• Semantic Prioritization of Novel Causative Genomic Variants in Mendelian and Oligogenic Diseases

      Boudellioua, Imane (2019-03-21) [Dissertation]
      Advisor: Hoehndorf, Robert
      Committee members: Gao, Xin; Arold, Stefan T.; Rebholz-Schuhmann, Dietrich
      Recent advances in Next Generation Sequencing (NGS) technologies have facilitated the generation of massive amounts of genomic data which in turn is bringing the promise that personalized medicine will soon become widely available. As a result, there is an increasing pressure to develop computational tools to analyze and interpret genomic data. In this dissertation, we present a systematic approach for interrogating patients’ genomes to identify candidate causal genomic variants of Mendelian and oligogenic diseases. To achieve that, we leverage the use of biomedical data available from extensive biological experiments along with machine learning techniques to build predictive models that rival the currently adopted approaches in the field. We integrate a collection of features representing molecular information about the genomic variants and information derived from biological networks. Furthermore, we incorporate genotype-phenotype relations by exploiting semantic technologies and automated reasoning inferred throughout a cross-species phenotypic ontology network obtained from human, mouse, and zebra fish studies. In our first developed method, named PhenomeNet Variant Predictor (PVP), we perform an extensive evaluation of a large set of synthetic exomes and genomes of diverse Mendelian diseases and phenotypes. Moreover, we evaluate PVP on a set of real patients’ exomes suffering from congenital hypothyroidism. We show that PVP successfully outperforms state-of-the-art methods, and provides a promising tool for accurate variant prioritization for Mendelian diseases. Next, we update the PVP method using a deep neural network architecture as a backbone for learning and illustrate the enhanced performance of the new method, DeepPVP on synthetic exomes and genomes. Furthermore, we propose OligoPVP, an extension of DeepPVP that prioritizes candidate oligogenic combinations in personal exomes and genomes by integrating knowledge from protein-protein interaction networks and we evaluate the performance of OligoPVP on synthetic genomes created by known disease-causing digenic combinations. Finally, we discuss some limitations and future steps for extending the applicability of our proposed methods to identify the genetic underpinning for Mendelian and oligogenic diseases.
    • Engineering of Pseudocapacitive Materials and Device Architecture for On-Chip Energy Storage

      jiang, qiu (2019-03-05) [Dissertation]
      Advisor: Alshareef, Husam N.
      Committee members: Zhang, Xixiang; Salama, Khaled N.; Fan, Hongjin
      The emergence of micropower-type applications such as self-powered sensors and miniaturized electronic systems has increased interest in on-chip electrochemical energy storage such as microsupercapacitors. Microsupercapacitors (MSCs) are high rate and high power yet miniaturized versions of macroscopic supercapacitors. MSCs with planar configuration have higher power density at potentially comparable energy density to thin-film batteries, while possessing essentially infinite cycle life. They could also offer compatible integration with smart electronic devices on an integrated chip (IC). In this dissertation, state-of-the-art microsupercapacitors based on Ti3C2Tx MXene and other pseudocapacitive electrode materials are proposed. The proposed strategies involve engineering both intrinsic properties of materials, fabrication methods and device architecture.
    • Spectral Density Function Estimation with Applications in Clustering and Classification

      Chen, Tianbo (2019-03-03) [Dissertation]
      Advisor: Sun, Ying
      Committee members: Ombao, Hernando; Al-Naffouri, Tareq Y.; Senturk, Damla
      Spectral density function (SDF) plays a critical role in spatio-temporal data analysis, where the data are analyzed in the frequency domain. Although many methods have been proposed for SDF estimation, real-world applications in many research fields, such as neuroscience and environmental science, call for better methodologies. In this thesis, we focus on the spectral density functions for time series and spatial data, develop new estimation algorithms, and use the estimators as features for clustering and classification purposes. The first topic is motivated by clustering electroencephalogram (EEG) data in the spectral domain. To identify synchronized brain regions that share similar oscillations and waveforms, we develop two robust clustering methods based on the functional data ranking of the estimated SDFs. The two proposed clustering methods use different dissimilarity measures and their performance is examined by simulation studies in which two types of contaminations are included to show the robustness. We apply the methods to two sets of resting-state EEG data collected from a male college student. Then, we propose an efficient collective estimation algorithm for a group of SDFs. We use two sets of basis functions to represent the SDFs for dimension reduction, and then, the scores (the coefficients of the basis) estimated by maximizing the penalized Whittle likelihood are used for clustering the SDFs in a much lower dimension. For spatial data, an additional penalty is applied to the likelihood to encourage the spatial homogeneity of the clusters. The proposed methods are applied to cluster the EEG data and the soil moisture data. Finally, we propose a parametric estimation method for the quantile spectrum. We approximate the quantile spectrum by the ordinary spectral density of an AR process at each quantile level. The AR coefficients are estimated by solving Yule- Walker equations using the Levinson algorithm. Numerical results from simulation studies show that the proposed method outperforms other conventional smoothing techniques. We build a convolutional neural network (CNN) to classify the estimated quantile spectra of the earthquake data in Oklahoma and achieve a 99.25% accuracy on testing sets, which is 1.25% higher than using ordinary periodograms.
    • Understanding Human Activities at Large Scale

      Caba Heilbron, Fabian David (2019-03) [Dissertation]
      Advisor: Ghanem, Bernard
      Committee members: Heidrich, Wolfgang; Wonka, Peter; Snoek, Cees
      With the growth of online media, surveillance and mobile cameras, the amount and size of video databases are increasing at an incredible pace. For example, YouTube reported that over 400 hours of video are uploaded every minute to their servers. Arguably, people are the most important and interesting subjects of such videos. The computer vision community has embraced this observation to validate the crucial role that human action recognition plays in building smarter surveillance systems, semantically aware video indexes and more natural human-computer interfaces. However, despite the explosion of video data, the ability to automatically recognize and understand human activities is still somewhat limited. In this work, I address four different challenges at scaling up action understanding. First, I tackle existing dataset limitations by using a flexible framework that allows continuous acquisition, crowdsourced annotation, and segmentation of online videos, thus, culminating in a large-scale, rich, and easy-to-use activity dataset, known as ActivityNet. Second, I develop an action proposal model that takes a video and directly generates temporal segments that are likely to contain human actions. The model has two appealing properties: (a) it retrieves temporal locations of activities with high recall, and (b) it produces these proposals quickly. Thirdly, I introduce a model, which exploits action-object and action-scene relationships to improve the localization quality of a fast generic action proposal method and to prune out irrelevant activities in a cascade fashion quickly. These two features lead to an efficient and accurate cascade pipeline for temporal activity localization. Lastly, I introduce a novel active learning framework for temporal localization that aims to mitigate the data dependency issue of contemporary action detectors. By creating a large-scale video benchmark, designing efficient action scanning methods, enriching approaches with high-level semantics for activity localization, and an effective strategy to build action detectors with limited data, this thesis is making a step closer towards general video understanding.
    • A functional group approach for predicting fuel properties

      Abdul Jameel, Abdul Gani (2019-03) [Dissertation]
      Advisor: Sarathy, Mani
      Committee members: Roberts, William L.; Szekely, Gyorgy; Won, Sang Hee
      Experimental measurement of fuel properties are expensive, require sophisticated instrumentation and are time consuming. Mathematical models and approaches for predicting fuel properties can help reduce time and costs. A new approach for characterizing petroleum fuels called the functional group approach was developed by disassembling the innumerable fuel molecules into a finite number of molecular fragments or ‘functional groups’. This thesis proposes and tests the following hypothesis, Can a fuels functional groups be used to predict its combustion properties? Analytical techniques like NMR spectroscopy that are ideally suited to identify and quantify the various functional groups present in the fuels was used. Branching index (BI), a new parameter that quantifies the degree and quality of branching in a molecule was defined. The proposed hypothesis was tested on three classes of fuels namely gasolines, diesel and heavy fuel oil. Five key functional groups namely paraffinic CH3, paraffinic CH2, paraffinic CH, naphthenic CH-CH2 and aromatic C-CH groups along with BI were used as matching targets to formulate simple surrogates of one or two molecules that reproduce the combustion characteristics. Using this approach, termed as the minimalist functional group (MFG) approach surrogates were formulated for a number of standard gasoline, diesel and jet fuels. The surrogates were experimentally validated using measurements from Ignition quality tester (IQT), Rapid compression machine (RCM) and smoke point (SP) apparatus. The functional group approach was also employed to predict research octane number (RON) and motor octane number (MON) of fuels blended with ethanol using artificial neural networks (ANN). A multiple linear regression (MLR) based model for predicting derived cetane number (DCN) of hydrocarbon fuels was also developed. The functional group approach was also extended to study heavy fuel oil (HFO), a viscous residual fuel that contains heteroatoms like S, N and O. It is used in ships as marine fuel and also in boilers for electricity generation. 1H NMR and 13C NMR measurements were made to analyze the average molecular parameters (AMP) of HFO molecules. The fuel was divided into 19 different functional groups and their concentrations were calculated from the AMP values. A surrogate molecule that represents the average structure of HFO was then formulated and its properties were predicted using QSPR approaches.
    • Algorithms and Frameworks for Graph Analytics at Scale

      Jamour, Fuad (2019-02-28) [Dissertation]
      Advisor: Kalnis, Panos
      Committee members: Keyes, David E.; Hadwiger, Markus; Amer-Yahia, Sihem
      Graph queries typically involve retrieving entities with certain properties and connectivity patterns. One popular property is betweenness centrality, which is a quantitative measure of importance used in many applications such as identifying influential users in social networks. Solving graph queries that involve retrieving important entities with user-defined connectivity patterns in large graphs requires efficient com- putation of betweenness centrality and efficient graph query engines. The first part of this thesis studies the betweenness centrality problem, while the second part presents a framework for building efficient graph query engines. Computing betweenness centrality entails computing all-pairs shortest paths; thus, exact computation is costly. The performance of existing approximation algorithms is not well understood due to the lack of an established benchmark. Since graphs in many applications are inherently evolving, several incremental algorithms were proposed. However, they cannot scale to large graphs: they either require excessive memory or perform unnecessary computations rendering them prohibitively slow. Existing graph query engines rely on exhaustive indices for accelerating query evaluation. The time and memory required to build these indices can be prohibitively high for large graphs. This thesis attempts to solve the aforementioned limitations in the graph analytics literature as follows. First, we present a benchmark for evaluating betweenness centrality approximation algorithms. Our benchmark includes ground-truth data for large graphs in addition to a systematic evaluation methodology. This benchmark is the first attempt to standardize evaluating betweenness centrality approximation algorithms and it is currently being used by several research groups working on approximate between- ness in large graphs. Then, we present a linear-space parallel incremental algorithm for updating betweenness centrality in large evolving graphs. Our algorithm uses biconnected components decomposition to localize processing graph updates, and it performs incremental computation even within affected components. Our algorithm is up to an order of magnitude faster than the state-of-the-art parallel incremental algorithm. Finally, we present a framework for building low memory footprint graph query engines. Our framework avoids building exhaustive indices and uses highly optimized matrix algebra operations instead. Our framework loads datasets, and evaluates data-intensive queries up to an order of magnitude faster than existing engines.
    • Efficient Numerical Methods for High-Dimensional Approximation Problems

      Wolfers, Sören (2019-02-06) [Dissertation]
      Advisor: Tempone, Raul
      Committee members: Keyes, David E.; Mai, Paul Martin; Gobet, Emmanuel
      In the field of uncertainty quantification, the effects of parameter uncertainties on scientific simulations may be studied by integrating or approximating a quantity of interest as a function over the parameter space. If this is done numerically, using regular grids with a fixed resolution, the required computational work increases exponentially with respect to the number of uncertain parameters – a phenomenon known as the curse of dimensionality. We study two methods that can help break this curse: discrete least squares polynomial approximation and kernel-based approximation. For the former, we adaptively determine sparse polynomial bases and use evaluations in random, quasi-optimally distributed evaluation nodes; for the latter, we use evaluations in sparse grids, as introduced by Smolyak. To mitigate the additional cost of solving differential equations at each evaluation node, we extend multilevel methods to the approximation of response surfaces. For this purpose, we provide a general analysis that exhibits multilevel algorithms as special cases of an abstract version of Smolyak’s algorithm. In financial mathematics, high-dimensional approximation problems occur in the pricing of derivatives with multiple underlying assets. The value function of American options can theoretically be determined backwards in time using the dynamic programming principle. Numerical implementations, however, face the curse of dimensionality because each asset corresponds to a dimension in the domain of the value function. Lack of regularity of the value function at the optimal exercise boundary further increases the computational complexity. As an alternative, we propose a novel method that determines an optimal exercise strategy as the solution of a stochastic optimization problem and subsequently computes the option value by simple Monte Carlo simulation. For this purpose, we represent the American option price as the supremum of the expected payoff over a set of randomized exercise strategies. Unlike the corresponding classical representation over subsets of Euclidean space, this relax- ation gives rise to a well-behaved objective function that can be globally optimized using standard optimization routines.
    • Design, fabrication and application of fractional-order capacitors

      Agambayev, Agamyrat (2019-02) [Dissertation]
      Advisor: Bagci, Hakan
      Committee members: Salama, Khaled N.; Baran, Derya; Biswas,Karabi
      The fractional–order capacitors add an additional degree of freedom over conventional capacitors in circuit design and facilitate circuit configurations that would be impractical or impossible to implement with conventional capacitors. We propose a generic strategy for fractional-order capacitor fabrication that integrates layers of conductive, semiconductor and ferroelectric polymer materials to create a composite with significantly improved constant phase angle, constant phase zone, and phase angle variation performance. Our approach involves a combination of dissolving the polymer powders, mixing distinct phases and making a film and capacitor of it. The resulting stack consisting of ferroelectric polymer-based composites shows constant phase angle over a broad range of frequencies. To prove the viability of this method, we have successfully fabricated fractional-order capacitors with the following: nanoparticles such as multiwall carbon nanotube (MWCNT), Molybdenum sulfide (MoS2) inserted ferroelectric polymers and PVDF based ferroelectric polymer blends. They show better performance in terms of fabrication cost and dynamic range of constant phase angle compared to fractional order capacitor from graphene percolated polymer composites. These results can be explained by a universal percolation model, where the combination of electron transport in fillers and the dielectric relaxation time distribution of the permanent dipoles of ferroelectric polymers increase the constant phase angle level and constant phase zone of fractional-order capacitors. This approach opens up a new avenue in fabricating fractional capacitors involving a variety of heterostructures combining the different fillers and different matrixes.
    • Highly efficient photoleletrochemical water splitting by optical, electrical and catalysis concurrent management

      Fu, Hui-Chun (2019-02) [Dissertation]
      Advisor: He, Jr-Hau
      Committee members: Ooi, Boon S.; Huang, Kuo-Wei; Zhang, Wenjun
      One way of harnessing and storing our most abundant and renewable energy source, sunlight, is by utilizing it to split water for the hydrogen generation as a storable form of fuel. Si, the most investigated material for solar-to-hydrogen technology has great potential as the single photoelectrode. While some success has been achieved in Si-Based photoelectrochemical (PEC) systems, they suffer from low efficiency and short longevity. Moreover, in order for hydrogen to be commercially viable, the existing challenges of electrical, optical, and catalysis management must be addressed concurrently. Herein, we work on the simultaneous improvement in light harvesting, charge carrier separation/transfer, and catalysis management of Si-based photocathodes, achieving best-in-class efficiency with stable electrochemical performance. By decoupling the light harvesting side from the electrocatalytic surface we nullify parasitic light absorption. We developed a Si bifacial (SiBF) PEC photocathode to absorb light on both sides of PEC devices, which exhibits a current density of 39.01 mA/cm2. Unlike conventional monofacial PEC cells, our bifacial design demonstrates excellent omnidirectional light harvesting capability. Furthermore, back buried junction photoelectrochemical (BBJ-PEC) cells were fabricated that can realize efficient decoupling of photon. This scheme enables maximum light-harvesting without any metal contact, which prevents the shadow effect during the water splitting reaction. The highest hydrogen evolution current density (41.76 mA/cm2) was demonstrated based on a single BBJ-PEC device. Additionally, wireless water splitting can be achieved when three BBJ-PEC cells were connected in series. The efficient PEC cell design described herein demonstrates promising performance, taking us a step closer to real-world solar-to-hydrogen production.
    • Ultrafast Spectroscopy of Polymer: Non-fullerene Small Molecule Acceptor Bulk Heterojunction Organic Solar Cells

      Alamoudi, Maha A (2019-01-07) [Dissertation]
      Advisor: Laquai, Frédéric
      Committee members: McCulloch, Iain; Anthopoulos, Thomas D.; Andrienko, Denis; Cabanetos, Clement
      Organic photovoltaics has emerged as a promising technology for electricity generation. The essential component in an organic solar cell is the bulk heterojunction absorber layer, typically a blend of an electron donor and an electron acceptor. Efforts have been made to design new materials such as donor polymers and novel acceptors to improve the power conversion efficiencies. New fullerene free acceptors providing low cost synthesis routes and tenability of their optoelectronic and electrochemical properties have been designed. Despite the efforts, still not much is known about the photopysical processes in these blends that limit the performance. In this respect, time-resolved spectroscopy such as transient absorption and time-resolved photoluminescence, can provide in-depth insight into the various (photo) physical processes in bulk heterojunction solar cell. In this thesis, PCE10 was used as donor and paired with different non fullerene acceptors. In the first part of this thesis the impact of the core structure (cyclopenta-[2, 1-b:3, 4-b’]dithiophene (CDT) versus indacenodithiophene (IDTT)) of malononitrile (BM)-terminated acceptors, abbreviated as CDTBM and IDTTBM, on the photophysical characteristics of BHJ solar cells is reported. The IDTT-based acceptor achieves power conversion efficiencies of 8.4%, higher than the CDT-based acceptor (5.6%), due to concurrent increase in short-circuit current and open-circuit voltage. Using (ultra)fast transient spectroscopy we demonstrate that reduced geminate recombination in PCE10: IDTTBM blends is the reason for the difference in short-circuit currents. External quantum efficiency measurements indicate that the higher energy of interfacial charge-transfer states observed for the IDTT-based acceptor blends is the origin of the higher open-circuit voltage. In the second part of this thesis, I report the impact of acceptor side chains on the photo-physical processes of BHJ solar cells using three different IDT-based acceptors, namely O-IDTBR, EH-IDTBR and O-IDTBCN blended with PCE10. Power conversion efficiencies as high as 10 % were achieved. The transient absorption spectroscopy experiments provide insight into sub-picosecond exciton dissociation and charge generation which is followed by nanosecond triplet state formation in PCE10:O-DTBR and PCE10:EH-IDTBR blends, while in O-IDTBCN triplets are not observed. Time delayed collection field experiments (TDCF) were performed to address the charge carrier generation and examine its dependence on the electric field.
    • Development of bismuth (oxy)sulfide-based materials for photocatalytic applications

      BaQais, Amal (2019-01-07) [Dissertation]
      Advisor: Takanabe, Kazuhiro
      Committee members: Cavallo, Luigi; Wang, Peng; Abdi, Fatwa F.
      Technologies based on alternative and sustainable energy sources present a vital solution in the present and for the future. These technologies are strongly driven by the increased global energy demand and need to reduce environmental issues created by fossil fuel. Solar energy is an abundant, clean and free-access resource, but it requires harvesting and storage for a sustainable future. Direct conversion and storage of solar energy using heterogeneous photocatalysts have been identified as parts of a promising paradigm for generating green fuels from sunlight and water. This thesis focused on developing semiconductor absorbers in a visible light region for photocatalytic hydrogen production reaction. In addition, theoretical studies are combined with experimental results for a deep understanding of the intrinsic optoelectronic properties of the obtained materials. The study presents a novel family of oxysulfide BiAgOS, produced by applying a full substitution strategy of Cu by Ag in BiCuOS. I was interested to address how the total substitution of Cu by Ag in a BiCuOS system affects its crystal structure, optical and electronic properties using experimental characterizations and theoretical calculations. Single-phase bismuth silver oxysulfide BiAgOS was prepared via a hydrothermal method. Rietveld refinement of the powder confirmed that BiAgOS is an isostructural BiCuOS. The diffraction peak positions of BiAgOS, relative to those of BiCuOS, were shifted toward lower angles, indicating an increase in the cell parameters. BiCuOS and BiAgOS were found to have indirect bandgaps of 1.1 and 1.5 eV, respectively. The difference in the bandgap results from the difference in the valence band compositions. The hybrid level of the S and Ag orbitals in BiAgOS is located at a more positive potential than that of S and Cu, leading to a widened bandgap. Both materials possess high dielectric constants and low electron and hole effective masses, making them interesting for photoconversion applications. BiAgOS has a potential for photocatalytic hydrogen evolution reaction in the presence of sacrificial reagents; however, it is inactive toward water oxidation. BiCuOS and BiAgOS can be considered interesting starting compositions for the development of new semiconductors for PV or Z-scheme photocatalytic applications. The second study investigates the synthesis and characterization of NaBiS2, this contains Bi3+, which belongs to the p-block electronic configuration Bi3+ 6s26p0, and NaLaS2, which contains La3+ with electronic configuration 6s05d0. Solid-state reactions from oxide precursor starting materials were applied for synthesis the materials. The sulfurization process was conducted by pressurizing a saturated vapor of CS2. The obtained black material of NaBiS2 has an indirect transition with high absorption coefficients in the visible region of the spectrum and the absorption edge is determined at 1.21 eV. However, NaBiS2 did not show photocatalytic activity toward hydrogen production. NaLaS2 is characterized by an indirect transition with a bandgap in the UV region at 3.15 eV and can drive the photocatalytic hydrogen evolution reaction in Na2S/Na2SO3 solution. Utilizing the solid solution NaLa1-xBixS2 strategy, the absorption properties and band edge position for photocatalytic hydrogen evolution reaction were optimized. The results indicated that the bismuth content is critical parameter for maintaining the photocatalytic activity. The incorporation of low Bi content up to 6% in NaLaS2 leads to extending the photon absorption from the UV to the visible region and enhancing the photocatalytic activity of hydrogen production. In contrast, all the solid solutions that have Bi content of more than 12% present absorption edges close to that of pure NaBiS2, and they are inactive for photocatalytic hydrogen production. Combining the experimental measurements with density functional theory calculations, such behavior can be explained by the degree of overlapping of Bi and La states on the conduction band minimum (CBM). Finally, self-assembly of Bi2S3 nanorods were grown on FG or FTO substrates. Bi2S3 thin films were prepared by sulfurization of Bi metal layer using the hydrothermal method. The results show that Bi2S3 has absorption up to 1.3 eV and has a moderate absorption coefficient in the visible region. The ultraviolet photoelectron spectroscopy and photoelectron spectroscopy in air results showed that the conduction band minimum of Bi2S3 is located slightly above the hydrogen redox potential. However, Pt/Bi2S3 did not evolve a detectable amount of hydrogen, suggesting the presence of surface states that can hinder the hydrogen reduction reaction.
    • Engineering of novel Biocatalysts with Functionalities beyond Nature

      Gespers (Akal), Anastassja (2019-01) [Dissertation]
      Advisor: Rueping, Magnus
      Committee members: Arold, Stefan T.; Hamdan, Samir; Stingl, Ulrich
      Novel biocatalysts are highly demanded in the white biotechnology. Hence, the development of highly stable and enantioselective biocatalysts with novel functionalities is an ongoing research topic. Here, an osmium ligating single-site ArM was created based on the biotinstreptavidin technology for the dihydroxylation of olefins. For the creation of the artificial catalytic metal center in the streptavidin (SAV) cavity, efficient osmium tetroxide (OsO4) chelating biotin-ligands were created. The unspecific metal binding of the host scaffold was diminished through genetical and chemical modification of the host protein. The created single-site OsO4 chelating ArM was successfully applied in the asymmetric cyclopropanation, revealing a stable and tunable catalytic hybrid system for application. The structural analysis of protein-ligand complexes is essential for the advanced rational design and engineering of artificial metalloenzymes. In previous studies, a SAV-dirhodium ArM was created and successfully applied in the asymmetric cyclopropanation reaction. To improve the selectivity of the SAV-dirhodium complex, the structural location of the organometallic complex in the SAV cavity was targeted and small-angle x-ray scattering (SAXS) was used to obtain the structural information. The SAXS analysis revealed valuable information of the molecular state of the complexes; hence, the method proved to be useful for the structural analysis of protein-ligand interactions. The discovery of novel enzymes from nature is still the major source for improved biocatalysts. One of the most important enzymes used in the molecular biology are DNA polymerases in PCR reactions. The halothermophilic brine-pool 3 polymerase (BR3 Pol) from the Atlantis II Red Sea brine pool showed optimal activities at 55 °C and salt concentrations up to 0.5 M NaCl, and was stable at temperatures above 95 °C. The comparison with the hyperthermophilic KOD polymerase revealed the haloadaptation of BR3 Pol due to an increased negative electrostatic surface charge and an overall higher structural flexibility. Engineered chimeric KOD polymerases with swapped single BR3 Pol domains revealed increased salt tolerance in the PCR, showing increased structural flexibility and a local negative surface charge. The understanding of the BR3 Pol haloadaptation might enable the development of a DNA polymerase tailored for specific PCR reactions with increased salt concentrations.
    • Communication Reducing Approaches and Shared-Memory Optimizations for the Hierarchical Fast Multipole Method on Distributed and Many-core Systems

      Abduljabbar, Mustafa (2018-12-06) [Dissertation]
      Advisor: Keyes, David E.
      Committee members: Bagci, Hakan; Hadwiger, Markus; Gropp, William D.
      We present algorithms and implementations that overcome obstacles in the migration of the Fast Multipole Method (FMM), one of the most important algorithms in computational science and engineering, to exascale computing. Emerging architectural approaches to exascale computing are all characterized by data movement rates that are slow relative to the demand of aggregate floating point capability, resulting in performance that is bandwidth limited. Practical parallel applications of FMM are impeded in their scaling by irregularity of domains and dominance of collective tree communication, which is known not to scale well. We introduce novel ideas that improve partitioning of the N-body problem with boundary distribution through a sampling-based mechanism that hybridizes two well-known partitioning techniques, Hashed Octree (HOT) and Orthogonal Recursive Bisection (ORB). To reduce communication cost, we employ two methodologies. First, we directly utilize features available in parallel runtime systems to enable asynchronous computing and overlap it with communication. Second, we present Hierarchical Sparse Data Exchange (HSDX), a new all-to-all algorithm that inherently relieves communication by relaying sparse data in a few steps of neighbor exchanges. HSDX exhibits superior scalability and improves relative performance compared to the default MPI alltoall and other relevant literature implementations. We test this algorithm alongside others on a Cray XC40 tightly coupled with the Aries network and on Intel Many Integrated Core Architecture (MIC) represented by Intel Knights Corner (KNC) and Intel Knights Landing (KNL) as modern shared-memory CPU environments. Tests include comparisons of thoroughly tuned handwritten versus auto-vectorization of FMM Particle-to-Particle (P2P) and Multipole-to-Local (M2L) kernels. Scalability of task-based parallelism is assessed with FMM’s tree traversal kernel using different threading libraries. The MIC tests show large performance gains after adopting the prescribed techniques, which are inevitable in a world that is moving towards many-core parallelism.
    • Spatio-Temporal Data Analysis by Transformed Gaussian Processes

      Yan, Yuan (2018-12-06) [Dissertation]
      Advisor: Genton, Marc G.
      Committee members: Alouini, Mohamed-Slim; Sun, Ying; Morgenthaler, Stephan
      In the analysis of spatio-temporal data, statistical inference based on the Gaussian assumption is ubiquitous due to its many attractive properties. However, data collected from different fields of science rarely meet the assumption of Gaussianity. One option is to apply a monotonic transformation to the data such that the transformed data have a distribution that is close to Gaussian. In this thesis, we focus on a flexible two-parameter family of transformations, the Tukey g-and-h (TGH) transformation. This family has the desirable properties that the two parameters g ∈ R and h ≥ 0 involved control skewness and tail-heaviness of the distribution, respectively. Applying the TGH transformation to a standard normal distribution results in the univariate TGH distribution. Extensions to the multivariate case and to a spatial process were developed recently. In this thesis, motivated by the need to exploit wind as renewable energy, we tackle the challenges of modeling big spatio-temporal data that are non-Gaussian by applying the TGH transformation to different types of Gaussian processes: spatial (random field), temporal (time series), spatio-temporal, and their multivariate extensions. We explore various aspects of spatio-temporal data modeling techniques using transformed Gaussian processes with the TGH transformation. First, we use the TGH transformation to generate non-Gaussian spatial data with the Matérn covariance function, and study the effect of non-Gaussianity on Gaussian likelihood inference for the parameters in the Matérn covariance via a sophisticatedly designed simulation study. Second, we build two autoregressive time series models using the TGH transformation. One model is applied to a dataset of observational wind speeds and shows advantaged in accurate forecasting; the other model is used to fit wind speed data from a climate model on gridded locations covering Saudi Arabia and to Gaussianize the data for each location. Third, we develop a parsimonious spatio-temporal model for time series data on a spatial grid and utilize the aforementioned Gaussianized climate model wind speed data to fit the latent Gaussian spatio-temporal process. Finally, we discuss issues under a unified framework of modeling multivariate trans-Gaussian processes and adopt one of the TGH autoregressive models to build a stochastic generator for global wind speed.
    • Molybdenum Disulfide as an Efficient Catalyst for Hydrogen Evolution Reaction

      Alarawi, Abeer A. (2018-12-02) [Dissertation]
      Advisor: He, Jr-Hau
      Committee members: Di Fabrizio, Enzo M.; Han, Yu; Jin, Song
      Hydrogen is a carrier energy gas that can be utilized as a clean energy source instead of oil and natural gas. Splitting the water into hydrogen and oxygen is one of the most favorable methods to generate hydrogen. The catalytic properties of molybdenum disulfide (MoS2) could be valuable in this role, particularly due to its unique structure and ability to be chemically modified, enabling its catalytic activity to be further enhanced or made comparable to that of Pt-based materials. In general, these modification strategies may involve either structural engineering of MoS2 or enhancing the kinetics of charge transfer, including by confining to single metal atoms and clusters or integrating with a conductive substrate. We present the results of synergetic integration of MoS2 films with a Si-heterojunction solar cell for generating H2 via the photochemical water splitting approach. The results of the photochemical measurements demonstrated an efficient photocurrent of 36. 3 mA cm-2 at 0 V vs. RHE and an onset potential of 0.56 V vs. RHE. In addition to 25 hours of continuous photon conversion to H2 generation, this study points out that the integration of the Si-HJ with MoS2 is an effective strategy for enhancing the internal conductivity of MoS2 towards efficient and stable hydrogen production. Moreover, we studied the effect of doping an atomic scale of Pt on the catalytic activity of MoS2. The electrochemical results indicated that the optimum single Pt atoms loading amount demonstrated a distinct enhancement in the hydrogen generating, in which the overpotential was minimized to -0.0505 V to reach a current density of 10 mA cm−2 using only 10 ALD cycles of Pt. The Tafel slopes of the ALD Pt/ML-MoS2 electrodes were in the range of 55–120 mV/decade, which indicates a fast improvement in the HER velocity as a result of the increased potential. Stability is another important parameter for evaluating a catalyst. The same (10 ALD cycles) Pt/ML-MoS2 electrode was able to continuously generate hydrogen molecules at for 150 hours. These superior results demonstrate that the low conductivity of semiconductive MoS2 can be enhanced by anchoring the film with Pt SAs and clusters, leading to sufficient charge transport and a decrease in the overpotential.
    • Antibiotic resistance genes and antibiotic resistant bacteria as emerging contaminants in wastewater: fate and persistence in engineered and natural environments

      Mantilla Calderon, David (2018-12) [Dissertation]
      Advisor: Hong, Pei-Ying
      Committee members: Plewa, Michael J.; Daffonchio, Daniele; Saikaly, Pascal
      The emergence and rapid spread of antimicrobial resistance (AMR) is a phenomenon that extends beyond clinical settings. AMR has been detected in multiple environmental compartments, including agricultural soils and water bodies impacted by wastewater discharges. The purpose of this research project was to evaluate what factors could influence the environmental persistence of antibiotic resistance genes (ARGs), as well as to identify potential strategies employed by human pathogens to survive in secondary environment outside the host. The first part of this dissertation describes the incidence of the New Delhi metallobeta lactamase gene (blaNDM-1) – an ARG conferring resistance to last resort antibiotics – in the influent of a wastewater treatment facility processing municipal wastewater from Jeddah, Saudi Arabia. Detection of blaNDM-1 was followed by the isolation of a multi-drug resistant strain of E. coli (denoted as strain PI7) at a frequency of ca. 3 x 104 CFU/m3 in the untreated municipal wastewater. Subsequently, we described the decay kinetics of E. coli PI7 in microcosm experiments simulating biological treatment units of wastewater treatment plants. We identified that transition to dormancy is the main strategy prolonging the persistence of E. coli PI7 in the microcosm experiments. Additionally, we observed slower decay of E. coli PI7 and prolonged stability of extracellular DNA in anoxic/anaerobic conditions. In the last chapter of this thesis, the fate of extracellular DNA is further explored. Using as a model Acinetobacter baylyi ADP1, we describe the stimulation of natural transformation frequencies in the presence of chlorination disinfection byproducts (DBPs). Moreover, we demonstrate the ability of BAA to stimulate transformation is associated with its capacity to cause DNA damage via oxidative stress. Overall, this dissertation addresses important knowledge gaps in our current understanding of ARB and extracellular ARG persistence in the environment. The results from this project highlight the importance of retrofitting the existing water treatment process with advance membrane filtration units, and the need to relook into the current disinfection strategies. Wastewater treatment technologies should be assessed for their efficacies in not only inactivating ARB and ARGs, but also whether unintended consequences such as stimulated horizontal gene transfer would occur.
    • Unravelling the Metabolic Interactions of the Aiptasia-Symbiodiniaceae Symbiosis

      Cui, Guoxin (2018-12) [Dissertation]
      Advisor: Aranda, Manuel
      Committee members: Gojobori, Takashi; Voolstra, Christian R.; Pringle, John R.; Weis, Virginia M.
      Many omics-level studies have been undertaken on Aiptasia, however, our understanding of the genes and processes associated with symbiosis regulation and maintenance is still limited. To gain deeper insights into the molecular processes underlying this association, we investigated this relationship using multipronged approaches combining next generation sequencing with metabolomics and immunohistochemistry. We identified 731 high-confident symbiosis-associated genes using meta-analysis. Coupled with metabolomic profiling, we exposed that symbiont-derived carbon enables host recycling of ammonium into nonessential amino acids, which may serve as a regulatory mechanism to control symbiont growth through a carbon-dependent negative feedback of nitrogen availability to the symbiont. We then characterized two symbiosis-associated ammonium transporters (AMTs). Both of the proteins exhibit gastrodermis-specific localization in symbiotic anemones. Their tissuespecific localization consistent with the higher ammonium assimilation rate in gastrodermis of symbiotic Aiptasia as shown by 15N labeling and nanoscale secondary ion mass spectrometry (NanoSIMS). Inspired by the tissue-specific localization of AMTs, we investigated spatial expression of genes in Aiptasia. Our results suggested that symbiosis with Symbiodiniaceae is the main driver for transcriptional changes in Aiptasia. We focused on the phagosome-associated genes and identified several key factors involved in phagocytosis and the formation of symbiosome. Our study provided the first insights into the tissue specific complexity of gene expression in Aiptasia. To investigate symbiosis-induced response in symbiont and to find further evidence for the hypotheses generated from our host-focused analyses, we explored the growth and gene expression changes of Symbiodiniaceae in response to the limitations of three essential nutrients: nitrogen, phosphate, and iron, respectively. Comparisons of the expression patterns of in hospite Symbiodiniaceae to these nutrient limiting conditions showed a strong and significant correlation of gene expression profiles to the nitrogen-limited culture condition. This confirmed the nitrogen-limited growing condition of Symbiodiniaceae in hospite, and further supported our hypothesis that the host limits the availability of nitrogen, possibly to regulate symbiont cell density. In summary, we investigated different molecular aspects of symbiosis from both the host’s and symbiont’s perspective. This dissertation provides novel insights into the function of nitrogen, and the potential underlying molecular mechanisms, in the metabolic interactions between Aiptasia and Symbiodiniaceae.
    • Molecular Basis for p85 Dimerization and Allosteric Ligand Recognition

      Aljedani, Safia (2018-12) [Dissertation]
      Advisor: Arold, Stefan T.
      Committee members: Bernado, Pau; Hamdan, Samir; Al-Babili, Salim
      The phosphatidylinositol-3-kinase α (PI3Kα) is a heterodimeric enzyme that is composed of a p85α regulatory subunit and a p110α catalytic subunit. PI3Kα plays a critical role in cell survival, growth and differentiation, and is the most frequently mutated pathway in human cancers. The PI3Kα pathway is also targeted by many viruses, such as the human immunodeficiency virus (HIV-1), the herpes simplex virus 1 (HSV-1) or the influenza A virus, to create favourable conditions for viral replication. The regulatory p85α stabilizes the catalytic p110α, but keeps it in an inhibited state. Various ligands can bind to p85α and allosterically activate p110α, but the mechanisms are still ill-defined. Intriguingly, p85α also binds to, and activates, the PTEN phosphatase, which is the antagonist of p110α. Previous studies indicated that only p85α monomers bind to the catalytic p110α subunit, whereas only p85α dimers bind to PTEN. These findings suggest that the balance of p85α monomers and dimers regulates the PI3Kα pathway, and that interrupting this equilibrium could lead to disease development. However, the molecular mechanism for p85α dimerization is controversial, and it is unknown why PTEN only binds to p85α dimers, whereas p110α only binds to p85α monomers. Here we set out to elucidate these questions, and to gain further understanding of how p85α ligands influence p85α dimerization and promote activation of p110α. We first established a comprehensive library of p85α fragments and protocols for their production and purification. By combining biophysical and structural methods such as small angle X-ray scattering, X-ray crystallography, nuclear magnetic resonance, microscale thermophoresis, and chemical crosslinking, we investigated the contributions of all p85α domains to dimerization and ligand binding. Contrarily to the prevailing thought in the field, we find that p85α dimerization and ligand recognition involves multiple domains, including those that directly bind to and inhibit p110α. This finding allows us to suggest a molecular mechanism that links p85α dimerization and allosteric p110α activation through ligands.
    • Using single molecule fluorescence to study substrate recognition by a structure-specific 5’ nuclease

      Rashid, Fahad (2018-12) [Dissertation]
      Advisor: Hamdan, Samir
      Committee members: Habuchi, Satoshi; Di Fabrizio, Enzo M.; Laporo, Joseph
      Nucleases are integral to all DNA processing pathways. The exact nature of substrate recognition and enzymatic specificity in structure-specific nucleases that are involved in DNA replication, repair and recombination has been under intensive debate. The nucleases that rely on the contours of their substrates, such as 5’ nucleases, hold a distinctive place in this debate. How this seemingly blind recognition takes place with immense discrimination is a thought-provoking question. Pertinent to this question is the observation that even minor variations in the substrate provoke extreme catalytic variance. Increasing structural evidence from 5’ nucleases and other structure-specific nuclease families suggest a common theme of substrate recognition involving distortion of the substrate to orient it for catalysis and protein ordering to assemble active sites. Using three single-molecule (sm)FRET approaches of temporal resolution from milliseconds to sub-milliseconds, along with various supporting techniques, I decoded a highly sophisticated mechanism that show how DNA bending and protein ordering control the catalytic selectivity in the prototypic system human Flap Endonuclease 1 (FEN1). Our results are consistent with a mutual induced-fit mechanism, with the protein bending the DNA and the DNA inducing a protein-conformational change, as opposed to functional or conformational selection mechanism. Furthermore, we show that FEN1 incision on the cognate substrate occurs with high efficiency and without missed opportunity. However, when FEN1 encounters substrates that vary in their physical attributes to the cognate substrate, cleavage happens after multiple trials During the course of my work on FEN1, I found a novel photophysical phenomena of protein-induced fluorescence quenching (PIFQ) of cyanine dyes, which is the opposite phenomenon of the well-known protein-induced fluorescence enhancement (PIFE). Our observation and characterization of PIFQ led us to further investigate the general mechanism of fluorescence modulation and how the initial fluorescence state of the DNA-dye complex plays a fundamental role in setting up the stage for the subsequent modulation by protein binding. Within this paradigm, we propose that enhancement and quenching of fluorescence upon protein binding are simply two different faces of the same process. Our observations and correlations eliminate the current inconvenient arbitrary nature of fluorescence modulation experimental design.
    • Cylindrical Magnetic Nanowires Towards Three Dimensional Data Storage

      Mohammed, Hanan (2018-12) [Dissertation]
      Advisor: Kosel, Jürgen
      Committee members: Manchon, Aurelien; Fariborzi, Hossein; Nielsch, Kornelius
      The past few decades have witnessed a race towards developing smaller, faster, cheaper and ultra high capacity data storage technologies. In particular, this race has been accelerated due to the emergence of the internet, consumer electronics, big data, cloud based storage and computing technologies. The enormous increase in data is paving the path to a data capacity gap wherein more data than can be stored is generated and existing storage technologies would be unable to bridge this data gap. A novel approach could be to shift away from current two dimensional architectures and onto three dimensional architectures wherein data can be stored vertically aligned on a substrate, thereby decreasing the device footprint. This thesis explores a data storage concept based on vertically aligned cylindrical magnetic nanowires which are promising candidates due to their low fabrication cost, lack of moving parts as well as predicted high operational speed. In the proposed concept, data is stored in magnetic nanowires in the form of magnetic domains or bits which can be moved along the nanowire to write/read heads situated at the bottom/top of the nanowire using spin polarized current. Cylindrical nanowires generally exhibit a single magnetic domain state i.e. a single bit, thus for these cylindrical nanowire to exhibit high density data storage, it is crucial to pack multiple domains within a nanowire. This dissertation demonstrates that by introducing compositional variation i.e. multiple segments along the nanowire, using materials with differing values of magnetization such as cobalt and nickel, it is possible to incorporate multiple domains in a nanowire. Since the fabrication of cylindrical nanowires is a batch process, examining the properties of a single nanowire is a challenging task. This dissertation deals with the fabrication, characterization and manipulation of magnetic domains in individual nanowires. The various properties of are investigated using electrical measurements, magnetic microscopy techniques and micromagnetic simulations. In addition to packing multiple domains in a cylindrical nanowire, this dissertation reports the current assisted motion of domain walls along multisegmented Co/Ni nanowires, which is a fundamental step towards achieving a high density cylindrical nanowire-based data storage device.