Planejamento probabilístico lida com a tomada de decisão sequencial em ambientes estocásticos e geralmente é modelado por um Processo de Decisão Markoviano (Markovian Decision Process - MDP). Um MDP modela a interação entre um agente e o seu ambiente: em cada estágio, o agente decide executar uma ação, com efeitos probabilísticos e um certo custo, que irá produzir um estado futuro. O objetivo do agente MDP é minimizar o custo esperado ao longo de uma sequência de escolhas de ação. O número de estágios que o agente atua no ambiente é chamado de horizonte, o qual pode ser finito, infinito ou indefinido. Um exemplo de MDP com horizonte indefinido é o Stochastic Shortest Path MDP (SSP MDP), que estende a definição de MDP adicionando um conjunto de estados meta (o agente para de agir ao alcançar um estado meta). Num SSP MDP é feita a suposição de que é sempre possível alcançar um estado meta a partir de qualquer estado do mundo. No entanto, essa é uma suposição muito forte e que não pode ser garantida em aplicações práticas. Estados a partir dos quais é impossível atingir a meta são chamados de becos-sem-saída. Um beco-sem-saída pode ser evitável ou inevitável (se nenhuma política leva do estado inicial para a meta com probabilidade um). Em trabalhos recentes foram propostas extensões para SSP MDP que permitem a existência de diferentes tipos de beco-sem-saída, bem como algoritmos para resolvê-los. No entanto, a detecção de becos-sem-saída é feita utilizando: (i) heurísticas que podem falhar para becos-sem-saída implícitos ou (ii) métodos mais confiáveis, mas que demandam alto custo computacional. Neste projeto fazemos uma caracterização formal de modelos de planejamento probabilístico com becos-sem-saída. Além disso, propomos uma nova técnica para detecção de becos-sem-saída baseada nessa caracterização e adaptamos algoritmos de planejamento probabilístico para utilizarem esse novo método de detecção. Os resultados empíricos mostram que o método proposto é capaz de detectar todos os becos-sem-saída de um dado conjunto de estados e, quando usado com planejadores probabilísticos, pode tornar esses planejadores mais eficientes em domínios com becos-sem-saída difíceis de serem detectados / Probabilistic planning deals with sequential decision making in stochastic environments and is modeled by a Markovian Decision Process (MDP). An MDP models the interaction between an agent and its environment: at each stage, the agent decides to execute an action, with probabilistic effects and a certain cost which produces a future state. The purpose of the MDP agent is to minimize the expected cost along a sequence of choices. The number of stages that the agent acts in the environment is called horizon, which can be finite, infinite or undefined. An example of MDP with undefined horizon is the Stochastic Shortest Path MDP, which extends the definition of MDP by adding a set of goal states (the agent stops acting after reaching a goal state). In an SSP MDP the assumption is made that it is always possible to achieve a goal state from every state of the world. However, this is a very strong assumption and cannot be guaranteed in practical applications. States from which it is impossible to reach the goal are called dead-ends. A dead-end may be avoidable or unavoidable (when no policy leads from the initial state to the goal with probability one). Recent work has proposed extensions to SSP MDP that allow the existence of different types of dead-ends as well as algorithms to solve them. However, the detection of dead-end is done using: (i) heuristics that may fail to detect implicitly dead-ends or (ii) more reliable methods that require a high computational cost. In this project we make a formal characterization of probabilistic planning models with dead-ends. In addition, we propose a new technique for dead-end detection based on this characterization and we adapt probabilistic planning algorithms to use this new detection method. The empirical results show that the proposed method is able to detect all dead-ends of a given set of states and, when used withprobabilistic planners, can make these planners more efficient in domains with difficult to detect dead-ends.

Poly(ADP-ribosylation) is a post-translational covalent modification of proteins and has been involved in processes such as DNA repair and gene expression. The poly(ADP-ribose) polymer (PAR) is mainly catalyzed by a family of enzymes termed poly(ADP-ribose) polymerases (PARPs). TcPARP of Trypanosoma cruzi, appears to play a role in DNA repair mechanism during Chagas’ disease and believed to be involved in controlling different phases of cell growth. In this study, cloning and characterization of mutants of TcPARP is reported. In first phase, five mutants of TcPARP, already cloned in pET-22b+, were expressed BL21 Rosetta2(DE3). Since expression in this vector was poor, all mutants were cloned in pNH-TrxT vector and expressed in BL21 Rosetta2(DE3). Protein purification of mutant TCP-c006 (124 residues deleted from N-terminus) was performed using His6 tagged FF crude IMAC column followed by gel filtration. Fluorescence-based activity assay of mutant TCP-c006 shows only 35% conversion of NAD+ to nictonamide compared to wild type. Moreover 50 times more protein is required for this conversion when compared with wild type. During DNA binding assay (EMSA), mutant TCP-c006 did not bind with DNA, clearly indicating that N-terminus is necessary for DNA binding and activity of TcPARP. In future, solubility of both wild type and mutants of TcPARP in different expression host system using various fusions such as MBP, T4 lysozyme etc could be tested. Structure of both wild type and mutants of TcPARP could be determined by using Static light scattering (SLS).

The purity of heterologous recombinant proteins is of utmost importance to the pharmaceutical sector since most are consumed as therapeutic agents by humans. Variability caused by co- and posttranslational modifications is a major concern in pharmaceutical production. In order to develop strategies which guarantee a homogeneous product in a robust production process, it is important to better understand the metabolic basis of the synthesis of related non-canonical amino acids. So far, studies have identified high glucose fluxes in connection to oxygen limitation and overexpression of leucine-rich proteins as possible reasons for the production of non-canonical amino acids and their incorporation into heterologous proteins expressed in Escherichia coli. The results presented in this work provide evidence that oscillations in the concentrations of glucose and oxygen as they occur in inhomogeneous industrial scale bioreactors potentiate the synthesis and incorporation of norvaline into the leucine-rich protein IL-2, heterologously expressed in E. coli W3110M, as observed in well-mixed homogenous cultures and perturbed shake flask cultivations. In order to represent the heterogeneities existing in large-scale bioreactors, two experimental setups were applied, using a simple shake flask scale-down model developed to monitor dissolved oxygen and pH online during a batch and fed-batch cultivation phases. Results here show that by applying repeated glucose pulses to the glucose limited culture, which consequently induce oscillations in dissolved oxygen, norvaline is accumulated. Analysis of inclusion bodies that resulted from the expressed IL-2 revealed the presence of norvaline in the protein. A higher concentration of norvaline was observed in the oscillating scale-down model compared to the non-perturbed culture, which suggests that the conditions as they typically occur in large scale bioreactors may be critical for product quality. The results and tools, developed in this work are a solid basis for future cell engineering approaches to overcome the challenges in view of product quality.

Cotransin is a cyclodepsipeptide inhibitor of protein translocation and has been demonstrated to inhibit cotranslational translocation of a variety of proteins by targeting the mammalian ER translocation channel Sec61 (Besemer, Harant et al. 2005, Garrison, Kunkel et al. 2005). Genetic screens in cancer cells found out the mutations near the luminal plug domain of Sec61 confer resistance to cotransin inhibition and thus outline the proposed binding site for cotransin (MacKinnon, Paavilainen et al. 2014). However molecular details of cotransin interactions remain unknown. Purpose of my first project was to express the human Sec61 translocation channel in correct stoichiometric ratios. To our knowledge, heterotrimeric expression of Sec61 has not been achieved previously. Baculovirus system was chosen express the Sec61 heterotrimeric complex. To vary expression levels of Sec61α and Sec61γ relative to Sec61β, separate baculovirus constructs were prepared. Proper co-transfection ratios between these viruses to express Sec61subunits in correct stoichiometric ratios were calculated during my pro gradu and insect cell expression was then scaled up using the determined virus ratios. Results from Sec61 expression have returned sufficient quantities of the translocation channel for biochemical analyses. Final expression seems to contain high lipid/protein ratio, which may have been caused due to insect cells not-fully adapted to the media. The expression should be repeated in well-adapted healthy insect cells and then accessed for protein quality. For its isolation for crystallization studies, co-immunoprecipitation may be a preferred way to pull down the Sec61 complex. Detergent based solubilization may be alternatively used to isolate Sec61. The second part of my pro gradu work included expression of SecYEβ translocation channel from Pyrococcus furiosus, along with its various humanized mutants, whose DNA constructs had been provided by Dr. Ville Paavilainen. A major goal of this work was to express the mutants which bind to cotransin. Via photo-crosslinking and click chemistry analyses, a mutant binding to cotransin was identified and scaled-up. Our photo-crosslinking studies were able to demonstrate that the wild-type SecYEβ does not bind to cotransin in vitro. Results from photo-crosslinking assays during this project also demonstrated that other known translocation inhibitors Mycolactone and Apratoxin A can bind to mammalian Sec61 channel. These results are consistent with unpublished work from Paavilainen lab (Paatero et al).

Bacterial proteins DsbB, VKOR, and the mammalian protein VKORC1 share similar functions involving electron transfer processes. While DsbB is not homologous to the bacterial and mammalian VKOR proteins, the three proteins share overall structural features. Based on the similarities between the three proteins and the finding that mtbVKOR can replace DsbB in E. coli, we considered the possibility of rat VKORC1 displaying similar functionality as mtbVKOR. Genetic selection and screening was done on an EMS mutagenized ΔdsbB strain expressing rat VKORC1wt from a plasmid for isolation of E.coli mutants that would facilitate complementation of the lack of DsbB by VKORC1wt. The principle for the genetic selection and screening is the restoration of disulfide bond forming pathway by replacement of DsbB. This phenotype of complementation can be assayed by restoration of motility, resistance to TCEP, and β-galactosidase inactivation on ∆dsbB strain. Results of the selection and screening process revealed mutations in the VKORC1 gene instead of the E.coli chromosome. On the other hand, we used a rational approach other than genetic screening. This approach involved targeting YidC hydrophilic groove, previously identified upon selection of mutants that facilitated functional expression of VKORC1∆AAR; a deletion of amino acids 31–33 (AAR), where other mutations inactivating protease HslV were also identified. For this approach, chromosomal mutations were introduced on selected residues in the YidC hydrophilic groove then functional expression of VKORC1wt or enhanced expression of VKORC1∆AAR in the new strains was assayed. We identified novel YidC mutants enhancing the expression of VKORC1∆AAR. From the analysis of these mutations and the VKORC1 mutations obtained from the screens, we concluded that the charge imbalance by VKORC1wt violates the positive-inside rule impeding its ability to substitute DsbB. The correct assembly of VKORC1∆AAR provided insight on the involvement of E.coli YidC in correct folding and insertion of foreign membrane proteins, with the hydrophilic groove being core for its membrane insertase functions. The improved functional expression of VKORC1∆AAR upon HslV inactivation provided insight on the mis-assembly of foreign membrane proteins as a quality control system. These findings suggested that the native VKORC1wt may not assemble properly in the E.coli inner membrane, and is degraded by proteases as a quality control mechanism.

Extracellular vesicles (EVs) are small nano-sized particles released constantly by cells into body fluids like blood, saliva, urine, plasma and milk. Depending in the size, EVs are divided into exosomes (30–100 nm), microvesicles (100–1000 nm) and apoptotic bodies (50–5000 nm). During this work, I extracted exosomes from kidney cancer cells using two different centrifugation methods (sequential centrifugation and sucrose gradient ultracentrifugation) under two different conditions (hypoxia and normoxia). The characterization was done by electron microscopy, western blot and nanoparticle tracking analysis (NTA). The effect of exosomes in normal kidney cells was studied in vitro by treating metanephric mesenchyme (MM) cells with exosomes for cell proliferation and motility. The exosomes were injected in chicken embryos in vivo together with renal carcinoma (Renca YFP) cells to see if they have some role in tumor growth. The protein and RNA contents of the exosomes were also analyzed. The cells release more vesicles when they are exposed to hypoxic conditions. Electron microscopy and western blot showed the presence of exosomes expressing CD63 and CD81 markers. Cell proliferation and motility was found to enhance when cells were treated with exosomes. Chicken embryos showed formation of bigger Renca YFP tumor after treatment with exosomes. Different proteins and miRNAs were detected in exosomes which may play active roles in biological processes. In summary, I successfully purified exosomes from kidney cancer cells and characterized them. We concluded that these vesicles play important roles in cell activity and have ability to enhance the tumor growth. The presence of proteins and miRNAs suggest the potential role in cellular communication.

In previous studies we have found that urinary fibrinogen-derived peptides are potential tumor markers for renal cell carcinoma. These peptides occur at low concentrations in urine. Identification of a low-abundant tumor marker requires optimal sample preparation and a highly sensitive analyzer. In this work different chromatographic and mass spectrometric methods were compared and evaluated for tumor marker discovery. We used urine samples from patients with renal cell carcinoma and prostate cancer. Our main targets were peptides derived from fibrinogen beta with unknown sequence that are produced by differential proteolysis. With our optimized workflow we discovered 26 fibrinogen beta derived peptides that have not been identified in urine previously.

<div lang="en" class="abs"> Abstract The increasing awareness of the causes and consequences of climate chance has led to actions to reduce the dependency on oil and other finite energy and raw material sources. Plant biomass is used in increasing amounts as a resource for biofuel, biochemical and fiber production. Carbohydrate enzymology has provided new ways to utilize and modify renewable carbon sources, especially the lignocellulolytic systems of fungi. Cellulolytic enzymes work in a synergistic manner on recalcitrant structure of cellulose, hydrolyzing it into soluble oligosaccharides, and eventually, glucose. Lytic polysaccharide monooxygenases (LPMOs) contribute to this system by oxidizing either C1- or C4-carbon from the carbohydrate chain on a crystalline cellulose with the help of copper-core induced radicals, thus creating available substrates for the other cellulolytic enzymes. Since their discovery in 2010, the research on their activities and specificities have increased rapidly, but the analytical methods to investigate this diverse group of enzymes is mostly limited to short and soluble products, which are only a fraction of the oxidation products. In addition, most of the methods require special equipment, wide range of standards and expertise to interpret the results. In this study, HPLC and HPAEC-PAD were tested, unsuccessfully, to quantify soluble products from LPMO-catalysis. A TTC-method, in which 2,3,5-triphenyl-2H-tetrazolium chloride is reduced into red and spectrophotometrically quantifiable formazan by reducing ends from insoluble LPMO-products, was successfully optimized and downscaled, and used to optimize reaction conditions for a type 3 LPMO from Trichoderma reesei, TrAA9A, with Whatman filter paper 1 as a substrate. Experiments were conducted to investigate the effects of pH, temperature, donor, time and the presence/absence of H₂O₂ to the accumulation of reducing ends. The results did not show any substantial differences in the accumulation of aldehydes in different reaction conditions. This study showed that cellulose degrades in the presence of TrAA9A and an electron donor. The greatest effects were observed with longer reaction times and the addition of H₂O₂, both increasing the amount of measured aldehydes in the insoluble products. The highest yield was recorded from the reactions with gallic acid as a donor at pH 6, and in the presence of 0.7 mM H₂O₂. The results from this study could lead to understanding the rate-limiting factors of the LPMOs and further improve the utilization of this enzyme in the degradation of lignocellulosic biomass.

Planejamento probabilístico lida com a tomada de decisão sequencial em ambientes estocásticos e geralmente é modelado por um Processo de Decisão Markoviano (Markovian Decision Process - MDP). Um MDP modela a interação entre um agente e o seu ambiente: em cada estágio, o agente decide executar uma ação, com efeitos probabilísticos e um certo custo, que irá produzir um estado futuro. O objetivo do agente MDP é minimizar o custo esperado ao longo de uma sequência de escolhas de ação. O número de estágios que o agente atua no ambiente é chamado de horizonte, o qual pode ser finito, infinito ou indefinido. Um exemplo de MDP com horizonte indefinido é o Stochastic Shortest Path MDP (SSP MDP), que estende a definição de MDP adicionando um conjunto de estados meta (o agente para de agir ao alcançar um estado meta). Num SSP MDP é feita a suposição de que é sempre possível alcançar um estado meta a partir de qualquer estado do mundo. No entanto, essa é uma suposição muito forte e que não pode ser garantida em aplicações práticas. Estados a partir dos quais é impossível atingir a meta são chamados de becos-sem-saída. Um beco-sem-saída pode ser evitável ou inevitável (se nenhuma política leva do estado inicial para a meta com probabilidade um). Em trabalhos recentes foram propostas extensões para SSP MDP que permitem a existência de diferentes tipos de beco-sem-saída, bem como algoritmos para resolvê-los. No entanto, a detecção de becos-sem-saída é feita utilizando: (i) heurísticas que podem falhar para becos-sem-saída implícitos ou (ii) métodos mais confiáveis, mas que demandam alto custo computacional. Neste projeto fazemos uma caracterização formal de modelos de planejamento probabilístico com becos-sem-saída. Além disso, propomos uma nova técnica para detecção de becos-sem-saída baseada nessa caracterização e adaptamos algoritmos de planejamento probabilístico para utilizarem esse novo método de detecção. Os resultados empíricos mostram que o método proposto é capaz de detectar todos os becos-sem-saída de um dado conjunto de estados e, quando usado com planejadores probabilísticos, pode tornar esses planejadores mais eficientes em domínios com becos-sem-saída difíceis de serem detectados / Probabilistic planning deals with sequential decision making in stochastic environments and is modeled by a Markovian Decision Process (MDP). An MDP models the interaction between an agent and its environment: at each stage, the agent decides to execute an action, with probabilistic effects and a certain cost which produces a future state. The purpose of the MDP agent is to minimize the expected cost along a sequence of choices. The number of stages that the agent acts in the environment is called horizon, which can be finite, infinite or undefined. An example of MDP with undefined horizon is the Stochastic Shortest Path MDP, which extends the definition of MDP by adding a set of goal states (the agent stops acting after reaching a goal state). In an SSP MDP the assumption is made that it is always possible to achieve a goal state from every state of the world. However, this is a very strong assumption and cannot be guaranteed in practical applications. States from which it is impossible to reach the goal are called dead-ends. A dead-end may be avoidable or unavoidable (when no policy leads from the initial state to the goal with probability one). Recent work has proposed extensions to SSP MDP that allow the existence of different types of dead-ends as well as algorithms to solve them. However, the detection of dead-end is done using: (i) heuristics that may fail to detect implicitly dead-ends or (ii) more reliable methods that require a high computational cost. In this project we make a formal characterization of probabilistic planning models with dead-ends. In addition, we propose a new technique for dead-end detection based on this characterization and we adapt probabilistic planning algorithms to use this new detection method. The empirical results show that the proposed method is able to detect all dead-ends of a given set of states and, when used withprobabilistic planners, can make these planners more efficient in domains with difficult to detect dead-ends.

Dois modelos principais de planejamento em inteligência artificial são os usados, respectivamente, em planejamento probabilístico (MDPs e suas generalizações) e em planejamento não-determinístico (baseado em model checking). Nessa dissertação será: (1) exibido que planejamento probabilístico e não-determinístico são extremos de um rico contínuo de problemas capaz de lidar simultaneamente com risco e incerteza (Knightiana); (2) obtido um modelo para unificar esses dois tipos de problemas usando MDPs imprecisos; (3) derivado uma versão simplificada do princípio ótimo de Bellman para esse novo modelo; (4) exibido como adaptar e analisar algoritmos do estado-da-arte, como (L)RTDP e LDFS, nesse modelo unificado. Também será discutido exemplos e relações entre modelos já propostos para planejamento sob incerteza e o modelo proposto. / Two noteworthy models of planning in AI are probabilistic planning (based on MDPs and its generalizations) and nondeterministic planning (mainly based on model checking). In this dissertation we: (1) show that probabilistic and nondeterministic planning are extremes of a rich continuum of problems that deal simultaneously with risk and (Knightian) uncertainty; (2) obtain a unifying model for these problems using imprecise MDPs; (3) derive a simplified Bellman\'s principle of optimality for our model; and (4) show how to adapt and analyze state-of-art algorithms such as (L)RTDP and LDFS in this unifying setup. We discuss examples and connections to various proposals for planning under (general) uncertainty.