No description available.

Current processor designs use additional transistors to add functionality that improves performance. These features tend to exploit instruction level parallelism. However, a point of diminishing returns has been reached in this effort. Instead, these additional transistors could be used to take advantage of thread level parallelism (TLP). This type of parallelism focuses on hundreds of instructions, rather than single instructions, executing in parallel. Additionally, as transistor sizes shrink, the wires on a chip become thinner. Fabricating a thinner wire means increasing the resistance and thus, the latency of that wire. In fact, in the near future, a signal may not reach a portion of the chip in a single clock cycle. So, in future designs, it will be important to limit the length of the wires on a chip. The SCMP parallel computer is a new architecture that is made up of small processing elements, called nodes, which are connected in a 2-D mesh with nearest neighbor connections. Nodes communicate with one another, via message passing, through a network, which uses dimension order worm-hole routing. To support TLP, each node is capable of supporting multiple threads, which execute in a non-preemptive round robin manner. The wire lengths of this system are limited since a node is only connected to its nearest neighbors. This paper focuses on the System C hardware design of the node that gets replicated across the chip. The result is a node implementation that can be used to create a hardware model of the SCMP parallel computer. / Master of Science

No description available.

No description available.

No description available.

Modern and future many-core systems represent large and complex architectures. The communication fabrics in these large systems play an important role in their performance and power consumption. Current simulation methodologies for evaluating networks-on-chip (NoCs) are not keeping pace with the increased complexity of our systems; architects often want to explore many different design knobs quickly. Methodologies that trade-off some accuracy but maintain important workload trends for faster simulation times are highly beneficial at early stages of architectural exploration. We propose a synthetic traffic generation methodology that captures both application behaviour and cache coherence traffic to rapidly evaluate NoCs. This allows designers to quickly indulge in detailed performance simulations without the cost of long-running full system simulation but still capture a full range of application and coherence behaviour. Our methodology has an average (geometric) error of 10.9% relative to full system simulation, and provides 50x speedup on average over full system simulation.

Many applications for the analysis and processing of biological materials require the enrichment of cell subpopulations. Conventional cell sorting systems are large and expensive with complex equipment that necessitates specialized personnel for operation. Employing microfluidics technology for lab-on-a-chip adaptation of these devices provides several benefits: improved transport control, reduced sample volumes, simplicity of operation, portability, greater accessibility, and reduced cost. The designs of microfluidic cell sorting chips vary widely in literature; evaluation and optimization efforts are rarely reported. This study intends to investigate the primary components of the design to understand the effect of various parameters and to improve the performance of the microfluidic chip. Optimized individual elements are incorporated into a proposed cell sorter chip with the ability to dynamically sort target cells from a non-homogeneous solution using electrical driving forces. Numerical and experimental results are used to evaluate the sample focusing element for controlled cell dispensing, the sorting configuration for target cell collection, and the flow elements for reduced pressure effects and prevention of flow blockages. Compact models are adapted to solve the potential field and flow field in the chip and to predict the focused sample stream width. A commercial CFD package is used to perform 2-D simulations of the potential, velocity, and concentration fields. A fluorescence microscopy visualization system is implemented to conduct experiments on several generations of chip designs. The data from sample focusing experiments, performed with fluorescent dye samples, is analyzed using a Gaussian distribution model proposed in this work. A technique for real-time monitoring of fluorescent microspheres in the microfluidic chip enables the use of dynamic cell sorting to emulate fully autonomous operation. The performance values obtained from these experiments are used to characterize the various design configurations. Sample focusing is shown to depend largely on the relative size of the sheath fluid channel and the sample channel, but is virtually independent of the junction shape. Savings in the applied potential can be achieved by utilizing the size dependency. The focusing performance also provides information for optimizing the widths of the channels relative to the cell size. Successful sorting of desired cells is demonstrated for several designs. Key parameters that affect the sorting performance are discussed; a design employing the use of supplemental fluid streams to direct the particle during collection is chosen due to a high sorting evaluation and a low sensitivity to flow anomalies. The necessary reduction of pressure influences to achieve reliable flow conditions is accomplished by introducing channel constrictions to increase the hydrodynamic resistance. Also, prolonged operation is realized by including particle filters in the proposed design to prevent blockages caused by the accumulation of larger particles. A greater understanding of the behaviour of various components is demonstrated and a design is presented that incorporates the elements with the best performance. The capability of the microfluidic chip is summarized based on experimental results of the tested designs and theoretical cell sorting relationships. Adaptation of this chip to a stand-alone, autonomous device can be accomplished by integrating an optical detection system and further miniaturization of the critical components.

Host-microbe interactions at the gastrointestinal interface have emerged as a key component in the governance of human health and disease. Advances in micro-physiological systems are providing researchers with unprecedented access and insights into this complex relationship. These systems combine the benefits of microengineering, microfluidics, and cell culture in a bid to recreate the environmental conditions prevalent in the human gut. Here we present the human-microbial cross talk (HuMiX) platform, one such system that leverages this multidisciplinary approach to provide a representative in vitro model of the human gastrointestinal interface. HuMiX presents a novel and robust means to study the molecular interactions at the host-microbe interface. We summarize our proof-of-concept results obtained using the platform and highlight its potential to greatly enhance our understanding of host-microbe interactions with a potential to greatly impact the pharmaceutical, food, nutrition, and healthcare industries in the future. A number of key questions and challenges facing these technologies are also discussed. (C) 2017 THE AUTHORS. Published by Elsevier LTD on behalf of the Chinese Academy of Engineering and Higher Education Press Limited Company. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Pour croître et proliférer, les cellules cancéreuses de la prostate activent des voies de signalisation dépendantes des androgènes. L'intervention thérapeutique en première ligne du cancer de la prostate (CaP) s'appuie donc d’abord sur le blocage de l'axe androgènes-récepteur aux androgènes (RA) mais rapidement, les patients développent des tumeurs résistantes (CRPC, Castration Resistant Prostate Cancer).Les récepteurs à activité tyrosine kinase de la famille ErbB semblent jouer un rôle dans cette résistance, en particulier le récepteur ErbB3. En effet, l'inactivation des voies en aval d'ErbB1 et ErbB2, en association avec les anti-androgéniques n'empêche pas la progression vers l'hormono-indépendance, et une accumulation nucléaire d'ErbB3 est observée dans les CRPC en même temps que la voie PI3K-Akt est réactivée.Dans ce contexte, nous avons validé l'expression d'une isoforme nucléaire ErbB380kDa chez les patients et dans des lignées hormono-sensible (LNCaP) et hormono-résistante (PC3). Par ChIP-on-chip, nous avons isolé 353 promoteurs cibles communs aux deux lignées, 245 spécifiques à la lignée LNCaP et 925 à la lignée PC3, et montré qu'ErbB380kDa est un co-régulateur transcriptionnel des gènes étudiés, parmi lesquels GATA2. L'analyse in silico de ces promoteurs révèle des sites de liaison pour les facteurs de transcription GATA2 et MZF1 au niveau des régions liant ErbB380kDa. Un complexe nucléaire GATA2-MZF1-ErbB380kDa est retrouvé dans les cellules LNCaP et PC3.Des travaux récents montrent que GATA2 s'associe au RA pour réguler l'expression de gènes et qu'il pourrait être participer à la dissémination métastatique dans le CaP.Nos résultats suggèrent qu'ErbB380kDa pourrait jouer un rôle régulateur, en amont de GATA2, dans les processus de résistance et l'apparition de métastases. Cette isoforme nucléaire insensible aux traitements actuels apparaît donc comme une cible privilégiée pour le ciblage thérapeutique. / Prostate cancer (PCa) is dependent on androgens and functional androgen-receptor (AR) for growth and proliferation. Androgen-directed therapy is used at the first stages of the disease but cancer cells frequently become resistant (CRPC) by inappropriate reactivation of AR activity. As ErbB receptors are expressed in PCa cells, therapies aiming at inactivate the pathways downstream have been tested in advanced prostate cancers alongside hormone-based therapy. Still, a significant proportion of CRPC treated by ErbB1/2 inhibitors resist to treatment. ErbB3 could be responsible for this failure through both its unexpected nuclear localization and the reactivation of the PI3K-Akt pathway in those advanced tumors.We have described a nuclear ErbB380kDa isoform, expressed in hormone-sensitive (LNCaP) and hormone-resistant (PC3) PCa cell lines that accumulates in the nucleus of tumor cells during cancer progression. ChIP-on-chip experiments led us to characterize 353 target promoters binding ErbB380kDa in both cell lines; 245 promoters specific to LNCaP and 925 specific to PC3 cells, among which the promoter of GATA2. We show that ErbB380kDa functions as a transcriptional co-regulator for the studied genes, potentially through its interaction with transcription factors. In silico analysis revealed binding sites for GATA2 and MZF1 transcription factors on the target promoters, and a complex GATA2-MZF1-ErbB380kDa has been found in LNCaP and PC3 cells. Recent publications have reported a role for GATA2 in the regulation of RA responsive-genes and in metastatic spreading. We propose that ErbB380kDa could act, upstream of GATA2, to induce resistance mechanisms and facilitate cancer progression. Thus, ErbB380kDa emerges as a putative target for the development of new therapies in prostate cancer.