Browsing by Author "Rynes, Mathew"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Principles of Computer Numerical Controlled Machining Applied to Small Research Animal Microsurgical Procedures
    (2017-12) Rynes, Mathew
    The palette of tools available for systems neuroscientists to measure and manipulate the brain during behavioral experiments has exploded in the last decade. Implementing these tools, from electrical to optical sensors require the removal of bone tissue without damage to the underlying brain tissue. This is typically a delicate procedure as the skulls of commonly used inbred mouse strains are very thin (~80-500 μm above the mice dorsal cortex). However, with increasing complexity, these microsurgical procedures have become art forms. It takes many months to become skilled at performing these operations. Automating some of the tissue removal processes would potentially enable more precise procedures to be performed. Here, we introduce the ‘Craniobot’, a microsurgery platform, assembled with off-the- shelf components, that combines automated skull surface profiling with a CNC milling machine to perform a variety of microsurgical procedures in mice. The Craniobot utilizes a low force contact sensor that can accurately measure the surface of the skull across the whole dorsal skull with a precision of 2.4 ± 8.5 µm and this information can be used to perform milling operations with comparable precision. We have used the Craniobot to perform skull thinning, small to large craniotomies, as well as drilling pilot holes for anchoring cranial implants. The system is implemented using open source and customizable machining practices, this approach can be expanded in the future to larger animal models, or for more complex procedures and a more comprehensive part of the pipeline of in vivo neuroscience.
  • Thumbnail Image
    Item
    Sequences of cortical activation states encode search strategies during spatial navigation
    (2023-05) Rynes, Mathew
    Successful goal-directed navigation requires an integration of multiple streams of sensory environments, incorporating those representations into goal-related plans, and actuating those goal-related plans. How this multi-sensory information is processed across the cortex during navigation is not well understood. To investigate this, we recorded cortex-wide calcium dynamics in Thy1-GCamp6f mice (n = 11) using the mini-mScope (Rynes*, Surinach* et al 2021), a head-mounted miniaturized microscope capable of mesoscale calcium imaging of large swathes of the dorsal cortex, in mice solving the Barnes maze. The Barnes maze consists of a single escape hole in the presence of mildly aversive sensory stimuli. Mice learned to successfully navigate to the escape hole predominantly using a random search strategy in early trials, progressing to serial search and spatial search strategies in later tests. A holistic view of the activity across the cortex during a rapid, ethologically relevant spatial learning task revealed neural states indicative of distinct sets of brain-wide circuits being recruited during navigation behaviors. Using an unsupervised algorithm, we clustered cortical activity via k-means clustering into sets of “states” characterized by instantaneous patterns of spatially distributed activity across the FOV of the cortex. We found 5-10 states in each mouse and 7 states across mice. This low-dimensional space allowed the discovery of search strategy-specific spatiotemporal sequences of activity. Trial initiation was marked by repeated macro timescale patterns of state activation with a prolonged duration of activation of states involving high calcium activity in the frontal regions of the cortex occurring shortly after trial initiation. This frontal cortex-activated state coincided with mice approaching the edge of the maze from the center and occurred reliably in nonrandom search strategies. These events were preceded by sequences of state transitions that were distinct for serial and spatial search strategies.