Dealing with Immigration in the Context of EU Enlargement: the Case of Transitional Provisions

This analysis addresses the issue of immigration in the context of the European Union enlargement. Focusing on the use of transitional provisions, it attempts to explain why and when EU leaders give workers from new member countries access to their labor market. Building on the observation that EU leaders seem not to use provisions in the spirit of the law, I gauge the importance of domestic political stakes in the use of those provisions. The empirical results suggest that although EU leaders implement and repeal provisions based on economic circumstances, political factors do intervene in the decision-making process. However, it remains uncertain whether those political factors are institutional or purely electoral.

Engineering Highly-functional, Self-regenerative Skeletal Muscle Tissues with Enhanced Vascularization and Survival in Vivo

Tissue engineering of biomimetic skeletal muscle may lead to development of new therapies for myogenic repair and generation of improved in vitro models for studies of muscle function, regeneration, and disease. For the optimal therapeutic and in vitro results, engineered muscle should recreate the force-generating and regenerative capacities of native muscle, enabled respectively by its two main cellular constituents, the mature myofibers and satellite cells (SCs). Still, after 20 years of research, engineered muscle tissues fall short of mimicking contractile function and self-repair capacity of native skeletal muscle. To overcome this limitation, we set the thesis goals to: 1) generate a highly functional, self-regenerative engineered skeletal muscle and 2) explore mechanisms governing its formation and regeneration in vitro and survival and vascularization in vivo.

By studying myogenic progenitors isolated from neonatal rats, we first discovered advantages of using an adherent cell fraction for engineering of skeletal muscles with robust structure and function and the formation of a SC pool. Specifically, when synergized with dynamic culture conditions, the use of adherent cells yielded muscle constructs capable of replicating the contractile output of native neonatal muscle, generating >40 mN/mm2 of specific force. Moreover, tissue structure and cellular heterogeneity of engineered muscle constructs closely resembled those of native muscle, consisting of aligned, striated myofibers embedded in a matrix of basal lamina proteins and SCs that resided in native-like niches. Importantly, we identified rapid formation of myofibers early during engineered muscle culture as a critical condition leading to SC homing and conversion to a quiescent, non-proliferative state. The SCs retained natural regenerative capacity and activated, proliferated, and differentiated to rebuild damaged myofibers and recover contractile function within 10 days after the muscle was injured by cardiotoxin (CTX). The resulting regenerative response was directly dependent on the abundance of SCs in the engineered muscle that we varied by expanding starting cell population under different levels of basic fibroblast growth factor (bFGF), an inhibitor of myogenic differentiation. Using a dorsal skinfold window chamber model in nude mice, we further demonstrated that within 2 weeks after implantation, initially avascular engineered muscle underwent robust vascularization and perfusion and exhibited improved structure and contractile function beyond what was achievable in vitro.

To enhance translational value of our approach, we transitioned to use of adult rat myogenic cells, but found that despite similar function to that of neonatal constructs, adult-derived muscle lacked regenerative capacity. Using a novel platform for live monitoring of calcium transients during construct culture, we rapidly screened for potential enhancers of regeneration to establish that many known pro-regenerative soluble factors were ineffective in stimulating in vitro engineered muscle recovery from CTX injury. This led us to introduce bone marrow-derived macrophages (BMDMs), an established non-myogenic contributor to muscle repair, to the adult-derived constructs and to demonstrate remarkable recovery of force generation (>80%) and muscle mass (>70%) following CTX injury. Mechanistically, while similar patterns of early SC activation and proliferation upon injury were observed in engineered muscles with and without BMDMs, a significant decrease in injury-induced apoptosis occurred only in the presence of BMDMs. The importance of preventing apoptosis was further demonstrated by showing that application of caspase inhibitor (Q-VD-OPh) yielded myofiber regrowth and functional recovery post-injury. Gene expression analysis suggested muscle-secreted tumor necrosis factor-α (TNFα) as a potential inducer of apoptosis as common for muscle degeneration in diseases and aging in vivo. Finally, we showed that BMDM incorporation in engineered muscle enhanced its growth, angiogenesis, and function following implantation in the dorsal window chambers in nude mice.

In summary, this thesis describes novel strategies to engineer highly contractile and regenerative skeletal muscle tissues starting from neonatal or adult rat myogenic cells. We find that age-dependent differences of myogenic cells distinctly affect the self-repair capacity but not contractile function of engineered muscle. Adult, but not neonatal, myogenic progenitors appear to require co-culture with other cells, such as bone marrow-derived macrophages, to allow robust muscle regeneration in vitro and rapid vascularization in vivo. Regarding the established roles of immune system cells in the repair of various muscle and non-muscle tissues, we expect that our work will stimulate the future applications of immune cells as pro-regenerative or anti-inflammatory constituents of engineered tissue grafts. Furthermore, we expect that rodent studies in this thesis will inspire successful engineering of biomimetic human muscle tissues for use in regenerative therapy and drug discovery applications.

Immunity and Arginine Deprivation in Alzheimer's Disease

The pathogenesis of Alzheimer’s disease (AD) is a critical unsolved question, and while recent studies have demonstrated a strong association between altered brain immune responses and disease progression, the mechanistic cause of neuronal dysfunction and death is unknown. We have previously described the unique CVN-AD mouse model of AD, in which immune-mediated nitric oxide is lowered to mimic human levels, resulting in a mouse model that demonstrates the cardinal features of AD, including amyloid deposition, hyperphosphorylated and aggregated tau, behavioral changes and age-dependent hippocampal neuronal loss. Using this mouse model, we studied longitudinal changes in brain immunity in relation to neuronal loss and, contrary to the predominant view that AD pathology is driven by pro-inflammatory factors, we find that the pathology in CVN-AD mice is driven by local immune suppression. Areas of hippocampal neuronal death are associated with the presence of immunosuppressive CD11c+ microglia and extracellular arginase, resulting in arginine catabolism and reduced levels of total brain arginine. Pharmacologic disruption of the arginine utilization pathway by an inhibitor of arginase and ornithine decarboxylase protected the mice from AD-like pathology and significantly decreased CD11c expression. Our findings strongly implicate local immune-mediated amino acid catabolism as a novel and potentially critical mechanism mediating the age-dependent and regional loss of neurons in humans with AD.

There is a large interest in identifying, lineage tracing, and determining the physiologic roles of monophagocytes in Alzheimer’s disease. While Cx3cr1 knock-in fluorescent reporting and Cre expressing mice have been critical for studying neuroimmunology, mice that are homozygous null or hemizygous for CX3CR1 have perturbed neural development and immune responses. There is, therefore, a need for similar tools in which mice are CX3CR1+/+. Here, we describe a mouse where Cre is driven by the Cx3cr1 promoter on a bacterial artificial chromosome (BAC) transgene (Cx3cr1-CreBT) and the Cx3cr1 locus is unperturbed. Similarly to Cx3cr1-Cre knock-in mice, these mice express Cre in Ly6C-, but not Ly6C+, monocytes and tissue macrophages, including microglia. These mice represent a novel tool that maintains the Cx3cr1 locus while allowing for selective gene targeting in monocytes and tissue macrophages.

The study of immunity in Alzheimer’s requires the ability to identify and quantify specific immune cell subsets by flow cytometry. While it is possible to identify lymphocyte subsets based on cell lineage-specific markers, the lack of such markers in brain myeloid cell subsets has prevented the study of monocytes, macrophages and dendritic cells. By improving on tissue homogenization, we present a comprehensive protocol for flow cytometric analysis, that allows for the identification of several cell types that have not been previously identified by flow cytometry. These cell types include F4/80hi macrophages, which may be meningeal macrophages, IA/IE+ macrophages, which may represent perivascular macrophages, and dendritic cells. The identification of these cell types now allows for their study by flow cytometry in homeostasis and disease.

Bayesian Inference in Large-scale Problems

Many modern applications fall into the category of "large-scale" statistical problems, in which both the number of observations n and the number of features or parameters p may be large. Many existing methods focus on point estimation, despite the continued relevance of uncertainty quantification in the sciences, where the number of parameters to estimate often exceeds the sample size, despite huge increases in the value of n typically seen in many fields. Thus, the tendency in some areas of industry to dispense with traditional statistical analysis on the basis that "n=all" is of little relevance outside of certain narrow applications. The main result of the Big Data revolution in most fields has instead been to make computation much harder without reducing the importance of uncertainty quantification. Bayesian methods excel at uncertainty quantification, but often scale poorly relative to alternatives. This conflict between the statistical advantages of Bayesian procedures and their substantial computational disadvantages is perhaps the greatest challenge facing modern Bayesian statistics, and is the primary motivation for the work presented here.

Two general strategies for scaling Bayesian inference are considered. The first is the development of methods that lend themselves to faster computation, and the second is design and characterization of computational algorithms that scale better in n or p. In the first instance, the focus is on joint inference outside of the standard problem of multivariate continuous data that has been a major focus of previous theoretical work in this area. In the second area, we pursue strategies for improving the speed of Markov chain Monte Carlo algorithms, and characterizing their performance in large-scale settings. Throughout, the focus is on rigorous theoretical evaluation combined with empirical demonstrations of performance and concordance with the theory.

One topic we consider is modeling the joint distribution of multivariate categorical data, often summarized in a contingency table. Contingency table analysis routinely relies on log-linear models, with latent structure analysis providing a common alternative. Latent structure models lead to a reduced rank tensor factorization of the probability mass function for multivariate categorical data, while log-linear models achieve dimensionality reduction through sparsity. Little is known about the relationship between these notions of dimensionality reduction in the two paradigms. In Chapter 2, we derive several results relating the support of a log-linear model to nonnegative ranks of the associated probability tensor. Motivated by these findings, we propose a new collapsed Tucker class of tensor decompositions, which bridge existing PARAFAC and Tucker decompositions, providing a more flexible framework for parsimoniously characterizing multivariate categorical data. Taking a Bayesian approach to inference, we illustrate empirical advantages of the new decompositions.

Latent class models for the joint distribution of multivariate categorical, such as the PARAFAC decomposition, data play an important role in the analysis of population structure. In this context, the number of latent classes is interpreted as the number of genetically distinct subpopulations of an organism, an important factor in the analysis of evolutionary processes and conservation status. Existing methods focus on point estimates of the number of subpopulations, and lack robust uncertainty quantification. Moreover, whether the number of latent classes in these models is even an identified parameter is an open question. In Chapter 3, we show that when the model is properly specified, the correct number of subpopulations can be recovered almost surely. We then propose an alternative method for estimating the number of latent subpopulations that provides good quantification of uncertainty, and provide a simple procedure for verifying that the proposed method is consistent for the number of subpopulations. The performance of the model in estimating the number of subpopulations and other common population structure inference problems is assessed in simulations and a real data application.

In contingency table analysis, sparse data is frequently encountered for even modest numbers of variables, resulting in non-existence of maximum likelihood estimates. A common solution is to obtain regularized estimates of the parameters of a log-linear model. Bayesian methods provide a coherent approach to regularization, but are often computationally intensive. Conjugate priors ease computational demands, but the conjugate Diaconis--Ylvisaker priors for the parameters of log-linear models do not give rise to closed form credible regions, complicating posterior inference. In Chapter 4 we derive the optimal Gaussian approximation to the posterior for log-linear models with Diaconis--Ylvisaker priors, and provide convergence rate and finite-sample bounds for the Kullback-Leibler divergence between the exact posterior and the optimal Gaussian approximation. We demonstrate empirically in simulations and a real data application that the approximation is highly accurate, even in relatively small samples. The proposed approximation provides a computationally scalable and principled approach to regularized estimation and approximate Bayesian inference for log-linear models.

Another challenging and somewhat non-standard joint modeling problem is inference on tail dependence in stochastic processes. In applications where extreme dependence is of interest, data are almost always time-indexed. Existing methods for inference and modeling in this setting often cluster extreme events or choose window sizes with the goal of preserving temporal information. In Chapter 5, we propose an alternative paradigm for inference on tail dependence in stochastic processes with arbitrary temporal dependence structure in the extremes, based on the idea that the information on strength of tail dependence and the temporal structure in this dependence are both encoded in waiting times between exceedances of high thresholds. We construct a class of time-indexed stochastic processes with tail dependence obtained by endowing the support points in de Haan's spectral representation of max-stable processes with velocities and lifetimes. We extend Smith's model to these max-stable velocity processes and obtain the distribution of waiting times between extreme events at multiple locations. Motivated by this result, a new definition of tail dependence is proposed that is a function of the distribution of waiting times between threshold exceedances, and an inferential framework is constructed for estimating the strength of extremal dependence and quantifying uncertainty in this paradigm. The method is applied to climatological, financial, and electrophysiology data.

The remainder of this thesis focuses on posterior computation by Markov chain Monte Carlo. The Markov Chain Monte Carlo method is the dominant paradigm for posterior computation in Bayesian analysis. It has long been common to control computation time by making approximations to the Markov transition kernel. Comparatively little attention has been paid to convergence and estimation error in these approximating Markov Chains. In Chapter 6, we propose a framework for assessing when to use approximations in MCMC algorithms, and how much error in the transition kernel should be tolerated to obtain optimal estimation performance with respect to a specified loss function and computational budget. The results require only ergodicity of the exact kernel and control of the kernel approximation accuracy. The theoretical framework is applied to approximations based on random subsets of data, low-rank approximations of Gaussian processes, and a novel approximating Markov chain for discrete mixture models.

Data augmentation Gibbs samplers are arguably the most popular class of algorithm for approximately sampling from the posterior distribution for the parameters of generalized linear models. The truncated Normal and Polya-Gamma data augmentation samplers are standard examples for probit and logit links, respectively. Motivated by an important problem in quantitative advertising, in Chapter 7 we consider the application of these algorithms to modeling rare events. We show that when the sample size is large but the observed number of successes is small, these data augmentation samplers mix very slowly, with a spectral gap that converges to zero at a rate at least proportional to the reciprocal of the square root of the sample size up to a log factor. In simulation studies, moderate sample sizes result in high autocorrelations and small effective sample sizes. Similar empirical results are observed for related data augmentation samplers for multinomial logit and probit models. When applied to a real quantitative advertising dataset, the data augmentation samplers mix very poorly. Conversely, Hamiltonian Monte Carlo and a type of independence chain Metropolis algorithm show good mixing on the same dataset.