62588 research outputs found
Sort by
Mapping electrically coupled networks in thalamic reticular nucleus
No full textThe thalamic reticular nucleus (TRN) is a key mediator of the thalamocortical relay system that is involved in attention to the sensory surround, sleep states, and wakes states. In thalamocortical relay, thalamus relays sensory information to cortex; upon collecting input from both of those areas, TRN provides inhibition to thalamus that consequently gates or modulates thalamic sensory signals before they are sent to cortex. The neural mechanisms that dictate how TRN determines the required inhibitory signals for different wake and attentional states remains to be determined.Two contributors to TRN function that require further study include the role of electrical synapses between TRN neurons and subcortical modulatory inputs that act on TRN. Previous work has established that electrical synapses in TRN facilitate synchronous activity under simultaneous excitatory drive, but their function for processing various transient signals has not been firmly established. A required advancement in pursuit of revealing the function of electrical synapses in TRN is to establish the pattern of connectivity within TRN. As for subcortical modulatory inputs of interest, The TRN is innervated by dopaminergic inputs from VTA, SNc, and RRF, and the TRN expresses two different dopamine receptors, D1 and D4 receptors. However, how activation of postsynaptic dopamine receptors affects TRN neurons remains to be determined. Dopaminergic inputs to TRN are of particular interest because dopamine modulates electrical synapse strength in retina and striatum. Therefore, it is possible that DA affects TRN function by modulating intra-nucleus communication.
My dissertation provides insight into the neural mechanisms of TRN function through two approaches. First, I establish how activation of specific dopamine receptors affects TRN excitability and electrical synapse strength. I accomplish this by recording from electrically coupled pairs of TRN neurons in vitro and measure their excitability and electrical synapse strength. By applying a D1 or a D4 agonist to the bath and measuring how TRN excitability and electrical synapses change; I found that D1 receptors broadly increase excitability and spiking output of TRN neurons, and therefore, dopaminergic inputs that target D1 receptors are well-poised to increase inhibition delivered to thalamus. D4 receptor activation also increases the excitability of TRN neurons, but that increase in excitability is accompanied by a reduction in spiking gain and spike output. D4 receptors make TRN neurons more likely to send inhibition, but they may not send more inhibition overall due to the accompanied reduction in spiking gain. Interestingly, we found that D4 receptors depressed electrical synapses. We expect that depression of electrical synapses reduces the impact of electrical synapses on TRN neurons, but it is difficult to interpret the function of that depression more broadly, since the function of electrical synapses in TRN is not clear.
A second set of experiments were designed to reveal patterns of connectivity in TRN. Because electrical synapses respond to differences in voltages between neurons, how inputs overlap or differ between coupled neurons is a critical factor for how electrical synapses affect spiking output. For a simple example, an electrical synapse between neurons with completely matching inputs will primarily act to synchronize spike times. On the other hand, an electrical synapse between neurons with completely distinct inputs will act as a coincident detector and prevent or slow spiking when the distinct inputs do not arrive at the same time. Recent discoveries that neurons of different genetics subtypes in TRN have distinct thalamic inputs enabled me to begin addressing how electrical synapses connect TRN with similar or distinct thalamic connectivity. Somatostatin-expressing TRN neurons send inhibition to and are excited by higher-order thalamus, while calbindin-expressing TRN neurons send inhibition to and are excited by first-order thalamus. Consequently, somatostatin and calbindin neurons participate in separate thalamocortical channels. If TRN neurons of different genetic subtypes form electrical synapses within TRN, then those electrical synapses functionally bridge thalamocortical relay and feedback channels. I used two methods to test for homocellular and heterocellular electrical synapses in TRN. The first method utilized patching pairs of genetically labeled TRN neurons, either matched or mismatched in subtype, to measure individual electrical synapses. The second method utilized a novel optogenetic method that measures electrical synapses by their ability to shift spike latencies; this method enables mapping of multiple electrical synapses connected to one neuron – ie, a network – for the first time in living tissue. Together, these two experiments established that TRN neurons form both homocellular and heterocellular electrical synapses in small clusters of 1 to 4 synapses. The discovery of heterocellular electrical synapses in TRN is of particular interest, as heterocellular electrical synapses should coordinate activity of TRN neurons participating in separate thalamocortical channels.
</p
Holographic Techniques for Strongly Coupled Matter
No full textThe holographic AdS/CFT correspondence has provided novel techniques for studying the dynamical behavior of strongly interacting quantum systems and emergent phases of matter. In this thesis we examine the applications of the holographic correspondence in these systems, with a specific focus on computing their transport properties. We present a detailed study of the effects of broken rotational symmetry on the temperature dependence of the shear viscosity of a strongly interacting fluid. This has applications for general anisotropic strongly coupled fluids and more notably for the physics of the quark gluon plasma. We also introduce an efficient holographic method for computing shear and bulk viscosities that takes into account the effects of finite coupling and a finite number of colors. This is achieved working in a gravitational theory that includes higher derivative corrections to the low- energy Einstein action.</p
Quasisymmetric Characteristic Compatible Sn-Modules
No full textThe Frobenius characteristic map connects the representation theory of Sn with symmetric function theory. In particular, the irreducible Sn-submodules are mapped to the Schur functions, a basis of symmetric functions Sym. In the late 1970s, P.N. Norton studied the representation theory of the 0-Hecke algebra Hn(0), a deformation of Sn that is not semisimple. In her work, Norton classified both the projective indecomposable Hn(0)-modules and the irreducible Hn(0)-modules. Mirroring the representation theory of Sn, Krob and Thibon in 1996 defined two Frobenius-type maps on the representations of Hn(0). They defined a noncommutative Frobenius characteristic map, linking noncommutative symmetric functions (NSym) to the projective indecomposable Hn(0)-modules, and a quasisymmetric Frobenius map, linking irreducible Hn(0)-modules to the quasisymmetric functions QSym. In this dissertation, a direct link between the classical symmetric Frobenius characteristic and the more modern quasisymmetric Frobenius characteristic is explored. In particular, a general approach to deforming a Sn-action to a Hn(0)-action such that the quasisymmetric Frobenius characteristic is equal to the symmetric Frobenius characteristic is shown to work for every Sn-module. This approach suggests a quasisymmetric compatibility, which is defined and discussed in the context of several examples.</p
Phase Tropical Homology and Intersection Lattices
No full textWe address a number of topics related to Tropical Geometry and Very Affine Hypersurfaces.We proved a conjecture by Arnal [1] regarding the relationship between simplicial algebraic hypersurfaces and their coamoeaba. We also provide a description of how to lift tropical intersection
lattices to intersection lattices on a phase-tropical hypersurface. Lastly, we develop code which can
compute tropical intersection lattices, and detail a number of example computations.
</p
Applications of Operations Research in Appointment Scheduling and Readmission Prediction
No full textDue to increasing healthcare expenditures and an increasing demand for healthcare services,hospitals are under growing pressure to improve the efficiency of their operations.
In the first part of this dissertation, we study a particular healthcare scheduling problem, dynamic
and stochastic appointment scheduling in the presence of patient unpunctuality, no shows,
walk-ins, and provider lateness with a single served and fixed patient sequence. The aim is to optimize
the decision of scheduling an arrival time for each patient, or equivalently, assigning appointment
intervals between patients, along with the first patient\u27s arrival time. We develop an
efficient and novel solution algorithm, Sample Average Approximation Gradient Descent based
on perturbation analysis.
In the second part, we remove the assumption that the probability distributions of the uncertain
parameters in the scenario creation are known. We also assume that the early patients are
treated as punctual, and the number of patients is static. We develop a distributionally robust
optimization model with a moment ambiguity set constructed to represent the uncertainty of
the various distributions. We derive an equivalent mixed integer linear programming (MILP) reformulation
of our DRO model and utilize Bender\u27s decomposition algorithm.
The third part of this dissertation focuses on a particular intersection between pharmaceutical
industry and hospitals. We present our methodology on characterizing the contribution of pneumonia
and influenza to the burden of readmissions among a population of Medicare Advantage
patients aged 65 and over. We use machine learning algorithms for predicting 30-day preventable
readmissions of Medicare patients, and in order to identify patient characteristics that are risk
factors of their preventable readmissions. Understanding patient risk characteristics can lead to the interventions and policymaking decisions.</p
Out of Body: Depersonalization in Toni Morrison\u27s Novels
No full textMy thesis will explore how Morrison uses formal techniques to bring the depersonalization many of her characters experience to life for her audience: specifically, nonlinear narrative structure, fragmented sentence structure, and internal dialogue highlight the characters\u27 depersonalization. I will explore this idea through Morrison\u27s novels The Bluest Eye, Beloved, and A Mercy, which feature characters who have experienced extreme traumas and seem to have developed depersonalization as a result
Functional Nano-Drug Conjugates For Wall Repair In Abdominal Aortic Aneurysms
No full textAbdominal aortic aneurysms (AAAs) are localized expansion of the abdominal aorta, characterized by chronic proteolytic breakdown of the structural elastic matrix (elastin and collagen) by matrix metalloproteases (MMPs). This expansion culminates to a fatal pre-rupture stage necessitating immediate surgical intervention but is often associated with significant risks. Challenges in regenerative repair of the elastic matrix include poor electrogenicity of diseased or adult smooth muscle cells, intricate pathophysiology, and reduced nitric oxide (NO) levels due to endothelial dysfunction. The pro-inflammatory milieu of AAAs further complicates this process, emphasizing the urgent need for alternative treatment strategies that provide continuous stimuli promoting elastogenesis and counteracting proteolytic breakdown of the elastic matrix. Our previous study, where we demonstrated the exogenous activation of abdominal aortic smooth muscle cells (aHASMCs) with the nitric oxide (NO) donor drug, sodium nitroprusside (SNP), resulted in increased NO bioavailability and enhanced regenerative repair of the elastic matrix in vitro. However, translating this success to real-time AAA scenarios faces challenges when attempting oral SNP delivery, given the short half-life of released NO and risk of degradation by biological and biochemical barriers, ultimately diminishing drug bioavailability to the target tissue. To tackle these challenges, this thesis introduces lipid nanoparticle charge grafts (LCGs) with the goal of facilitating a slow, long-term release of SNP upon it periadventitial placement, providing a sustained and continuous stimulus for regenerative repair efforts in the AAA wall. The thesis presents strong evidence supporting LCGs as an effective treatment option in in-vitro aHASMC cultures, complemented by a limited proof of concept in an ex-vivo AAA model. This ex-vivo model involves a decellularized porcine carotid artery recellularized with aHASMCs, showcasing LCGs\u27 proficiency in elastin regeneration and inhibition of proteolytic activities of MMP2. The thesis accentuates the potential of LCGs in addressing the challenges of regenerative repair within the complex AAA environment, marking a significant stride toward more effective treatments for this life-threatening condition.The primary goal of this thesis was to assess the feasibility of an alternative treatment strategy aimed at impeding or reversing AAA pathophysiology. This was achieved by demonstrating the regenerative potential of SNP and SNP-releasing LCGs through in-vitro and ex-vivo AAA models. The study sought to elucidate the underlying mechanisms behind these regenerative capabilities. Our in-vitro results demonstrated that exogenous delivery of SNP exhibits the ability to upregulate elastic matrix assembly, crosslinking, and maturation while concurrently inhibiting matrix metalloproteinases (particularly MMP2) through NO-mediated mitogen-activated protein kinase (MAPK) inhibition. The utilization of LCGs demonstrates a significant downregulation of MMP2, coupled with an upregulation of key proteins crucial for ECM maintenance (LOX, elastin, TIMP2 and TIMP4) and vascular health. In an ex-vivo decellularized porcine carotid artery (dPCA) AAA model, recellularized with aHASMCs, LCGs exhibit cytocompatibility, allowing cell growth, and evidence of anti-proteolysis (MMP2 downregulation) and elastic matrix neoassembly (upregulated elastin). The sustained release behavior of SNP from LCGs, as demonstrated through the Korsmeyer-Peppas model, suggests controlled and prolonged therapeutic SNP delivery. Addressing the short half-life of NO through LCGs presents an opportunity for more sustained and effective ECM repair. This study introduces a novel strategy for LCG development, successfully conjugating bLNPs onto bPCL meshes to create efficient functional nanodrug conjugates for localized SNP delivery. The promising results signify a significant advancement towards more effective treatments for AAA, offering insights into the dynamic role of NO in maintaining vascular health and ECM homeostasis
Hydrodynamics of schooling: Stability and performance of oscillatory swimmers at high Reynolds numbers
No full textThis dissertation work examines the hydrodynamic interactions that occur in fish schools. Oscillatory foils and fish-like robots are used as simple models to represent fish in a school with varying degrees of fidelity. Experiments are designed to study the stability and energetics of schooling formations at high Reynolds numbers. This methodology allows for a detailed observation of the flowfields, forces, and energetics and, consequently, a complete hydrodynamic description of the interactions that occur among schooling individuals. A novel robotic platform is developed to study the two-dimensional stability of self-propelled oscillatory swimmers. For the first time for high Reynolds number swimmers, the spontaneous self-organization of two two-dimensionally unconstrained pitching foils into a side-by-side formation is observed and the swimmers further enjoy at 15\% increase in their swimming speed. A force map framework, constructed from time-averaged force measurements of constrained swimmers, is shown to successfully predict the stability features of a school of unconstrained swimmers. Three constrained 3D pitching foils (\AR = 3) are used to study the mechanisms and energetics of a small school, specifically probing the classic diamond formation, a canonical arrangement largely studied due to its symmetry and hypothesized performance benefits. Significant collective performance gains of the trio are observed of approximately a doubling in thrust and a 60\% increase in efficiency over three foils in isolation. Three distinct hydrodynamic mechanisms are also examined: vortex-body interactions, body-body interactions and the drafting mechanism. Compact schooling formations yield the highest measured energetic enhancements, and the drafting mechanism does not provide a hydrodynamic benefit for pitching hydrofoils. When a foil is directly in the jet flow downstream of another individual its performance is affected by the oncoming wake, and the spatial phase is traditionally used to define the synchronization between the wake flow and the foil kinematics. It is shown that: (1) a spatial phase model must use the actual wake wavelength rather than an assumed wake wavelength (freestream velocity times the period of oscillation) to predict performance peaks and troughs, and (2) a spatial phase model can predict the performance of a follower foil with the flowfield information from \textit{any upstream leader foils alone}, i.e. the optimal placement of a follower downstream of leaders is known from the flows produced by the leaders. Finally, a high-fidelity experiment with 3D biorobots, inspired by the morphological features of the yellow-fin tuna, is performed to examine the streamwise stability of two interacting fish-like swimmers. A side-by-side formation is verified to be one-dimensionally stable in the streamwise direction. The stable formation was observed for both in-phase and out-of-phase tail beat synchronizations, in contrast to previous results for heaving hydrofoils for which the in-phase synchrony is unstable. The findings reported in this dissertation provide novel insights on the self-organization of fish schools and contribute to describe the hydrodynamic mechanisms that determine the energetics of swimming in a group
Modulated light transmission via magneto-responsive Janus particle chains
No full textJanus particles, amphiphilic particles having two different physical and/or chemical attributes on different hemispheres, have intrigued researchers since their original conception by Pierre De Gennes in 1991. Fabrication of magnetic Janus particles having an iron oxide cap enables control of their orientation and assembly, with potential applications in displaying devices and drug delivery. In this thesis, an effective large-scale fabrication route of magnetic Janus particles is described, particle suspensions of magnetically responsive Janus particles are studied as variable emissivity fluids, and the microscopic structure of Janus particle chains in suspension is analyzed. The promise of Janus particles as bulk additives for responsive complex fluids has been limited by the inability to scale up Janus particle functionalization. Magnetic Janus particles having the high fidelity and monodispersity in both size and surface functionalization are fabricated utilizing particle monolayer formation and physical vapor deposition (PVD). The production rate of Janus particles is enhanced with the roll-to-roll route of depositing particle monolayers of silica or polymer colloids with various metals. Utilizing this new process for fabrication, bulk suspensions of magnetically responsive Janus particles under varying concentrations and magnetic field strengths have been utilized for variable emissivity fluids. Experimental measurements of light transmission agree well with linear models and ray tracing simulations. This work lays the foundation for further investigations of chain dynamics in manipulated magnetic fields in both Newtonian and non-Newtonian fluids. The optical response to an external magnetic field can be qualitatively explained by a kinematic model, and the time-related details of their response lead to a dynamic analysis of the averaged behavior of the Janus particle suspension microstructure. In the kinematic model, a variation of the Beer-Lambert model and ray tracing simulations capture the behavior of the experimentally measured difference in intensity between magnetically activated and nonactivated Brownian suspensions. In the dynamic analysis, different terms in a Langevin equation are discussed. Likewise, the Mason number, defined as the ratio of hydrodynamic force and magnetic force, describes the different amplitudes observed. Microscopic images obtained in both two-dimensional and three-dimensional scans via confocal microscopy reveal Janus particle chaining and microstructure evolution under static and dynamic fields. Slowing the dynamics and sedimentation using a highly viscous fluid enables in-situ observation of the dynamics of the microstructure. The shape, the size, the aspect ratio, and the orientation of each aggregate are obtained. The results describe the dynamics of chain growth, chain coalescence, chain realignment, chain distortion, and chain breaking. In summary, these findings can provide guidance on developing high performance magneto-optical suspensions and gels