Our lab has recently implemented a virtual reality (VR) experimental platform for research with human patients undergoing invasive, intracranial neural monitoring for seizure localization. The goal of our project is to design a software interface enabling high-dimensional kinematic tracking during commercial VR gameplay. This will allow us to collect a large dataset of synchronized neural signals and continuous kinematic variables such as hand location and angular velocity during the dynamic/naturalistic human movements evoked by VR gameplay, that can be used to support neural decoding of 3D human movement for brain computer interface (BCI) development in labs at UW. It can also be published as a benchmark dataset for human movement decoding to the larger community of computational neuroscientists. Currently, only 2D datasets of human movement tracking during concurrent intracranial neural recording have been provided to the larger computational community. The distribution of a 3D movement dataset with intracranial neural recording is thus vital and would facilitate many avenues of clinical, computational, and neural network (ANN) modeling research. Specifically, our project is to create a software interface, called a “layer”, that can record VR controller/headset positions and other tracked variables while a subject plays a commercial VR game that utilizes the industry-standard OpenXR framework. Our layer works by intercepting different functions in the OpenXR framework, and uses this to track movement data. Specifically, our layer will include a manifest file which holds our layer’s identification, and intercepts commands such as xrNegotiateLoaderApiLayerInterface, xrNegotiateLoaderInfo, xrCreateActionSpace, and more to specifically track the movement. The positions and data collected will be outputted to a .json file which will then be used in the clinical research detailed above.

Skin is a very important organ that facilitates our sense of touch. The touch system is complex and versatile. For example, specialized cells detect a variety of tactile stimuli, including temperature, pain, and textures. Merkel cells are specialized skin cells responsible for detecting light touch and textures in most vertebrates. Despite being discovered over a hundred years ago, their development remains poorly understood. Zebrafish are a good model organism for studying Merkel cells because the fish skin is transparent and easy to image. In wildtype zebrafish, Merkel cells are distributed in clusters, corresponding to the location of scales. The ectodysplasin (Eda) signaling pathway regulates the formation of many types of skin appendages, including mammalian hair follicles and zebrafish scales. In eda mutants lacking scales, Merkel cells appear at a lower density and are uniformly distributed across the trunk. This suggests that blocking Eda-dependent scale formation inhibited Merkel cell development, but what would be the effect of altering scale shape and size? To address this question, we examined the skin of hagoromo mutant fish. The hagoromo mutation is a viral insertion that causes an overexpression of fgf8a, another important signal pathway for scale formation. By imaging transgenic zebrafish with marked Merkel cells and osteoblasts, we used ImageJ to calculate Merkel cell density, scale area, scale Feret’s diameter, and scale aspect ratio. We found that juvenile zebrafish with the hagoromo mutation had highly variable scale shapes and sizes. Interestingly, the clusters of Merkel cells expanded or shrank to match the new scale shapes in the mutants. However, there seemed to be no significant difference in Merkel cell density. We expect to see similar results in adult zebrafish. Our research will help us understand the development of Merkel cells, therefore helping us understand skin development in vertebrates better. 

Touch is an extremely important sense to any organism’s understanding of their environment. The anatomy of the touch system is well-characterized: sensory neurons project axons to the skin where they form complexes with many specialized cells. One such specialized skin cell is the Merkel cell, known for its ability to sense gentle touch and texture. Although Merkel cells have been well studied in rodents during adult stages, little is known about Merkel cells during development. The Rasmussen lab recently identified a novel population of zebrafish skin cells sharing many characteristics of mammalian Merkel cells. In both zebrafish and mammals, Merkel cells arise from basal keratinocytes, a stem cell population in the epidermis, during normal skin development. However, the precise mechanisms involved in this process remain unclear. Stem cells can undergo two types of cell division: asymmetric or symmetric division. I hypothesize that Merkel cells develop from asymmetric division of basal keratinocytes, which would allow the production of new Merkel cells while also maintaining a certain level of stem cells. To test my hypothesis, I examined the Merkel cell lineage during zebrafish scale regeneration (to simulate recovery after injury) which we have established as an experimentally tractable system to study Merkel cell differentiation. Testing of my hypothesis has been conducted through two methods: EdU labeling/antibody staining of Merkel cells and photoconversion of Merkel cells. Preliminary results suggest asymmetric division to be occurring, with no “doublets” of recently divided Merkel cells shown in either method to support the occurrence of symmetric division. My results provide novel insights into the lineage connecting basal keratinocytes to differentiation of specialized sensory cells in the skin. I hope these findings can be used to better our understanding of the restoration of touch systems after injury.

This work proposes a novel generative design tool for passive grippers—robot end effectors that have no additional actuation and instead leverage the existing degrees of freedom in a robotic arm to perform grasping tasks. Passive grippers offer interesting trade-offs between cost and capabilities. However, existing designs are limited in the types of shapes that can be grasped. This work proposes to use rapid-manufacturing and design optimization to expand the space of shapes that can be passively grasped. Our novel generative design algorithm takes in an object and its orientation with respect to a robotic arm and generates a 3D printable passive gripper that can stably pick the object up. To achieve this, we address the key challenge by jointly optimizing the gripper shape and the insert trajectory to enlarge the set of objects that can be pasively grasped. We evaluate our method on a testing suite of 23 objects, all of which were evaluated with physical experiments to bridge the virtual-to-real gap. Inspired by the true cost of repurposing infrastructures in assembly lines following the recent changes in demand early in the COVID-19 crisis, our work allows a cost effective solution to rapidly generate, fabricate, and deploy custom passive grippers on the existing robot arms for the new products in demand.

Traditional foams are fabricated via stochastic chemical processes that yield homogeneous material properties. Foams can exhibit a wide range of material properties by varying process controls allowing them to be used in many industrial and commercial applications. Previously, additive manufacturing could only produce foam approximations in the form of traditional lattice infill. My work employs viscous thread printing (VTP) of thermoplastic polyurethane (TPU) on a fused filament fabrication (FFF) printer, exploiting the semi-viscous nature of extruded filament to coil producing a new type of printed foam. Specimens were tested under compression to determine uniformity along principal axes and behavior under strain when compared to infill patterns, such as grid and cubic. My work establishes that VTP, using elastic materials, can be used to manufacture programmable stiffness foams as a function of density, suited to a variety of needs and should be considered as an alternative to traditional foams and other printed lattice geometries.

Alzheimer’s disease (AD) is an age-related neurodegenerative disease characterized histopathologically by amyloid plaques and neurofibrillary tangles in the brains of affected individuals. The impairment of cerebrospinal fluid (CSF)-mediated clearance of proteins including amyloid beta and tau from the brain is proposed to underlie the development of AD pathology. Sleep and circadian disruption are both linked to the development of AD. CSF clearance is regulated through both sleep and circadian processes, while CSF production by the choroid plexus (CP) is diurnally regulated. The CP acts as a blood-CSF-barrier, provides nutrient delivery, and clears toxic proteins. Studies to date document reduction in CSF production, impaired blood-CSF-barrier function, and altered protein uptake in both aged and AD conditions. It is currently unknown whether the CP is itself regulated by sleep and circadian rhythms and whether disruption of these two governing processes contributes to disease development. Our initial analysis across awake, asleep, and acutely sleep-deprived young mice indicated no gene expression differences between sleep states; however, significant circadian-dependent transcriptional changes were observed. We then examined the circadian-dependent gene expression profile of the CP in aged (12-14 months) mice and in an AD mouse model. Preliminary analysis reveals a shift in the transcriptional profile of aged mice and a near complete loss of circadian regulation in the AD model. Ongoing analysis and validation are being carried out to reveal functional pathways disrupted in the CP that could provide a further understanding of AD pathology.

Chronic sleep disruption, present in 25-60% of patients suffering from Alzheimer's Disease (AD), often precedes cardinal disease symptoms. While little is known about the mechanisms underlying chronic sleep disruption and the development of clinical pathology, both acute and chronic sleep deprivation have been found to increase biomarkers of AD including neuroinflammation and amyloid-beta accumulation. Additionally, in people without AD, sleep deprivation can result in a deterioration of working memory and attention. In this study, we examine the impact of chronic sleep disruption on cognition both at baseline and in the 5xFAD mouse model of AD. The 5xFAD mouse model is a transgenic mouse model of familial amyloidosis which expresses neurocognitive impairment as early as 2 months. To define the effect of chronic sleep disruption on cognition in the absence of AD pathology, 60 wild type mice were exposed to chronic sleep disruption or sham procedure for 8 weeks between 10 and 18 weeks of age. At 18 weeks of age, I evaluated the animals for changes in spatial memory (Barnes maze), short-term memory (Y-maze), locomotion and anxiety (open field test), and activities of daily living (burrowing trials). To test whether chronic sleep disruption specifically exacerbates AD-related neurocognitive decline, the same cognitive tests were performed on 60 5xFAD+ animals exposed to 8 weeks of sleep disruption or sham treatment. I then analyzed the collected data to isolate any trends of cognitive performance, finding that chronic sleep disruption impaired cognitive performance in 5xFAD+ and littermate controls, with a more significant impact on 5xFAD+ animals. These findings highlight the critical association between dysfunctional sleep and the development of cognitive impairment with AD disease progression which then guides us toward better preventative care and treatments.

Alzheimer’s Disease (AD) is a neurodegenerative disease that affects more than 5 million Americans. The glymphatic system (a network of perivascular spaces that facilitate fluid movement and solute clearance from the brain) and its dysfunction associated with aging has been implicated in the development of AD. The water channel aquaporin 4 (AQP4), located in astrocytic endfeet bordering the perivascular spaces, is crucial for the proper functioning of the glymphatic system. Data suggests that loss of AQP4 localization results in amyloid-ß deposition, a hallmark of AD pathology, and loss of AQP4 localization accompanies aging in rodents as well as AD in humans. In this study, we quantitatively analyze the expression of aquaporin- 4ex (AQP4ex)—a translational readthrough variant of AQP4 believed to play a role in its localization—to identify any correlation with aging and AD pathology. Selective deletion of AQP4ex results in the mislocalization of AQP4 all over the astrocytic membrane, indicating that AQP4ex is a crucial element in the localization of AQP4. We analyze young, old and AD groups in the murine (mouse) brain as well as AD versus control in a human case series. Currently, we see a trend towards decline in cortical perivascular AQP4ex in the AD group, with more analysis ongoing. This is the first characterization of AQP4ex expression in the murine brain and in a human case series, and these data will contribute to the small but growing body of research on AQP4ex and its relationship with AQP4 localization, creating opportunities to identify a new novel mechanism and novel target in AD pathology.

Mild traumatic brain injury (mTBI) is a major public health issue, frequently resulting in long-term sequelae such as sleep disruption, headaches, and cognitive impacts. In recent years, mTBI has emerged as a risk factor for the development of neurodegenerative diseases, such as Alzheimer’s disease (AD). Blast-related mTBI has been experienced by large numbers of Servicemembers during the conflicts in Afghanistan and Iraq; therefore, their potential vulnerability to downstream neurodegeneration is a major concern among Veteran populations. Recent evidence demonstrates that TBI impairs the glymphatic system, a brain-wide network of perivascular channels along which cerebrospinal fluid (CSF) and interstitial fluid (ISF) exchange facilitates the clearance of interstitial solutes such as amyloid β and tau. However, these findings were in an impact TBI model, which is a brain injury caused by a blow to the head; therefore, relatively little is known about the possible effects of blast mTBI. Here, we hypothesize that glymphatic function is impaired following repetitive blast mTBI. Using a murine blast model, we measured glymphatic function at both 7-day and 28-day timepoints following a repetitive blast induced TBI. Glymphatic function was quantified using intracisternal fluorescent tracer injection and measuring the fluorescent intensity of CSF tracer movement. We found a delayed impairment in glymphatic function at 28 days post-injury. These findings may provide further insight on the mechanisms that may render the blast-injured brain vulnerable to neurodegeneration and may give rise to improved treatments for patients exposed to blast injury.

The glymphatic system, which is primarily active during sleep, is a network of astroglial perivascular channels within the brain that allow for Cerebrospinal Fluid (CSF) influx and exchange. Glymphatics play a crucial role in the waste clearance of amyloid beta, a hallmark in the development of Alzheimer’s Disease and neurodegeneration. Recently, a bidirectional relationship between Alzheimer's Disease and sleep has also been suggested with the aggregation of amyloid beta associated with mid-life sleep disruption. However, the mechanistic link between sleep disruption, particularly over chronic time scales, and the development of Alzheimer’s pathology remains unclear. This study investigates whether chronic sleep disruption, similar to that experienced in humans, will impact downstream neuropathology. We hypothesize chronic sleep disruption will result in decreased glymphatic function and subsequently increased amyloid plaque burden. This experiment utilizes a chronic sleep fragmentation model in 120 5xFAD mice from 8 weeks to 16 weeks of age. In the Lafayette Sleep Fragmentation chambers, 60 animals are disturbed every two minutes during normal sleeping periods (daylight hours). 60 mice were placed in normal sleeping conditions. After eight weeks of sleep fragmentation or sham exposure, glymphatic function is assessed by in vivo near infrared imaging following stereotactic CSF tracer injection. Animals are perfusion fixed, cryosectioned, and glymphatic function is assessed by measurement of fluorescent cerebrospinal fluid tracers in brain tissue. Aquaporin-4 localization, amyloid plaque deposition, and markers of astroglial and microglial activation are assessed by immunofluorescence. In this project, I specifically work on cryosectioning the tissue, and understanding glymphatic function through the processes of immunofluorescence imaging and analysis. The collected data demonstrated that sleep disruption did increase neuropathological outcomes.The measured impact of glymphatic function was also correlated with these downstream pathological effects. These findings could be an indicator of interactions between neurological disease progression and an inflammatory expression after sleep disruption. They can also shed more light on the complex relationship between Alzheimer’s disease progression, the glymphatic system, and chronic sleep disruption.

The Rubber Hand Illusion (RHI) is a perceptual illusion used to examine the sense of body ownership. Traditionally, the illusion is induced by stroking a visible rubber hand synchronously to a participant’s own hand, which is placed out of view. Our research is aimed at developing a virtual reality (VR) RHI task to more closely investigate how body ownership is experienced by participants in a fully immersive setting. Following the implementation of our prototype, we will conduct a signals analysis of either an EEG or invasive electrode dataset, depending on clinical patient availability. The results of this analysis will provide greater insight into the neural correlates of body ownership, as well as provide the foundation for future studies related to perception and integration of stimulation in VR environments. The VR simulation is created using Unity 3D and the SteamVR SDK. The experiment is split into four parts: (1) initial proprioceptive drift measurement, (2) induction of the illusion, (3) final proprioceptive drift measurement, (4) movement to “break” the illusion, repeat. During the proprioceptive drift measurements, participants are asked to point and click on a virtual ruler that is placed in front of them. For the illusion phase, we render a virtual hand and colocated touch of the virtual hand in synchrony with touch of the subject’s real hand -- the induction is thus accomplished in “real-time” (no pre-recorded animation and sync up), which requires precise and accurate tracking of the hand, fingers, and touch object. To achieve the precision and accuracy needed, we will use Hi5 VR gloves alongside SteamVR’s pose system. While more difficult to design, real-time induction will allow us to more flexibly explore novel ways to induce the illusion, opening new pathways for future RHI studies.

For over 40 years, various biological activities, like gene regulation, have been tracked by protein reporter systems. However, the number of uniquely addressable protein reporters that can be used together is limited due to their overlapping readout signal. This prevents simultaneous measurement of multiple reporters (multiplexing), which not only impacts scalability and convenience, but also the potential complexity of the system of interest. To overcome this problem, we have previously designed a new class of reporter proteins, Nanopore-addressable protein Tags Engineered as Reporters. These are more scalable and multiplexable than traditional reporter strategies and are read out on a commercial nanopore array. They are detected and distinguished based on their peptide barcode regions which yield distinct ionic current blockades as they dwell within the nanopore sensitive region. Here, we optimize this system for mammalian cell systems, investigate phosphorylation motif-barcodes, and analyze new random and designed barcode sequences. We demonstrate this system in human embryonic kidney 293 cells transfected with differentially barcoded genetic circuits. We aim to utilize this novel reporter system to investigate complex mammalian processes such as chromatin regulation dynamics and the determination of cellular phenotypes using customized reporter cassettes. Ultimately, our technology will increase the scale and complexity at which systems such as these can be studied, leading to deeper understanding of biological programming and thus more robust synthetic gene circuit development.

Essential Tremor (ET) is the most common neurological movement disorder, impacting approximately 1% of the global population. Symptoms of the disorder are characterized by involuntary rhythmic motions of affected body parts and range greatly in tremor severity. As the disease progresses, pharmacological treatments often fail, requiring implantation of a deep brain stimulation (DBS) device to suppress symptoms. Effective treatment requires exhaustive physician tuning of stimulation parameters, which include numerous clinician visits for patients. One way that clinicians assess symptoms is to ask patients to perform behavioral tasks such as drawing spirals, which indicate non ideal stimulation through tremor patterns in the drawing. The goal of this study was to explore the feasibility and options for remote collection of symptom assessments, as well as to explore methods for detecting and characterizing tremor based on spiral features. Consented patients performed spiral drawing tasks multiple times a day on a personal digital assistant such as a smartphone with the collected data securely stored on the cloud. To ensure that access to a mobile device wouldn’t disqualify participants, we developed a device loaning process used in addition to an application developed by Runelabs as supporting infrastructure to collect data. The results of our research showed that we could not only collect data remotely over extended periods of time, but also replicated HOG (Histogram of Oriented Gradients) based classification algorithms on existing datasets to distinguish healthy vs tremor spirals with up to 94.5% accuracy. Previous lab-published results on the same dataset demonstrated a 98.3% accuracy using Principal Component Analysis, which illustrates the strength of our lab's prior work, though the higher efficiency of HOG classification is promising for larger datasets. This demonstrates our methods’ potential to allow for larger patient cohorts and possible integration with other inertial sensor data into a tremor classification model for future studies.