy, similar questions can be formulated in designing nanoparticle-based drug delivery vessels. Many ingenious experiments using a diverse array of biochemical and biological tools have been devised to address them, and the insights have led to translational research that has direct therapeutic impacts. A direct observation and recording of these dynamical events that a viral particle may exhibit in its cell-invasion and multiplication cycle could potentially offer much more. Recently, our laboratory put forward a proof-of-principle imaging platform that allows one to do just that: While an integrated two-photon laser-scanning microscope continuously provides 3D sections of the environmental context of a moving virus-like nanoparticle, the nanoparticle is tracked in 3D. This is performed by moving the sample in order to keep the particle at the center of a microscope objective focus with a super-resolution localization precision (~10 nm) in all three dimensions and at a 10-microsecond time resolution. Even for events as simple as a virus-like nanoparticle approaching and landing on a cell, this prototype multiresolution microscope has allowed us to uncover that, unexpectedly, a virus-like particle tends to slow down significantly before its landing on a cell surface. While the prototype instrumentation demonstrates that direct 3D high-resolution visualization could indeed provide uniquely new information that has been inaccessible using conventional methods, its trajectory throughput is too low to be of widespread practical use. This developmental project is intended to test whether a new microscopy design would make this approach higher throughput. This is to be achieved by a new instrumentation design and a machine learning computational backend for dynamic content filtering. Quantitative measures for benchmarking and feasibility testing are also described. If feasible, the community would have a completely new and practical way of looking at viral particle actions and nano-carrier delivery performances.