acase@tulane.edu

acase@tulane.edu

This thesis reports the studies of the structure and mechanism of peptide-induced membrane pores by antimicrobial peptide alamethicin and by a peptide named Baxalpha5, which is derived from Bax protein. Alamethicin is one of best known antimicrobial peptides, which are ubiquitous throughout the biological world. Bax-alpha5 peptide is the pore-forming domain of apoptosis regulator protein Bax, which activates pore formation on outer mitochondrial membrane to release cytochrome c to initiate programmed cell death. Both peptides as well as many other pore-forming peptides, induce pores in membrane, however the structure and mechanism of the pore formation were unknown. By utilizing grazing angle x-ray diffraction, I was able to reconstruct the electron density profile of the membrane pores induced by both peptides. The fully hydrated multiple bilayers of peptide-lipid mixture on solid substrate were prepared in the condition that pores were present, as established previously by neutron in-plane scattering and oriented circular dichroism. At dehydrated conditions, the inter bilayer distance of the sample shortened and the interactions between bilayers caused the membrane pores to become long-ranged correlated and formed a periodically ordered lattice of rhombohedral symmetry, so that x-ray diffraction can be applied. To help solving the phase problem of diffraction, a brominated lipid was used and multi-wavelength anomalous diffraction was performed below the bromine K-edge. The reconstructed electron density profiles unambiguously revealed that the alamethicin-induced membrane pore is of barrel-stave type, while the Bax-alpha5 induced pore is of lipidic toroidal (wormhole) type. The underlying mechanism of pore formation was resolved by observing the time-dependent process of pore formation in vesicles exposed to Bax-alpha5 solutions, as well as the membrane thinning experiment. This demonstrated that Bax-alpha5 exhibited the same sigmoidal concentration dependence as alamethicin: below a threshold concentration, peptide only binds to membrane surface, and thins the membrane; when the concentration exceeds a critical value, pores are formed. The structure and mechanism of peptide-induced membrane pores provide insight onto how alpha-pore-forming proteins and peptides interact with membrane. The results also suggest that formation of lipidic pores is the major mechanism for alpha-pore-forming proteins.

In this thesis, properties of particle trajectories associated with the de Broglie-Bohm causal interpretation of quantum mechanics are studied. These trajectories are shown to exhibit deterministic chaos and adiabatic invariance under certain conditions. The very basic elements of the causal interpretation are presented in the first chapter. These include the equations of motion for the particle and the quantum potential. A brief discussion of the philosophically agreeable features of the theory is also included. In Chapter 2, properties of chaotic systems are studied. We define deterministic chaos for a flow and present methods for calculating the maximum Lyapunov exponent. The properties of the different types of systems and the conditions that lead to chaos in these systems are analyzed. We study in detail the specific example of the two-dimensional harmonic oscillator in Chapter 3. We find that different types of trajectories include those which are periodic and chaotic. The necessary conditions for obtaining chaos are determined for a superposition of stationary states. Systems which are qualitatively similar to the harmonic oscillator are covered in Chapter 4. These include the two-dimensional infinite well, an infinite well bisected by a finite barrier, and a Rydberg atom in an external electromagnetic field. In Chapter 5, the effect of a spin 1/2 wavefunction is considered. The causal equations of motion for a spin 1/2 particle are introduced. We find that chaotic trajectories are easily obtained. The causal analogue of the geometric phase is defined in Chapter 6. This phase is shown to be an adiabatic invariant for periodic trajectories. We define the geometric frequency for both periodic and aperiodic trajectories. Finally, in Chapter 7 we examine trajectories associated with stationary states. We define necessary conditions for chaos to arise in the trajectories. The properties of entangled boson and fermion systems are analyzed.

The various devices (transistors, resistors, etc.) in an integrated semiconductor circuit have very highly coupled or correlated parametric inter-relationships. Adding to the complexity, are changes in the parametric values as the sizes and spacings between the devices change. This coupling is not in the form of interaction fields or forces but rather takes place through the correlation of parameters between different devices. These parametric correlations occur because of the processing of the semiconductor wafers through its manufacturing stages. The devices on each wafer have many n-type or p-type doped semiconductor layers in common because of being processed at the same temperature, or in the same gaseous environments, or in the same implantation sessions. In addition, each doped layer has variations over its different regions. All this results in very complex parametric interrelationships between the various devices within the integrated circuit. In turn these have very influential effects on the variation of key circuit characteristics. In spite of the tremendous importance of knowing and predicting these relationships, accurate methods of predicting these complex relationships between devices have evaded the semiconductor industry. The current methods used, such as statistically independent Monte Carlo simulation and Corner Models, either severely underestimate or severely overestimate the variation of key integrated circuit characteristics of interest. Either way, the current methods are very inaccurate. In order to meet this challenge, the methods covered in this dissertation have been developed and applied to the case at hand. They are based on applications of probability, statistics, stochastic, and random field theory, and various computer algorithms. Because of the accuracy, the ease with which device correlations are specified, and the use of computer algorithms, it is expected that the techniques described in this dissertation will be the way that accurate statistical integrated circuit simulations will be done by everyone in the industry. In addition, many of the concepts developed here can be applied to other complex correlated systems not necessarily involving semiconductors.

The behavior of particulate media, in contrast to the behavior of continuous media, is governed by the interaction of particles/grains at contacts. The current approach to model particulate media is through a disc model assembly composed of rigid discs linked by normal and tangential springs at contacts. In this project, a novel approach using a lattice type model is investigated. The mathematical concepts related to the formulation of the lattice model are developed and are numerically implemented by a computer code. Numerical experiments were conducted on a two dimensional assembly of spheres subjected to a vertical load and a simple shear strain. The results of numerical simulation are in general agreement with observations from simple shear tests.

<p> <b>Context:</b> The majority (~85%) of the matter in the universe is composed of dark matter, a mysterious particle that does not interact via the electromagnetic force yet does interact with all other matter via the gravitational force. Many direct detection experiments have been devoted to finding interactions of dark matter with baryonic matter via the weak force. It is still possible that dark matter interacts with itself via a strong scale force and has a self-scattering cross-section of ~0.5 cm²g^-1. In fact such a strong scale scattering force could resolve several outstanding astronomical mysteries: a discrepancy between the cuspy density profiles seen in &Lambda;CDM simulations and the cored density profiles observed in low surface brightness galaxies, dwarf spheroidal galaxies, and galaxy clusters, as well as the discrepancy between the significant number of massive Milky Way dwarf spheroidal halos predicted by &Lambda;CDM and the dearth of observed Milky Way dwarf spheroidal halos. <b>Need:</b> While such observations are in conflict with &Lambda;CDM and suggest that dark matter may self-scatter, each suffers from a baryonic degeneracy, where the observations might be explained by various baryonic processes (e.g., AGN or supernove feedback, stellar winds, etc.) rather than self-interacting dark matter (SIDM). If dark matter lags behind the effectively collisionless galaxies then this is clear evidence that dark matter self-interacts. The expected galaxy-dark matter offset is typically >25 kpc (for cross-sections that would explain the other aforementioned issues with &Lambda;CDM), this is larger than the scales of that are plagued by the baryonic degeneracies. <b>Task:</b> To test whether dark matter self-interacts we have carried out a comprehensive survey of the dissociative merging galaxy cluster DLSCL J0916.2+2951 (also known as the Musket Ball Cluster). This survey includes photometric and spectroscopic observations to quantify the position and velocity of the cluster galaxies, weak gravitational lensing observations to map and weigh the mass (i.e., dark matter which comprises ~85% of the mass) of the cluster, Sunyaev-Zel'dovich effect and X-ray observations to map and quantify the intracluster gas, and finally radio observations to search for associated radio relics, which had they been observed would have helped constrain the properties of the merger. Using this information in conjunction with a Monte Carlo analysis model I quantify the dynamic properties of the merger, necessary to properly interpret constraints on the SIDM cross-section. I compare the locations of the galaxies, dark matter and gas to constrain the SIDM cross-section. This dissertation presents this work. <b>Findings:</b> We find that the Musket Ball is a merger with total mass of 4.8^+3.2_-1.5&times;10¹⁴M_sun. However, the dynamic analysis shows that the Musket Ball is being observed 1.1^+1.3_-0.4 Gyr after first pass through and is much further progressed in its merger process than previously identified dissociative mergers (for example it is 3.4^+3.8_-1.4 times further progressed that the Bullet Cluster). By observing that the dark matter is significantly offset from the gas we are able to place an upper limit on the dark matter cross-section of &sigma;_SIDMm^-1_DM &lt; 8 cm²g^-1. However, we find an that the galaxies appear to be leading the weak lensing (WL) mass distribution by 20.5" (129 kpc at z=0.53) in southern subcluster, which might be expected to occur if dark matter self-interacts. Contrary to this finding though the WL mass centroid appears to be leading the galaxy centroid by 7.4" (47 kpc at z=0.53) in the northern subcluster. <b>Conclusion:</b> The southern offset alone suggests that dark matter self-interacts with ~83% confidence. However, when we account for the observation that the galaxy centroid appears to trail the WL centroid in the north the confidence falls to ~55%. While the SIDM scenario is slightly preferred over the CDM scenario it is not significantly so. <b>Perspectives:</b> The galaxy-dark matter offset measurement is dominated by random errors in each cluster. Thus measuring this offset in other dissociative mergers holds the promise of reducing our uncertainty and enabling us to: 1) state confidently whether dark matter self-interacts via a new dark sector force, or 2) constrain the dark matter cross-section to such a degree that SIDM cannot explain the aforementioned mysteries. To this end we have established the Merging Cluster Collaboration to observe and simulate an ensemble of dissociative merging clusters. We are currently in the process of analyzing six dissociative mergers with existing data, and carrying out multi-wavelength observations of a new sample of 15 radio relic identified dissociative mergers. (Abstract shortened by UMI.)</p>

<p>The role of autonomy in the student experience in a large-enrollment undergraduate introductory physics course was studied from a Self-Determination Theory perspective with two studies. Study I, a correlational study, investigated whether certain aspects of the student experience correlated with how autonomy supportive (vs. controlling) students perceived their instructors to be. An autonomy supportive instructor acknowledges students' perspectives, feelings, and perceptions and provides students with information and opportunities for choice, while minimizing external pressures. It was found that the degree to which students perceived their instructors as autonomy supportive was positively correlated with student interest and enjoyment in learning physics (beta=0.31***) and negatively correlated with student anxiety about taking physics (beta=-0.23**). It was also positively correlated with how autonomous (vs. controlled) students' reasons for studying physics became over the duration of the course (i.e., studying physics more because they wanted to vs. had to; beta=0.24***). This change in autonomous reasons for studying physics was in turn positively correlated with student performance in the course (beta=0.17*). Additionally, the degree to which students perceived their instructors as autonomy supportive was directly correlated with performance for those students entering the course with relatively autonomous reasons for studying physics (beta=0.25**). In summary, students who perceived their instructors as more autonomy supportive tended to have a more favorable experience in the course. If greater autonomy support was in fact the cause of a more favorable student experience, as suggested by Self-determination Theory and experimental studies in other contexts, these results would have implications for instruction and instructor professional development in similar contexts. I discuss these implications. Study II, an experimental study, investigated the effect, on the student experience, of the number of opportunities for choice built into the course format. This was done by comparing two sets of classes. In one set of classes, students spent each class period working through a required series of activities. In the other set of classes, with additional choice, students were free to choose what to work on during nearly half of each class. It was found that the effect of additional choice on student interest and enjoyment in learning physics was significantly different for men vs. women, with a Cohen's d of 0.62 (0.16-1.08; 95% CI). Men became somewhat more interested with additional choice and women became less interested. This gender difference in interest and enjoyment as a result of additional choice could not be accounted for by differences in performance. It was also found that only in classes with additional choice did performance in the course correlate with the degree to which students reasons for studying physics became more autonomous during the quarter (beta=0.30*). I discuss the implications that these effects of additional choice have for instruction and course design in similar contexts.

<p>In this dissertation, we describe observational and theoretical work related to the large-scale clustering of matter in the universe. Such work is crucial in constraining models of the Universe in future surveys and is one of the most powerful probes of the nature of dark energy. In Chapter \ref{magnification}, we present work performed using the Deep Lens Survey (DLS) to measure the growth of structure over cosmic time using weak lensing magnification. This is the first time such a measurement has been performed and represents a significant step forward for this relatively new probe of large-scale structure (LSS) which is known to be complementary to other weak lensing measurements. Later in Chapter \ref{conclusions}, we discuss steps needed for magnification become a competitive, precision probe of cosmology. Chapter \ref{covariance} presents a model for the emulation cosmology dependent error covariances in LSS probes. Estimating these covariances are necessary in order to compare models to the data and require a large amount of computational time to create the simulations required. Tools to reduce the number of simulations required and model the cosmology dependence are needed. We utilize a novel decomposition of LSS error covariances that allows for construction of a emulator that fulfills both of these criteria. In order for future surveys to reach their goals, methods to model measurement error and new probes of LSS complementary to those planned are required. The conclusions of this dissertation in Chapter \ref{conclusions} address the future outlook for this work and research that must be done between now and when the next set of survey data is available. Many systematic errors need to be addressed in magnification before it can be considered a precision cosmology tool. For the error covariances, additional methods to reduce the required number of simulations to estimate the matrices are required. In the Appendix, we present a high level description of an open sourced software package that we developed and implemented over the course of these two projects.

<p> The abstract is not available for copy and paste.</p>