Post-traumatic stress disorder (PTSD) is initiated by traumatic-stress exposure and manifests into a collection of symptoms including increased anxiety, sleep disturbances, enhanced response to triggers, and increased sympathetic nervous system arousal. PTSD often co-occurs with alcohol use disorder. Only some individuals experiencing traumatic stress develop PTSD and a subset of individuals with PTSD develop co-occurring alcohol use disorder. Both men and women are at risk to develop PTSD and co-occurring alcohol use disorder when exposed to a traumatic event. Age at which the traumatic event occurred is also a major factor in developing PTSD and co-occurring alcohol use disorder. If exposure occurs during childhood or adolescence, individuals may be more resilient to these stresses compared to older individuals. Factors including sex and age have shown individual differences in developing PTSD and co-occurring alcohol use disorder and severity of disorders. However, what factors following traumatic stress exposure that predict resilience or vulnerability remain unknown. To investigate the basis of the individual responses to traumatic stress, single prolonged stress (SPS) a validated rodent model of traumatic stress was applied to young adult male and female rats and adolescent female rats. Individual behavioral responses to traumatic stress were characterized using anxiety-like behaviors with open field and elevated plus maze tests, fear-like behaviors with cue-reactivity, and depression-like behaviors with the forced swim test. Ethanol consumption following traumatic stress or control handling was measured by allowing individual rats to self-administer ethanol using an intermittent two bottle choice procedure for 8 weeks. Correlations within age and sex were used to determine which behavioral factors were predictive of ethanol consumption. Results demonstrate that different behavioral endpoints were predictive of subsequent drinking in males and females, and in adult and adolescent groups. Fear-like behavior was predictive of drinking in young adult males. Depression-like behavior was predictive of adolescent female ethanol consumption. Anxiety-like behavior was predictive of ethanol drinking in young adult females. These results indicate that resilience and vulnerability manifest differently after traumatic stress exposure depending on age and biological sex. Young adult females were further analyzed using an artificial intelligence algorithm that was developed to predict resilient and vulnerable individuals based on data from anxiety testing and ethanol consumption. Using the algorithm with the factors of time in center on the open field test and open arm entries on the elevated plus maze revealed that the population consisted of 3 groups with 24% classified as resilient and 41% classified as susceptible to high ethanol drinking. The artificial intelligence model was implemented in a second experiment to identify resilient and vulnerable adult female rats before ethanol exposure. Using the resilient and vulnerable animals identified from the artificial intelligence algorithm, analyses of neuropeptide Y (NPY) and its receptors Y1 and Y2 in the central nucleus of the amygdala (CeA), basolateral amygdala (BLA), and bed nucleus stria terminalis (BNST) were performed. The CeA, BLA, and BNST are important regions for PTSD and co-occurring alcohol use disorder. Results demonstrate that resilient rats had higher expression of Y2 mRNA in the CeA compared with vulnerable and control rats. In the BLA, the vulnerable rats had higher levels of Y1 compared to controls. In the BNST, NPY was elevated in resilient animals compared to controls. The results of the study show that an artificial intelligence algorithm can identify individual differences in response to traumatic stress and subsequent ethanol drinking, and the NPY pathway is differentially altered following traumatic stress exposure in resilient and vulnerable populations. Understanding neurochemical alterations following traumatic-stress exposure is critical in developing prevention strategies for the vulnerable phenotype and will help further development of novel therapeutic approaches for individuals suffering from PTSD and alcohol use disorder. / Biomedical Neuroscience

Despite tremendous improvement in the development of antiretroviral therapy (ART) for treatment of human immunodeficiency virus‐1 (HIV‐1), a cure is currently missing. One of the main challenges in HIV-1 treatment is postulated to be due to the accumulation of HIV-1 particles in the latent tissue reservoirs, where they are protected from both antiretrovirals (ARVs) and immune surveillance mechanisms. Interestingly, it has been shown that binding of the monosialodihexosylgangliosid (GM3) to CD169 (Siglec‐1) plays an important role in the glycoprotein‐independent sequestration of HIV‐1 particles in non-lysosomal virus‐containing compartments (VCCs) in CD169+ myeloid cells. Therefore, VCCs represent potential virus latent tissue reservoirs and provide protection for virus from immunological surveillance. In this dissertation, GM3‐functionalized lipid-coated polymer nanoparticles (NPs) referred to as HIV-1 mimicking lipid-coated polymer NPs were designed to target the VCCs and deliver ARVs to achieve full virus eradication. HIV-1 mimicking lipid-coated polymer NPs were assembled using poly(lactic‐co‐glycolic) acid (PLGA) as well as polylactic acid (PLA) polymer cores. After a systematic characterization of HIV-1 mimicking lipid-coated polymer NPs, the NPs were applied as a virus mimicking model to investigate the effect of core stiffness on NP binding, uptake, and intracellular fate mediated by GM3‐CD169 binding in CD169+ macrophages. Our results suggest that GM3‐CD169‐mediated sequestration of NPs in non-lysosomal VCC-like compartments is not only regulated by ligand–receptor interactions but also determined by the core stiffness of polymer NPs. Subsequently, ARV-loaded HIV-1 mimicking polymer NPs were developed as long-acting nanocarriers for delivery of a combination of Rilpivirine (RPV) and Cabotegravir (CAB) to VVCs in CD169+ primary human monocyte-derived macrophages (MDMs) for a duration of at least 28 days. Cellular drug concentrations and inhibitory effects obtained for GM3- or phosphatidylserine (PS) -mediated binding to CD169+ MDMs were quantified, and GM3-presenting PLA NPs were found to provide the most efficient and longest lasting inhibition of viral infection. Our findings pave the path towards designing a new class of polymer NPs aimed to target VCCs in CD169+ macrophages and to utilize CD169+ tissue resident macrophages as cellular drug depots for eradicating viral latent tissue reservoirs or as long-acting prevention and treatment strategies against HIV-1 infection.

High-precision neuromodulation with high efficacy poses great importance in neuroscience. Also, ultrasound frequency in sub-MHz gives greater penetration depth of neuromodulation in vivo, and has many different biomedical applications, such as ocular drug delivery, arte- rial hyperplasia reduction, etc. Here we fabricated a series of carbon-nanotube (CNT) and polydimethylsiloxane (PDMS) mixture films with different CNT concentration and thick- ness, which can generate highly efficient sub-MHz ultrasound. We discovered that 0.8% CNT concentration CNT/PDMS films gave approximately 1 MHz frequency, below which we achieved tunable frequency in sub-MHz area. Additionally, we adjusted the PDMS base to curing agent ratio and discovered that the 2:1 ratio’s optoacoustic conversion ef- ficiency is 1.6 times greater than the typical 10:1 ratio. Our work offers a potential way to increase photoacoustic conversion rate by optimizing CNT/PDMS films’ fabrication for further photoacoustic-related applications.

While combination anti-retroviral therapy (cART) has improved the length and quality of life of individuals living with HIV-1 infection, the prevalence of HIV-associated neurocognitive disorders (HAND) has increased and remains a significant clinical concern. The neuropathogenesis of HAND is not completely understood, however, latent HIV infection in the central nervous system (CNS) and chronic neuroinflammation are believed to play a prominent role. CNS-associated macrophages and resident microglia are significant contributors to CNS inflammation and constitute the chief reservoir of HIV-1 infection in the CNS. Previous studies from our lab suggest monocyte/macrophage invasion of the CNS in HIV may be driven by altered monocyte/macrophage homeostasis. We have reported expansion of a monocyte subset (CD14+CD16+CD163+) in peripheral blood of HIV+ patients that is phenotypically similar to macrophages/microglia that accumulate in the CNS as seen in post-mortem tissue. The factors driving the expansion of this monocyte subset are unknown, however, signaling through cFMS, a type III receptor tyrosine kinase (RTK), may play a role. Macrophage-colony stimulating factor (M-CSF), a ligand of cFMS, has been shown to be elevated in the cerebral spinal fluid (CSF) of individuals with the most severe form of HAND, HIV-associated dementia (HAD). M-CSF promotes a Macrophage-2-like phenotype and increases CD16 and CD163 expression in cultured monocytes. M-CSF has also been shown to increase the susceptibility of macrophages to HIV infection and enhance virus production. These findings, in addition to the known function of M-CSF in promoting macrophage survival, supports a role for M-CSF in the development and maintenance of macrophage viral reservoirs in tissues where these cells accumulate, including the CNS. Interestingly, a second ligand for cFMS, IL-34, was recently identified and reported to share some functions with M-CSF, suggesting that both ligands may contribute to HIV-associated CNS injury and AIDS pathogenesis. Through immunohistochemical studies using a relevant animal model of HIV infection, SIV infected rhesus macaques, we reported the presence of M-CSF and IL-34 in the brains of seronegative and SIV+ animals, for the first time, and identified spatial differences in the expression of these ligands. Important to our interest in viral persistence in the CNS, we observed the predominance of M-CSF expression in brain to be by cells that comprise perivascular cuffs and nodular lesions, which contain monocytes/ macrophages that have migrated into the CNS. IL-34 appeared to be a tissue-specific ligand expressed by resident microglia. Like M-CSF, we found that IL-34 also increased the frequency of CD16+CD163+ monocytes in vitro. We further investigated the potential of cFMS inhibition as a means to abrogate macrophage-2-like immune polarization using the small molecule tyrosine kinase inhibitor (TKI), GW2580. The addition of GW2580 abolished cFMS ligand-mediated increases in CD16+CD163+ monocyte frequency in human peripheral blood mononuclear cells (PBMC) as well as virus production in HIV infected primary human microglia. Furthermore, we found cFMS-mediated upregulation of CD16 and CD163 to be relevant to an additional disease process, high-grade astrocytomas, suggesting that M-CSF and IL-34 may be mediators of other neuroinflammatory diseases, as well. We hope these findings will provide insight into the role of altered monocyte/macrophage homeostasis in HIV disease and identify a novel strategy for targeting long-lived cellular reservoirs of HIV infection through restored immune homeostasis. / Biomedical Neuroscience

<p>Be-doped InP nanowires were grown by the gold-assisted vapour-liquid-solid mechanism in a gas source molecular beam epitaxy system. The nanowires were characterized by scanning and transmission electron microscopy. With increased doping, the dependence of the length on diameter [L(D)] underwent an unusual transition from the diffusion-limited 1/D² relationship to one that increased before saturating. Doping influences on crystal structure and radial growth have been reported previously, but in the absence of these effects it is speculated that the beryllium introduces an increase in the steady-state chemical potential of the catalyst, and a barrier to nucleation. A model is presented relating the diffusion- and nucleation-limited regimes.</p> <p>Additionally, the progressive increase of dopant incorporation was verified by secondary ion mass spectrometry. Samples were transformed into a "bulk-like" material by spin-coating with cyclotene to enable depth profiling. Carrier concentrations were inferred through comparison with a thin film reference, and agreed in order of magnitude with the nominal doping values.</p> <p>Dopant activation was investigated through micro-photoluminescence experiments, and showed peak emissions between 1.49 eV and 1.50 eV for undoped samples, transitioning with increased doping to 1.45-1.46 eV. The difference between the dominant peak energies was consistent with differences reported for comparable epitaxially-grown thin film samples. Bandgap narrowing was also observed at high levels of doping, and was consistent with theoretical predictions.</p> <p>As a whole, the work presented here provides a different perspective on the effects of doping on nanowire growth, demonstrated through the specific system of Be-doped InP. The findings have implications for predictable and consistent nanowire device design, and suggestions for avenues of future research are provided.</p> / Master of Applied Science (MASc)

Spike-timing dependent plasticity (STDP), a widespread synaptic modification mechanism, is sensitive to correlations between presynaptic spike trains, and organizes neural circuits in functionally useful ways. In this dissertation, I study the structures arising from STDP in a population of synapses with an emphasis on the interplay between synaptic stability and Hebbian competition, explained in Chapter 1. Starting from the simplest description of STDP which relates synaptic modification to the intervals between pairs of pre- and postsynaptic spikes, I show in Chapter 2 that stability and Hebbian competition are incompatible in this class of ``pair-based'' STDP models, either when hard bounds or soft bounds are imposed to the synapses. In chapter 3, I propose an alternative biophysically inspired method for imposing bounds to synapses, i.e. introducing a small temporal shift in the STDP window. Shifted STDP overcomes the incompatibility of synaptic stability and competition and can implement both Hebbian and anti-Hebbian forms of competitive plasticity. In light of experiments the explored a variety of spike patterns, STDP models have been augmented to account for interactions between multiple pre- and postsynaptic action potentials. In chapter 4, I study the stability/competition interplay in three different proposed multi-spike models of STDP. I show that the ``triplet model'' leads to a partially steady-state distribution of synaptic weights and induces Hebbian competition. The ``suppression model'' develops a stable distribution of weights when the average weight is high and shows predominantly anti-Hebbian competition. The "NMDAR-based" model can lead to either stable or partially stable synaptic weight distribution and exhibits both Hebbian and anti-Hebbian competition, depending on the parameters. I conclude that multi-spike STDP models can produce radically different effects at the population level depending on how they implement multi-spike interactions. Finally in chapter 5, I focus on the types of global structures that arise from STDP in a recurrent network. By analyzing pairwise interactions of neurons through STDP and also numerical simulations of a large network, I show that conventional pair-based STDP functions as a loop-eliminating mechanism in a network of spiking neurons and organizes neurons into in- and out-hubs. Loop-elimination increases when depression dominates and decreases when potentiation dominates. STDP with dominant depression implements a buffering mechanism for network firing rates, and shifted STDP can generate recurrent connections in a network, and also functions as a homeostatic mechanism that maintains a roughly constant average value of the synaptic strengths. In conclusion, studying pairwise interactions of neurons through STDP provides a number of important insights about the structures that arise from this plasticity rule in large networks. This approach can be extended to networks with more complex STDP models and more structured external input.

This Thesis addresses the fundamental problem of controlling transport through a metal-organic interface by studying electronic and mechanical properties of single organic molecule-metal junctions. Using a Scanning Tunneling Microscope (STM) we image, probe energy-level alignment and perform STM-based break junction (BJ) measurements on molecules bound to a gold surface. Using Scanning Tunneling Microscope-based break-junction (STM-BJ) techniques, we explore the effect of binding geometry on single-molecule conductance by varying the structure of the molecules, metal-molecule binding chemistry and by applying sub-nanometer manipulation control to the junction. These experiments are performed both in ambient conditions and in ultra high vacuum (UHV) at cryogenic temperatures. First, using STM imaging and scanning tunneling spectroscopy (STS) measurements we explore binding configurations and electronic properties of an amine-terminated benzene derivative on gold. We find that details of metal-molecule binding affect energy-level alignment at the interface. Next, using the STM-BJ technique, we form and rupture metal-molecule-metal junctions ~104 times to obtain conductance-vs-extension curves and extract most likely conductance values for each molecule. With these measurements, we demonstrated that the control of junction conductance is possible through a choice of metal-molecule binding chemistry and sub-nanometer positioning. First, we show that molecules terminated with amines, sulfides and phosphines bind selectively on gold and therefore demonstrate constant conductance levels even as the junction is elongated and the metal-molecule attachment point is modified. Such well-defined conductance is also obtained with paracyclophane molecules which bind to gold directly through the ð system. Next, we are able to create metal-molecule-metal junctions with more than one reproducible conductance signatures that can be accessed by changing junction geometry. In the case of pyridine-linked molecules, conductance can be reliably switched between two distinct conductance states using sub-nanometer mechanical manipulation. Using a methyl sulfide linker attached to an oligoene backbone, we are able to create a 3-nm-long molecular potentiometer, whose resistance can be tuned exponentially with Angstom-scale modulations in metal-molecule configuration. These experiments points to a new paradigm for attaining reproducible electrical characteristics of metal-organic devices which involves controlling linker-metal chemistry rather than fabricating identically structured metal-molecule interfaces. By choosing a linker group which is either insensitive to or responds reproducibly to changes in metal-molecule configuration, one can design single molecule devices with functionality more complex than a simple resistor. These ambient temperature experiments were combined with UHV conductance measurements performed in a commercial STM on amine-terminated benzene derivatives which conduct through a non-resonant tunneling mechanism, at temperatures varying from 5 to 300 Kelvin. Our results indicate that while amine-gold binding remains selective irrespective of environment, conductance is not temperature independent, in contrast to what is expected for a tunneling mechanism. Furthermore, using temperature-dependent measurements in ambient conditions we find that HOMO-conducting amines and LUMO-conducting pyridines show opposite dependence of conductance on temperature. These results indicate that energy-level alignment between the molecule and the electrodes changes as a result of varying electrode structure at different temperatures. We find that temperature can serve as a knob with which to tune transport properties of single molecule-metal junctions.

Graphenes transport properties have been extensively studied in the 10 years since its discovery in 2004, with ground-breaking experimental observations such as Klein tunneling, fractional quantum Hall effect and Hofstadters butterfly. Though, so far, it turned out to be rather poor on complex correlated electronic ground states and phase transitions, despite various theoretical predictions. The purpose of this thesis is to help understanding the underlying theoretical and experimental reasons for the lack of strong electronic interactions in graphene, and, employing graphenes high tunability and versatility, to identify and alter experimental parameters that could help to induce stronger correlations. In particular graphene holds one last, not yet experimentally discovered prediction, namely exhibiting intrinsic superconductivity. With its vanishingly small Fermi surface at the Dirac point, graphene is a semi-metal with very weak electronic interactions. Though, if it is doped into the metallic regime, where the size of the Fermi surface becomes comparable to the size of the Brillouin zone, the density of states becomes sizeable and electronic interactions are predicted to be dramatically enhanced, resulting in competing correlated ground states such as superconductivity, magnetism and charge density wave formation. Following these predictions, this thesis first describes the creation of metallic graphene at high carrier doping via electrostatic doping techniques based on electrolytic gates. Due to graphenes surface only properties, we are able to induce carrier densities above n>10¹⁴cm⁻²(εF>1eV) into the chemically inert graphene. While at these record high carrier densities we yet do not observe superconductivity, we do observe fundamentally altered transport properties as compared to semi-metallic graphene. Here, detailed measurements of the low temperature resistivity reveal that the electron-phonon interactions are governed by a reduced, density dependent effective Debey temperature - the so-called Bloch-Grüneisen temperature ΘBG. We also probe the transport properties of the high energy sub-bands in bilayer graphene by electrolyte gating. Furthermore we demonstrate that electrolyte gates can be used to drive intercalation reactions in graphite and present an all optical study of the reaction kinetics during the creation of the graphene derived graphite intercalation compound LiC₆, and show the general applicability of the electrolyte gates to other 2-dimensional materials such as thin films of complex oxides, where we demonstrate gating dependent conductance changes in the spin-orbit Mott insulator Sr₂IrO₄. Another, entirely different approach to induce superconducting correlations into graphene is by bringing it into proximity to a superconductor. Although not intrinsic to graphene, Cooper pairs can leak in from the superconductor and exist in graphene in the form of phase-coherent electron-hole states, the so-called Andreev states. Here we demonstrate a new way of fabricating highly transparent graphene/superconductor junctions by vertical stacking of graphene and the type-II van der Waals superconductor NbSe₂. Due to NbSe₂'s high upper critical field of Hc₂= 4 T we are able to test a long proposed and yet not well understood regime, where proximity effect and quantum Hall effect coexist.

<p> This thesis describes the synthesis of water-soluble dendron-conjugated gold nanoparticles (Den-AuNPs) with various average core sizes and the evaluation of stability, cytotoxicity, and cell permeability and uptake of these materials. The characterization of Den-AuNPs using various instruments including transmission electron microscopy (TEM), matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF-MS), 1H NMR, FT-IR, and UV-vis spectroscopy confirms the dendron conjugation to the glutathione-capped gold nanoparticles (AuNPs). The stability of AuNPs and Den-AuNPs in solutions of different pH and salt concentration was determined by monitoring changes in surface plasmon bands of gold using UV-vis spectroscopy. The Den-AuNPs were found to be more stable than the precursor AuNPs maintaining their solubility at the pHs higher than 4 and with the salt concentrations of up to 100 mM. The improved stability of Den-AuNPs suggests that the post-functionalization of thiol-capped gold nanoparticle surfaces with dondrons can further improve the physiological stability and biocompatibility of gold nanoparticle-based materials. Cytotoxicity studies with AuNPs and Den-AuNPs with and without flourophores were also performed by examining cell viability for 3T3 fibroblasts using a MTT cell proliferation assay. The conjugation of dendrons to the AuNPs with flourophores was able to decrease the cytotoxicity brought about by the flourophores. The successful uptake of Den-AuNPs in mouse fibroblast 3T3 cells shows the physiological viability of the hybrid materials.</p>

<p> Natural manganese oxides are generally formed in surficial environments that are near ambient temperature and water-rich, and may be exposed to wet-dry cycles and a variety of adsorbate species that influence dramatically their level of hydration. Manganese oxide minerals are often poorly crystalline, nanophase, and hydrous. In the near-surface environment they are involved in processes that are important to life, such as water column oxygen cycling, biomineralization, and transport of minerals/nutrients through soils and water. These processes, often involving transformations among manganese oxide polymorphs, are governed by a complex interplay between thermodynamics and kinetics. Manganese oxides are also used in technology as catalysts, and for other applications. </p><p> The major goal of this dissertation is to examine the energetics of bulk and nanophase manganese oxide phases as a function of particle size, composition, and surface hydration. Careful synthesis and characterization of manganese oxide phases with different surface areas provided samples for the study of enthalpies of formation by high temperature oxide melt solution calorimetry and of the energetics of water adsorption on their surfaces. These data provide a quantitative picture of phase stability and how it changes at the nanoscale. </p><p> The surface energy of the hydrous surface of Mn₃O₄ is 0.96 &plusmn; 0.08 J/m², of Mn₂O₃ is 1.29 &plusmn; 0.10 J/m², and of MnO₂ is 1.64 &plusmn; 0.10 J/m². The surface energy of the anhydrous surface of Mn₃O₄ is 1.62 &plusmn; 0.08 J/m², of Mn₂O₃ is 1.77 &plusmn; 0.10 J/m², and of MnO₂ is 2.05 &plusmn; 0.10 J/m². Supporting preliminary findings (Navrotsky et al., 2010), the spinel phase (Mn₃O₄) has a lower surface energy (more stabilizing) than bixbyite, while the latter has a smaller surface energy than pyrolusite. These differences significantly change the positions in oxygen fugacity&mdash;temperature space of the redox couples Mn₃O₄-Mn₂O₃ and Mn₂O₃-MnO₂ favoring the lower surface enthalpy phase (the spinel Mn₃O₄) for smaller particle size and in the presence of surface hydration. </p><p> Chemisorption of water onto anhydrous nanophase Mn₂O₃ surfaces promotes rapidly reversible redox phase changes at room temperature as confirmed by calorimetry, X-ray diffraction, and titration for manganese average oxidation state. Water adsorption microcalorimetry (in situ) at room temperature measured the strongly exothermic integral enthalpy of water adsorption (-103.5 kJ/mol) and monitored the energetics of the redox phase transformation. Hydration-driven redox transformation of anhydrous nanophase Mn(III)₂O₃, (high surface enthalpy of anhydrous surfaces 1.77 &plusmn; 0.10 J/m²) to Mn(II,III)₃O₄ (lower surface enthalpy 0.96 &plusmn; 0.08 J/m²) occurred during the first few doses of water vapor. Surface reduction of nanoparticle bixbyite (Mn₂O₃) to hausmannite (Mn₃O₄) occurs under conditions where no such reactions are seen or expected on grounds of bulk thermodynamics in coarse-grained materials. </p><p> Layered structure manganese oxides contain alkali or alkaline earth cations and water, are generally fine-grained, and have considerable thermodynamic stability. The surface enthalpies (SE) of layered and tunnel structure complex manganese oxides are significantly lower than those of the binary manganese oxide phases. The SE for hydrous surfaces and overall manganese average oxidation state (AOS) (value in parentheses) are: cryptomelane 0.77 &plusmn; 0.10 J/m² (3.78), sodium birnessite 0.69 &plusmn; 0.13 J/m² (3.56), potassium birnessite 0.55 &plusmn; 0.11 J/m² (3.52), and calcium birnessite 0.41 &plusmn; 0.11 J/m² (3.50). Surface enthalpies of hydrous surfaces of the calcium manganese oxide nanosheets are: &delta;Ca_0.39MnO_2.3nH₂O 0.75 &plusmn; 0.10 J/m² (3.89) and &delta;Ca_0.43MnO_2.3nH₂O 0.57 &plusmn; 0.12 J/m² (3.68). The surface enthalpy of the complex manganese oxides appears to decrease with decreasing manganese average oxidation state, that is, with greater mixed valence manganese (Mn^3+/4+). Low surface energy suggests loose binding of H₂O on the internal and external surfaces and may be critical to catalysis in both natural and technological settings.</p>