Slow dynamics and aging in metastable systems

Dhillon, Param Pash Harvard University 2007 해외박사(DDOD)

This thesis focuses on slow dynamics in glass-forming systems. We tackle a range of problems, including the microscopic nature of elasticity, aging, and the development of a microscopic theoretical description of these systems. First, we connect dynamical heterogeneity and elasticity in a supercooled liquid near the glass transition. Using standard molecular dynamics techniques, we simulate a polydisperse system of hard spheres, and focus on dynamically slow particles. We find that in a supercooled liquid, a cluster of slow particles spans the system at short time scales, but dissociates on longer time scales. However, in the glass, spanning clusters persist for all observable time scales. The lifetime of the spanning cluster is quantitatively connected to the viscoelastic moduli of the system, thus establishing a link between microscopic dynamical heterogeneity and elasticity, the macroscopic feature that distinguishes a glass from a liquid. We then shift to the aging behavior in a stripe-forming system quenched below the order-disorder transition, simulated with Langevin dynamics. In order to study fluctuation-dissipation relations, we develop a field-free method to calculate the integrated response function. This allows us characterize the evolution of this metastable system towards equilibrium. We find that the aging behavior of this system resembles neither simple coarsening nor simple glassy behavior. The aging is more rich than either of these scenarios, bearing some resemblance to that of critical coarsening systems. Finally, we focus on the theoretical description of supercooled liquids near the glass transition, employing a new time-local mode-coupling approach which is motivated by recent work on turbulent systems. We present the results of this approach on a toy model inspired by microscopic considerations, and find some noteable improvements over mode-coupling theory, including the disappearance of the dynamically sharp glass transition and an exponential dependence of the relaxation times on the coupling strength.

Stress and tone in Indo-Aryan languages

Dhillon, Rajdip Kaur Yale University 2010 해외박사(DDOD)

The current work offers a comprehensive examination of stress and tone in ten Indo-Aryan languages, providing novel analyses within Optimality Theory. The languages are divided into three categories: those in which tone is attracted to stress; those in which stress is attracted to tone; and those in which no interaction between stress and tone occurs. The typology which emerges points to a property of prosodic balance occurring in this group of languages, whereby one prosodic feature behaves in a manner which appears complex, while the other feature behaves more simply. Also considered is the use of the notion of pitch accent in classifying these languages. Using Hyman's (2007a, 2007b, 2006) model of prosodic typology, the very idea of pitch accent is questioned and ultimately rejected, in favor of Hyman's prosodic model which only allows for the existence of tone and stress --- two features which can be clearly defined. Finally, a theoretical exposition of the interaction of stress and tone is offered. De Lacy's (2007, 2004a, 2004b) Stringent Markedness approach is evaluated. While possessing the beneficial feature of allowing for the ability to conflate categories, this approach is far too limited as a result of the singular tonal hierarchy it employs, which fails to cover a great deal of tonal phenomena, especially within the Indo-Aryan languages discussed here. Consequently, a multi-hierarchical theory of the interaction of stress and tone is proposed, which calls for the implementation of multiple attested hierarchies in order to accommodate a wider range of stress-tone phenomena and thus provide more cross-linguistic support than the Stringent Markedness approach. The advantage of this work is threefold: the large collection of data on ten languages contained within illustrates the descriptive contribution; the Optimality Theoretic analyses of the languages, along with the emergent typology constitutes the analytical value; and the use of Optimality Theory, the evaluation of the Stringent Markedness approach (de Lacy, 2007; 2004a; 2004b), and the construction of a more comprehensive alternative to Stringent Markedness comprises the theoretical contribution. Not only is this work of appeal to scholars of Indo-Aryan, but also to scholars of Optimality Theory.

Hierarchical optimization of digital CMOS circuits for power, performance and reliability

Dhillon, Yuvraj Singh Georgia Institute of Technology 2005 해외박사(DDOD)

Power consumption and soft-error tolerance have become major constraints in the design of deep-sub-micron (DSM) complimentary metal-oxide-semiconductor (CMOS) circuits. With continued technology scaling, the impact of these parameters is expected to gain in significance. Furthermore, the design complexity continues to increase rapidly due to the tremendous increase in number of components (gates/transistors) on an integrated circuit (IC) every technology generation. This research describes an efficient and general computer-aided-design (CAD) framework for the optimization of critical circuit characteristics such as power consumption and soft-error tolerance under delay constraints with supply/threshold voltages and/or gate sizes as variables. A general technique called Delay-Assignment-Variation (DAV) based optimization was formulated for the delay-constrained optimization of directed acyclic graphs. Exact mathematical conditions on the supply and threshold voltages of circuit modules were developed that lead to minimum overall dynamic and static power consumption of the circuit under delay constraints. A DAV search based method was used to obtain the optimal supply and threshold voltages that minimized power consumption. To handle the complexity of design of reliable, low-power circuits at the gate level, a hierarchical application of DAV based optimization was explored. The effectiveness of the hierarchical approach in reducing circuit power and unreliability, while being highly efficient is demonstrated. The usage of the technique for improving upon already optimized designs is described. An accurate and efficient model for analyzing the soft-error tolerance of CMOS circuits is also developed.

Bias due to exposure misclassification and rising screening levels: A case-control study of prostate-specific antigen (PSA) screening efficacy

Dhillon, Preet Kaur University of Washington 2002 해외박사(DDOD)

We conducted a case-control study to evaluate the association between PSA screening and prostate cancer mortality among 594 male members of Kaiser Permanente of Northern California. Cases were comprised of 297 men aged 50–84 years who died as a result of prostate cancer between 1995–1997 and controls included 297 men who were alive and free of prostate cancer at the time of the matched case's diagnosis. Controls were individually matched on to collect information on screening and follow-up testing over the 10-year reference period leading up to the case's diagnosis (and the corresponding time period in matched controls). To distinguish between screening tests and diagnostic tests, an algorithm was developed, and all tests were classified as probable, possible, unlikely and non-screens. Our results revealed a higher proportion of PSA “screening” in cases relative to controls. This anomalous result—which would suggest that screening could <italic>increase </italic> mortality from prostate cancer—could be the result of bias to the concurrent use of a more widely used screening test, the DRE. This can result in misclassification of PSA's that were ordered in response to a suspicious DRE but mislabeled as “screens” because information on suspicious signs or symptoms was not recorded by the clinician. Nonetheless, using the most stringent criteria possible for screening, we still obtained an adjusted odds ratio greater than one associated with PSA. Another explanation is the rise in the occurrence of PSA testing over time in the population, which differentially increases screening in cases relative to controls. This serves to artificially inflate the odds ratio and underestimate the potential benefit associated with screening. Simulation models were used to assess the possible distortion due to increasing screening rates. The methodological considerations discussed in this paper are relevant for case-control studies of screening efficacy, especially those evaluating screening tests newly introduced into a population.

The effects of including almonds in an energy-restricted diet on weight, body composition, visceral adipose tissue, blood pressure and cognitive function

Dhillon, Jaapna Purdue University ProQuest Dissertations & Theses 2016 해외박사(DDOD)

Inclusion of almonds in an energy restricted diet has been reported to enhance or have no effect on weight loss. Their effects specifically on visceral fat stores during energy restriction have not been widely examined. Additionally, almond consumption has been associated with reduced blood pressure, but whether this is linked to or is independent of changes of body composition has not been examined. Moreover, almond consumption during energy restriction may be an effective strategy for reversing the negative effects of dieting on cognitive performance. The unique nutrient profile of almonds also has the potential to influence cognitive function post-prandially. The post-lunch dip in cognition is a well-established phenomenon of decreased alertness, memory and vigilance after lunch consumption and can be affected by lunch composition. Almonds which are higher in fat and lower in carbohydrate may be able to reduce this post lunch dip in cognition. Consequently, this dissertation had three primary aims. The first aim was to evaluate the effects of almond consumption as part of an energy-restricted diet on weight, visceral and subcutaneous adipose depots and blood pressure compared to a nut-free energy restricted diet. The second aim was to evaluate the effects of almond consumption as part of an energy-restricted diet on cognitive function. The third aim was to evaluate the acute effects of almond consumption on the post-lunch dip in cognitive function. A secondary objective of this dissertation was to develop an analytical approach to identify metabolic profiles associated with almond consumption to ascertain compliance in long term clinical trials. (Abstract shortened by ProQuest.).

Geometric Representations of W-Algebras and Applications to the Quantum Langlands Correspondence

Dhillon, Gurbir ProQuest Dissertations & Theses Stanford Universit 2020 해외박사(DDOD)

Investigating the Strain Properties of Alpha-Synuclein: Implications for Pathogenesis

Dhillon, Jess-Karan Singh University of Florida ProQuest Dissertations & The 2019 해외박사(DDOD)

Synucleinopathies are a class of neurodegenerative disorders characterized by the intracellular deposition of the protein α-synuclein (αS) leading to multiple outcomes, including dementia and parkinsonism. Recent findings support the notion that across the spectrum of synucleinopathies there exist diverse but specific biochemical modifications and/or structural conformations of α-synuclein. These physical states could give rise to protein strain specific prion-like intercellular transmission, which is a proposed model that could, at least partially, explain the spectrum of synucleinopathies disease progression and heterogeneity of inclusion types. To investigate these varied inclusions and the underlying pathophysiology of these diseases requires adequate tools so that disease specific modifications can be appropriately appreciated. With this in mind, I developed and described a panel of antibodies which target several regions of the αS protein that display unique immunoreactivty for pathological forms of αS. Multiple system atrophy (MSA) represents one of the more distinctive synucleinopathies as αS inclusions can be found predominantly within glial cells, dissimilar to the more common synucleinopathies, Parkinson’s disease (PD) and dementia with LSynucleinopathies are a class of neurodegenerative disorders characterized by the intracellular deposition of the protein α-synuclein (αS) leading to multiple outcomes, including dementia and parkinsonism. Recent findings support the notion that across the spectrum of synucleinopathies there exist diverse but specific biochemical modifications and/or structural conformations of α-synuclein. These physical states could give rise to protein strain specific prion-like intercellular transmission, which is a proposed model that could, at least partially, explain the spectrum of synucleinopathies disease progression and heterogeneity of inclusion types. To investigate these varied inclusions and the underlying pathophysiology of these diseases requires adequate tools so that disease specific modifications can be appropriately appreciated. With this in mind, I developed and described a panel of antibodies which target several regions of the αS protein that display unique immunoreactivty for pathological forms of αS. Multiple system atrophy (MSA) represents one of the more distinctive synucleinopathies as αS inclusions can be found predominantly within glial cells, dissimilar to the more common synucleinopathies, Parkinson’s disease (PD) and dementia with Lewy bodies (DLB), which have primarily neuronal inclusions. Biochemical alterations specific to MSA have been described but a thorough investigation of these unique and disease specific inclusions is warranted, especially given the prion-like nature of synucleinopathies and the putative existence of protein strains that could play a fundamental role in synucleinopathy pathogenesis. Utilizing a comprehensive antibody panel I demonstrated the existence of diverse αS biochemical and immunohistochemical permutations differentiating MSA and DLB. Additionally, these studies revealed substantial variation across MSA patients and even within an individual MSA patient across brain regions through diverse immunoreactivity profiles which highlights the necessity of using several antibodies for evaluation of αS pathology. Finally, I evaluate the notion that αS pathology within MSA represents an “authentic prion” by assessing propagation through a new host and maintenance of strain-like properties. My studies highlight that although seeded conformational templating can participate in the induction and spread of αS inclusion pathology, MSA aggregated αS does not fully recapitulate cellular specific and conserved strain properties of an authentic prion.

Advances in spectral learning with applications to text analysis and brain imaging

Dhillon, Paramveer Singh University of Pennsylvania 2014 해외박사(DDOD)

Spectral learning algorithms are becoming increasingly popular in data-rich domains, driven in part by recent advances in large scale randomized SVD, and in spectral estimation of Hidden Markov Models. Extensions of these methods lead to statistical estimation algorithms which are not only fast, scalable, and useful on real data sets, but are also provably correct. Following this line of research, we make two contributions. First, we propose a set of spectral algorithms for text analysis and natural language processing. In particular, we propose fast and scalable spectral algorithms for learning word embeddings -- low dimensional real vectors (called Eigenwords) that capture the "meaning" of words from their context. Second, we show how similar spectral methods can be applied to analyzing brain images. State-of-the-art approaches to learning word embeddings are slow to train or lack theoretical grounding; We propose three spectral algorithms that overcome these limitations. All three algorithms harness the multi-view nature of text data i.e. the left and right context of each word, and share three characteristics. 1). They are fast to train and are scalable. 2). They have strong theoretical properties. 3). They can induce context-specific embeddings i.e. different embedding for "river bank" or "Bank of America". They also have lower sample complexity and hence higher statistical power for rare words. We provide theory which establishes relationships between these algorithms and optimality criteria for the estimates they provide. We also perform thorough qualitative and quantitative evaluation of Eigenwords and demonstrate their superior performance over state-of-the-art approaches. Next, we turn to the task of using spectral learning methods for brain imaging data. Methods like Sparse Principal Component Analysis (SPCA), Non-negative Matrix Factorization (NMF) and Independent Component Analysis (ICA) have been used to obtain state-of-the-art accuracies in a variety of problems in machine learning. However, their usage in brain imaging, though increasing, is limited by the fact that they are used as out-of-the-box techniques and are seldom tailored to the domain specific constraints and knowledge pertaining to medical imaging, which leads to difficulties in interpretation of results. In order to address the above shortcomings, we propose Eigenanatomy (EANAT), a general framework for sparse matrix factorization. Its goal is to statistically learn the boundaries of and connections between brain regions by weighing both the data and prior neuroanatomical knowledge. Although EANAT incorporates some neuroanatomical prior knowledge in the form of connectedness and smoothness constraints, it can still be difficult for clinicians to interpret the results in specific domains where network-specific hypotheses exist. We thus extend EANAT and present a novel framework for prior-constrained sparse decomposition of matrices derived from brain imaging data, called Prior Based Eigenanatomy (p-Eigen). We formulate our solution in terms of a prior-constrained ℓ1 penalized (sparse) principal component analysis. Experimental evaluation confirms that p-Eigen extracts biologically-relevant, patient-specific functional parcels and that it significantly aids classification of Mild Cognitive Impairment when compared to state-of-the-art competing approaches.

Micro-Columnated Loop Heat Pipe: The Future of Electronic Substrates

Dhillon, Navdeep Singh University of California, Berkeley 2012 해외박사(DDOD)

The modern world is run by semiconductor-based electronic systems. Due to continuous improvements in semiconductor device fabrication, there is a clear trend in the market towards the development of electronic devices and components that not only deliver enhanced computing power, but are also more compact. Thermal management has emerged as the primary challenge in this scenario where heat flux dissipation of electronic chips is increasing exponentially, but conventional cooling solutions such as conduction and convection are no longer feasible. To keep device junction temperatures within the safe operating limit, there is an urgent requirement for ultra-high-conductivity thermal substrates that not only absorb and transport large heat fluxes, but can also provide localized cooling to thermal hotspots. This dissertation describes the design, modeling, and fabrication of a phase change-based, planar, ultra-thin, passive thermal transport system that is inspired by the concept of loop heat pipes and capillary pumped loops. Fabricated on silicon and Pyrex wafers using microfabrication techniques, the micro-columnated loop heat pipe (muCLHP) can be integrated directly with densely packed or multiply-stacked electronic substrates, to provide localized high-heat-flux thermal management. The muCLHP employs a dual-scale coherent porous silicon(CPS)-based micro-columnated wicking structure, where the primary CPS wick provides large capillary forces for fluid transport, while a secondary surface-wick maximizes the rate of thin-film evaporation. To overcome the wick thickness limitation encountered in conventional loop heat pipes, strategies based on MEMS surface micromachining techniques were developed to reduce parasitic heat flow from the evaporator to the compensation chamber of the device. Finite element analysis was used to confirm this reduction in a planar evaporator design, thus enabling the generation of a large motive temperature head for continuous device operation. To predict the overall heat carrying capacity of the muCLHP in the capillary pumping limit, an analytical model was developed to account for a steady state pressure balance in the device flow loop. Based on this model, a design optimization study, employing monotonicity analysis and numerical optimization techniques, was undertaken. It was found that an optimized muCLHP device can absorb heat fluxes as large as 1293 W/cm2 when water is used as a working fluid. A finite volume method-based numerical model was also developed to compute the rates of thin-film evaporation from the patterned surface of the secondary wick. The numerical results indicated that, by properly optimizing the dual-scale wick topology, allowable evaporative heat fluxes can be made commensurate with the heat flux performance predicted by the capillary pumping limit. The latter part of the dissertation deals with the fabrication, packaging, and experimental testing of several in-plane-wicking micro loop heat pipe (muLHP) prototypes. These devices were fabricated on silicon and Pyrex substrates and closely resemble the muCLHP design philosophy, with the exception that the CPS wick is substituted with an easier to fabricate in-plane wick. A novel thermal-flux method was developed for the degassing and fluid charging of the muLHP prototypes. Experiments were conducted to study the process of evaporation and dynamics of the liquid and vapor phases in the device flow loop. Using these results, the overall device and individual component topologies critical to the operation of the two-phase flow loop were identified. A continuous two-phase device flow loop was demonstrated for applied evaporator heat fluxes as high as 41 W/cm2. The performance of these devices, currently found to be limited by the motive temperature head requirement, can be significantly improved by implementing the parasitic heat flow-reduction strategies developed in this work. The 3-D thin-film evaporation model, when integrated into the overall device modeling framework, will enable a design optimization of the micro-columnated wick for further device performance enhancements.

Characterizing the neutralizing antibody repertoires of asymptomatic HIV-1 infected individuals with broadly neutralizing sera

Dhillon, Amandeep K The Scripps Research Institute 2007 해외박사(DDOD)

Although the ability to elicit broadly neutralizing antibodies is considered an important feature of an effective HIV-1 vaccine, elicitation of such antibodies by vaccine antigens remains elusive. To better understand the regions of the HIV-1 envelope spike that are vulnerable to neutralizing antibodies, and the requirements for broad neutralization, the antibody repertoires of three asymptomatic, HIV-1 infected individuals with broadly neutralizing sera were characterized. We demonstrate that neutralization arises from the Protein-A reactive IgG fraction of the sera. Neutralization of a panel of clade B viruses was not inhibited by linear peptides corresponding to the third hypervariable (V3) loop of gp120, suggesting that V3-specific antibodies do not mediate cross-neutralization. The sera also failed to significantly neutralize chimeric SIV and HIV-2 viruses displaying conserved HIV-1 gp41 neutralizing epitopes, indicating that antibodies directed towards these epitopes also do not account for the broad neutralization. Fractionation of polyclonal IgG on recombinant monomeric clade B gp120 demonstrates that gp120-reactive antibodies mediate neutralization of some isolates, but not all. Taken together, the data suggest that neutralization results from multiple neutralizing specificities, although the total number of different specificities is likely small. The neutralization breadth most likely arises from recognition of the CD4 binding site of gp120, although carbohydrate epitopes and conformational epitopes specific to trimeric envelope cannot be excluded. To isolate the monoclonal specificities responsible for neutralization of HIV-1JR-FL, phage-displayed antibody libraries representing the repertoire of one donor were selected against clade B gp120 proteins used in gp120-fractionation experiments. However, only non-neutralizing antibodies specific for the CD4 binding site and V3-loop were recovered. Rescue of the broadly neutralizing monoclonal antibodies may require the development of novel screening antigens and the use of eukaryotic antibody rescue technologies. Lastly, the structure of the V3-specific antibody 447-52D, which exhibits a neutralization breadth uncharacteristic of V3-antibodies, in complex with a V3-peptide is reported. The structure confirms that the neutralization breadth is due to the use of main-chain hydrogen bonds in antigen binding. The similarity of this structure to previously reported V3-peptide structures suggests that structural features of the V3-loop crown are highly conserved. Thus, although V3-specific antibodies do not mediate broad serum neutralization, the V3-region might potentially provide some degree of protection in a multi-component vaccine.