MIT Librarieshttp://hdl.handle.net/1721.1/75812018-08-12T01:23:36Z2018-08-12T01:23:36ZScalable mobility support in future internet architecturesMwangi, Xavier Khttp://hdl.handle.net/1721.1/1173282018-08-10T06:18:36Z2018-01-01T00:00:00ZScalable mobility support in future internet architectures
Mwangi, Xavier K
In this thesis, we present MobileNDN, a scalable design for producer mobility support in the Named Data Network (NDN) architecture. While the initial design of NDN provided support for consumer mobility automatically via its stateful forwarding plane, a solution for producer mobility was left unspecified. The mobility support scheme we propose decouples the tasks of 1) detecting whether data may exist on a mobile producer, and of 2) forwarding interest packets towards the mobile producer. First, we use NDNS to detect zones that may be served by mobile producers. Second, based on insights from MobilityFirst, we introduce a scalable global mapping service to locate mobile producers. The combination of these two components yields a mobility solution that allows NDN's forwarding to operate and scale in the face of mobile consumers and producers.
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.; This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.; Cataloged student-submitted from PDF version of thesis.; Includes bibliographical references (pages 50-51).
2018-01-01T00:00:00ZData-driven strategies for vaccine designKaczorowski, Kevin Jhttp://hdl.handle.net/1721.1/1173272018-08-09T06:19:47Z2018-01-01T00:00:00ZData-driven strategies for vaccine design
Kaczorowski, Kevin J
Vaccination is one of the greatest achievements in immunology and in general medicine, and has virtually eradicated many infectious diseases that plagued humans in the past. Vaccination involves injecting an individual with some version of the pathogen in order to allow the individual to develop a memory immune response that will protect them from future challenge with the same pathogen. Until recently, vaccine development has largely followed empirical paradigms that have proven successful against many diseases. However, many pathogens have now evolved that defy success using the traditional approaches. Rational design of vaccines against such pathogens will likely require interdisciplinary approaches spanning engineering, immunology, and the physical sciences. In this thesis, we combine theoretical approaches with protein sequence and clinical data to address two contemporary problems in vaccinology: 1. Developing an antibody vaccine against HIV, an example of a highly mutable pathogen; and 2. Understanding how the many immune components work collectively to effect a systemic immune response, such as to vaccines. In HIV-infected individuals, antibodies produced by the immune system bind to specific parts of an HIV protein called Envelope (Env). However, the virus evades the immune response due to its high mutability, thus making effective vaccine design a huge challenge. To predict the mutational vulnerabilities of the virus, we developed a model (a fitness landscape) to translate sequence data into knowledge of viral fitness, a measure of the ability of the virus to replicate and thrive. The landscape accounts explicitly for coupling interactions between mutations at different positions within the protein, which often dictate how the virus evades the immune response. We developed new computational approaches that enabled us to tackle the large size and mutational variability of Env, since previous approaches have been unsuccessful in this case. A small fraction of HIV-infected individuals produce a class of antibodies called broadly neutralizing antibodies (bnAbs), which neutralize a diverse number of HIV strains and can thus tolerate many mutations in Env. To investigate the mechanisms underlying breadth of these bnAbs, we combined our landscape with 3D protein structures to gain insight into the spatial distribution of binding interactions between bnAbs and Env. Based on this, we designed an optimal set of immunogens (i.e. Env sequences), with mutations at key residues, that are potentially likely to lead to the elicitation of bnAbs via vaccination. We hope that these antigens will soon be tested in animal models. Even when the right immunogens are included in a vaccine, a potent immune response is not always induced. For example, some individuals do not respond to protective influenza vaccines as desired. The human immune system consists of many different immune cells that coordinate their actions to fight infections and respond to vaccines. The balance between these cell populations is determined by direct interactions and soluble factors such as cytokines, which serve as messengers between cells. A mechanistic understanding of how the various immune components cooperate to bring about the immune response can guide strategies to improve vaccine efficacy. To investigate whether differences in immune response could be explained by variation in immune cell compositions across individuals, we analyzed experimental measurements of various immune cell population frequencies in a cohort of healthy humans. We demonstrated that human immune variation in these parameters is continuous rather than discrete. Furthermore, we showed that key combinations of these immune parameters can be used to predict immune response to diverse stimulations, namely cytokine stimulation and vaccination. Thus, we defined the concept of an individual's "immunotype" as their location within the space of these key combinations of parameters. This result highlights a previously unappreciated connection between immune cell composition and systemic immune responses, and can guide future development of therapies that aim to collectively, rather than independently, manipulate immune cell frequencies.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, February 2018.; Cataloged from PDF version of thesis.; Includes bibliographical references.
2018-01-01T00:00:00ZAlgorithms, analysis and software for the global optimization of two-stage stochastic programsKannan, Rohithttp://hdl.handle.net/1721.1/1173262018-08-09T06:19:45Z2018-01-01T00:00:00ZAlgorithms, analysis and software for the global optimization of two-stage stochastic programs
Kannan, Rohit
Optimization models in the chemical process industries often include uncertain model parameters due to uncertainties in market forces and the environment, use of reduced-order and surrogate process models, and difficulty in measuring parameters accurately. Optimal solutions to formulations that simply ignore uncertainties in the model parameters can be economically worthless or even disastrous in safety-critical applications. Rigorously accounting for uncertainties in optimization models arising out of the process industries is usually computationally prohibitive because of their inherent nonconvex and combinatorial nature. This thesis develops branch-and-bound (B&B) algorithms and a software framework for the scalable solution of a rich class of optimization problems under parametric uncertainty called two-stage stochastic programs, which finds several applications within the petrochemical, pharmaceutical, and energy industries. Additionally, the convergence rates of broad classes of B&B algorithms for constrained optimization problems are analyzed to determine favorable characteristics of such algorithms that can help mitigate the cluster problem in constrained optimization. Two-stage stochastic programming formulations assume that a finite number of scenarios of the uncertain parameters may be realized, and provide a suitable framework for modeling applications with economic objectives. General-purpose B&B algorithms for two-stage stochastic programs suffer from a worst-case exponential increase in solution times with the number of scenarios, which makes the solution of practical applications impractical. This thesis presents a decomposable B&B algorithm for the efficient solution of large-scenario instances of a broad class of two-stage stochastic programs. Furthermore, this thesis details a software framework, named GOSSIP, that was developed for solving such problems. GOSSIP, which implements state-of-the-art decomposition techniques for the global solution of two stage stochastic programs, is shown to perform favorably on a diverse set of test problems from the process systems engineering literature, and is a step towards the efficient solution of two-stage stochastic programming applications from the chemical process industries. Branch-and-bound algorithms that do not possess good convergence properties suffer from the so-called cluster problem wherein a large number of boxes are visited in the vicinity of global optimizers. While the convergence rates of B&B algorithms for unconstrained problems and the cluster problem in unconstrained optimization had been well-studied prior to this thesis, the analyses for constrained problems were lacking, and are the focus of the second part of this thesis. This section of the thesis begins by developing a notion of convergence order for bounding schemes for B&B algorithms, establishes conditions under which first-order and second-order convergent bounding schemes may be sufficient to mitigate the cluster problem, and determines sufficient conditions for widely applicable bounding schemes to possess first-order and second-order convergence. In addition, this section analyzes the convergence orders of some reduced-space B&B algorithms in the literature and establishes that such algorithms may suffer from the cluster problem if domain reduction techniques are not employed. Determining sufficient conditions on the domain reduction techniques to be able to mitigate the above cluster problem can help identify efficient reduced-space B&B algorithms for solving two-stage stochastic programs.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2018.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 315-331).
2018-01-01T00:00:00ZSimulation, sensitivity analysis, and optimization of bioprocesses using dynamic flux balance analysisGomez, Jose Alberto, Ph. D. Massachusetts Institute of Technologyhttp://hdl.handle.net/1721.1/1173252018-08-09T06:19:42Z2018-01-01T00:00:00ZSimulation, sensitivity analysis, and optimization of bioprocesses using dynamic flux balance analysis
Gomez, Jose Alberto, Ph. D. Massachusetts Institute of Technology
Microbial communities are a critical component of natural ecosystems and industrial bioprocesses. In natural ecosystems, these communities can present abrupt and surprising responses to perturbations, which can have important consequences. For example, climate change can influence drastically the composition of microbial communities in the oceans, which in turn affects the entirety of the food chain, and changes in diet can affect drastically the composition of the human gut microbiome, making it stronger or more vulnerable to infection by pathogens. In industrial bioprocesses, engineers work with these communities to obtain desirable products such as biofuels, pharmaceuticals, and alcoholic beverages, or to achieve relevant environmental objectives such as wastewater treatment or carbon capture. Mathematical models of microbial communities are critical for the study of natural ecosystems and for the design and control of bioprocesses. Good mathematical models of microbial communities allow scientists to predict how robust an ecosystem is, how perturbed ecosystems can be remediated, how sensitive an ecosystem is with respect to specific perturbations, and in what ways and how fast it would react to environmental changes. Good mathematical models allow engineers to design better bioprocesses and control them to produce high-quality products that meet tight specifications. Despite the importance of microbial communities, mathematical models describing their behavior remain simplistic and only applicable to very simple and controlled bioprocesses. Therefore, the study of natural ecosystems and the design of complex bioprocesses is very challenging. As a result, the design of bioprocesses remains experiment-based, which is slow, expensive, and labor-intensive. With high throughput experiments large datasets are generated, but without reliable mathematical models critical links between the species in the community are often missed. The design of novel bioprocesses rely on informed guesses by scientists that can only be tested experimentally. The expenses incurred by these experiments can be difficult to justify. Predictive mathematical models of microbial communities can provide insights about the possible outcomes of novel bioprocesses and guide the experimental design, resulting in cheaper and faster bioprocess development. Most mathematical models describing microbial communities do not take into account the internal structure of the microorganisms. In recent years, new knowledge of the internal structures of these microorganisms has been generated using highthroughput DNA sequencing. Flux balance analysis (FBA) is a modeling framework that incorporates this new information into mathematical models of microbial communities. With FBA, growth and exchange flux predictions are made by solving linear programs (LPs) that are constructed based on the metabolic networks of the microorganisms. FBA can be combined with the mathematical models of dynamical biosystems, resulting in dynamic FBA (DFBA) models. DFBA models are difficult to simulate, sensitivity information is challenging to obtain, and reliable strategies to solve optimization problems with DFBA models embedded are lacking. Therefore, the use of DFBA models in science and industry remains very limited. This thesis makes DFBA simulation more accessible to scientists and engineers with DFBAlab, a fast, reliable, and efficient Matlab-based DFBA simulator. This simulator is used by more than a 100 academic users to simulate various processes such as chronic wound biofilms, gas fermentation in bubble column bioreactors, and beta-carotene production in microalgae. Also, novel combinations of microbial communities in raceway ponds have been studied. The performance of algal-yeast cocultures and more complex communities for biolipids production has been evaluated, gaining relevant insights that will soon be tested experimentally. These combinations could enable the production of lipids-rich biomass in locations far away from power plants and other concentrated CO 2 sources by utilizing lignocellulosic waste instead. Following reliable DFBA simulation, the mathematical theory required for sensitivity analysis of DFBA models, which happen to be nonsmooth, was developed. Methods to compute generalized derivative information for special compositions of functions, hierarchical LPs, and DFBA models were generated. Significant numerical challenges appeared during the sensitivity computation of DFBA models, some of which were resolved. Despite the challenges, sensitivity information for DFBA models was used to solve for the steady-state of a high-fidelity model of a bubble column bioreactor using nonsmooth equation-solving algorithms. Finally, local optimization strategies for different classes of problems with DFBA models embedded were generated. The classes of problems considered include parameter estimation and optimal batch, continuous steady-state, and continuous cyclic steady-state process design. These strategies were illustrated using toy metabolic networks as well as genome-scale metabolic networks. These optimization problems demonstrate the superior performance of optimizers when reliable sensitivity information is used, as opposed to approximate information obtained from finite differences. Future work includes the development of global optimization strategies, as well as increasing the robustness of the computation of sensitivities of DFBA models. Nevertheless, the application of DFBA models of microbial communities for the study of natural ecosystems and bioprocess design and control is closer to reality.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2018.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 301-312).
2018-01-01T00:00:00Z