ChipMulti-processor (CMP) architectures have become mainstream for designing processors. With a large number of cores, Network-On-Chip (NOC) provides a scalable communication method for CMPs. NOC must be carefully designed to provide low latencies and high throughput in the resource-constrained environment. To improve the network throughput, we propose the Very Long Packet Window (VLPW) architecture for the NOC router design that tries to close the throughput gap between state-of-the-art on-chip routers and the ideal interconnect fabric. To improve throughput, VLPW optimizes Switch Allocation (SA) efficiency. Existing SA normally applies Round-Robin scheduling to arbitrate among the packets targeting the same output port. However, this simple approach suffers from low arbitration efficiency and incurs low network throughput. Instead of relying solely on simple switch scheduling, the VLPW router design globally schedules all the input packets, resolves the output conflicts and achieves high throughput. With the VLPW architecture, we propose two scheduling schemes: Global Fairness and Global Diversity. Our simulation results show that the VLPW router achieves more than 20% throughput improvement without negative effects on zero-load latency.

Univ., Diss.--Saarbrücken, 2002.

Düsseldorf, Universiẗat, Diss., 2003.

No description available.

Doctor of Philosophy / Genetics Interdepartmental Program - Plant Pathology / Jesse A. Poland / A major challenge for 21st century plant geneticists is to predict plant performance based on genetic information. This is a daunting challenge, especially when there are thousands of genes that control complex traits as well as the extreme variation that results from the environment where plants are grown. Rapid advances in technology are assisting in overcoming the obstacle of connecting the genotype to phenotype. Next generation sequencing has provided a wealth of genomic information resulting in numerous completely sequenced genomes and the ability to quickly genotype thousands of individuals. The ability to pair the dense genotypic data with phenotypic data, the observed plant performance, will culminate in successfully predicting cultivar performance. While genomics has advanced rapidly, phenomics, the science and ability to measure plant phenotypes, has slowly progressed, resulting in an imbalance of genotypic to phenotypic data. The disproportion of high-throughput phenotyping (HTP) data is a bottleneck to many genetic and association mapping studies as well as genomic selection (GS). To alleviate the phenomics bottleneck, an affordable and portable phenotyping platform, Phenocart, was developed and evaluated. The Phenocart was capable of taking multiple types of georeferenced measurements including normalized difference vegetation index and canopy temperature, throughout the growing season. The Phenocart performed as well as existing manual measurements while increasing the amount of data exponentially. The deluge of phenotypic data offered opportunities to evaluate lines at specific time points, as well as combining data throughout the season to assess for genotypic differences. Finally in an effort to predict crop performance, the phenotypic data was used in GS models. The models combined molecular marker data from genotyping-by-sequencing with high-throughput phenotyping for plant phenotypic characterization. Utilizing HTP data, rather than just the often measured yield, increased the accuracy of GS models. Achieving the goal of connecting genotype to phenotype has direct impact on plant breeding by allowing selection of higher yielding crops as well as selecting crops that are adapted to local environments. This will allow for a faster rate of improvement in crops, which is imperative to meet the growing global population demand for plant products.

While tissue-engineering approaches of heart valves have made great strides towards creating functional tissues in vitro, the instruments used, named bioreactors, cannot efficiently integrate multiple stimuli to accurately emulate the physiological microenvironment. To address this, we conceptually designed and built a bioreactor system that applied a range of mechanical tension conditions, modulated matrix stiffness, and introduced biochemical signals in a combinatorial and high-throughput manner. Proof-of-concept experiments on PAVIC-seeded hydrogels were performed to assess the independent and combined effects of tensile strain, matrix stiffness and TGF-β1 on myofibroblast differentiation by measuring α-SMA expression, a marker that indicates a disease-associated phenotype. We found that matrix stiffness and TGF-β1 significantly increased α-SMA levels (p < 0.001), while the effect of mechanical strain was only significant on soft gels (~12 kPa) without TGF-β1. This study therefore demonstrated independent and integrated effects of multiple stimuli in regulating key cellular events in the aortic valve.

This thesis presents new spherical harmonic (SH) approaches for ligand-based high-throughput virtual screening (HTVS). If it is assumed that small drug molecules may be adequately superposed and distinguished by co-locating their centers of mass and by performing rotational correlations of their shapes, then to a good approximation each molecule may be represented very compactly using a two dimensional (2D) SH surface envelope. Of course, this assumes that the true molecular surface is star-like, or single-valued, with respect to radial rays projecting from the selected origin. However, this often holds to a very good approximation for small globular molecules. Even when this is not the case, it is nonetheless reasonable to suppose that similar molecules should give similar radial projections and, therefore, that they should share very similar SH representations. Following this premise, a new program called “SpotLight” was developed. The results obtained with this software show that SH-based global shape matching provides a powerful new way to perform HTVS. SH surface representations are increasingly being applied to a broad range of object recognition and registration tasks, and have also been used to model protein-ligand shape complementarity. Most current shape similarity techniques search for global similarities, and may therefore miss finding active compounds with different overall shapes and sizes but which share similar substructures or surface features. Existing molecular fragment matching algorithms can identify common covalent substructures but they are not well suited for performing scaffold-hopping shape-based database searches. This thesis introduces a novel SH fragment-based shape matching approach that can exploit knowledge of structures of existing protein-ligand complexes to perform virtual screening using as queries SH surface fragments derived from crystallographic ligand binding surfaces.

Chemical separation undoubtedly accounts for a large proportion of process industries' activities. In the past few decades, 10-15% of the world's energy consumed was resulted from separation process. Tremendous efforts have been made in separating the components of large quantities of chemical mixtures into pure or purer forms in most industrial chemists. In addition, industrial development and population growth would lead to a further increase in the global demand for energy in the future. This makes the effective and efficient energy separation process one of the most challenging tasks in engineering. Adsorptive separation using porous materials is widely used in industry today. In order for an adsorptive separation process to be efficient, the essential requirement is a selective adsorbent that possesses high surface area and preferentially adsorbs one component (or class of similar components). Metal-organic frameworks (MOFs) are promising materials for separation purposes as their diversity, due to their building block synthesis from metal clusters and organic linker, gives rise to a wide range of porous structures. Engineering of a separation process is a multi-disciplinary problem that requires a holistic approach. In particular, material selection for industrial applications in the field of MOFs is one of the most significant engineering challenges. The complexity of a screening exercise for adsorptive separations arises from the multitude of existing porous adsorbents including MOFs. There are more than 80,000 structures that have been synthesised so far, as well as the multivariate nature of that performance criteria that need to be considered when selecting or designing an optimal adsorbent for a separation process. However, it is infeasible to assess all the potential materials experimentally to identify the promising structure for a particular application. Recently, molecular simulation and mathematical modelling have seen an ever- growing contribution to the research field of MOFs. The development of these computational tools offers a unique platform for the characterisation, prediction and understanding of MOFs, complementary to experimental techniques. In the first part of this research, Monte Carlo molecular simulation and a number of advanced mathematical methods were used to investigate newly synthesised or not well-known MOFs. These computational techniques allowed not only to characterise materials with their textural properties, but also to predict and understand adsorption performances at the atomic level. Based on the insight gained from the molecular simulation, two computational high-throughput screening approaches were designed and assessed. A multi-scale approach has been proposed and used which combined high-throughput molecular simulation, data mining and advanced visualisation, process system modelling and experimental synthesis and testing. The focus here was on two main applications. On one hand, the challenging CO/N2 separation, which is critical for the petrochemical sector, where two molecules have very similar physical properties. On the other hand, the separation of chiral molecules. For CO/N2 separation, a database of 184 Cu- Cu paddle-wheels MOFs, which contains unsaturated metal centres as strong interaction sites, was extracted from CSD (Cambridge Structural Database) MOF subset for material screening. In the case of chiral separation, an efficient high-throughput approach based on calculation of Henry's constant was developed in this research. Owning to the nature of chirality, this separation of relevance to the pharmaceutical sector is crucially important. A database of 1407 homochiral MOFs was extracted, again, from CSD MOF subset for material screening of enantioselective adsorption. The results obtained in these computational high-throughput approaches allows the screening of interesting, existing structures, and would have a huge impact on making MOFs to be industrially interesting adsorbents as well as guiding the synthesis of these materials. From the many different possibilities, the ultimate interest of this work is in developing an integrated systematic study of the structure-adsorption performance relationship working with a limited library of candidate MOF structures in order to identify promising trends and materials for the specific applications mentioned above. In summary, the overall aim of this research was exploiting different computational techniques, developing novel high-throughput approaches in order to tackle important engineering challenges.

No description available.

The ubiquitin-proteasome system (UPS) controls the stability, activity and localisation of most of the proteome and regulates virtually all cellular processes through modification of proteins with ubiquitin. Ubiquitin conjugation is mediated by a conserved enzymatic cascade composed of E1, E2 and E3 enzymes, which cooperate to activate and transfer ubiquitin to substrate proteins. Dysfunction of the UPS is implicated in many disease states, including cancer, neurodegeneration, immune and cardiovascular disorders. Despite the central role of the UPS in cellular regulation, our understanding of the function, interactions and specificity of proteins that comprise the UPS is still limited. One approach to dissect and to study the UPS is to identify molecular probes, which can be used to specifically interrogate catalytic mechanisms and can be potentially considered as entry points for drug discovery. This work focuses on developing novel high-throughput screening methods for inhibitors of the ubiquitin-conjugating enzymes (E2s), using a unicellular organism Saccharomyces cerevisiae and in vitro technologies. S. cerevisiae is a model organism, commonly used in research as a valuable tool for genetic investigations and other high-throughput studies. In this work, we evaluated the toxicity of exogenously expressed human E2s on yeast cells and discovered that one of the E2s, Ube2U, significantly inhibited yeast growth. This inhibition was dependent on the Ube2U ubiquitin-conjugation activity, as demonstrated with a catalytically inactive Ube2U C89A control, which did not affect yeast growth. The growth defect induced by Ube2U allowed us to develop a screening setup for inhibitors of Ube2U, where the enzyme activity was coupled to cell growth readout. Potential Ube2U inhibitors would be identified as rescuers of the slow growing Ube2U-expressing yeast phenotype. Although screening methods in yeast are relatively straightforward to set up and run, the advantages of this system, namely simplicity of the detection signal and high-throughput, are limited by the fact that yeast is not a recognised large scale screening system in pharmaceutical industry, and that it is difficult to identify the target in a complex pathway such as the UPS. In vitro technologies are needed to provide the necessary structure-activity relationship for chemical optimisation. Therefore, we developed a novel, fluorescence-based, miniaturised assay technology, suitable for biochemical investigations and screening for inhibitors of a wide range of specific ubiquitination reactions within the UPS.