<p>Next generation sequencing technologies have provided us with a wealth of data profiling a diverse range of biological processes. In an effort to better understand the process of gene regulation, two predictive machine learning models specifically tailored for analyzing gene transcription and polyadenylation are presented.</p><p>Transcriptional enhancers are specific DNA sequences that act as ``information integration hubs" to confer regulatory requirements on a given cell. These non-coding DNA sequences can regulate genes from long distances, or across chromosomes, and their relationships with their target genes are not limited to one-to-one. With thousands of putative enhancers and less than 14,000 protein-coding genes, detecting enhancer-gene pairs becomes a very complex machine learning and data analysis challenge. </p><p>In order to predict these specific-sequences and link them to genes they regulate, we developed McEnhancer. Using DNAseI sensitivity data and annotated in-situ hybridization gene expression clusters, McEnhancer builds interpolated Markov models to learn enriched sequence content of known enhancer-gene pairs and predicts unknown interactions in a semi-supervised learning algorithm. Classification of predicted relationships were 73-98% accurate for gene sets with varying levels of initial known examples. Predicted interactions showed a great overlap when compared to Hi-C identified interactions. Enrichment of known functionally related TF binding motifs, enhancer-associated histone modification marks, along with corresponding developmental time point was highly evident.</p><p>On the other hand, pre-mRNA cleavage and polyadenylation is an essential step for 3'-end maturation and subsequent stability and degradation of mRNAs. This process is highly controlled by cis-regulatory elements surrounding the cleavage site (polyA site), which are frequently constrained by sequence content and position. More than 50\% of human transcripts have multiple functional polyA sites, and the specific use of alternative polyA sites (APA) results in isoforms with variable 3'-UTRs, thus potentially affecting gene regulation. Elucidating the regulatory mechanisms underlying differential polyA preferences in multiple cell types has been hindered by the lack of appropriate tests for determining APAs with significant differences across multiple libraries. </p><p>We specified a linear effects regression model to identify tissue-specific biases indicating regulated APA; the significance of differences between tissue types was assessed by an appropriately designed permutation test. This combination allowed us to identify highly specific subsets of APA events in the individual tissue types. Predictive kernel-based SVM models successfully classified constitutive polyA sites from a biologically relevant background (auROC = 99.6%), as well as tissue-specific regulated sets from each other. The main cis-regulatory elements described for polyadenylation were found to be a strong, and highly informative, hallmark for constitutive sites only. Tissue-specific regulated sites were found to contain other regulatory motifs, with the canonical PAS signal being nearly absent at brain-specific sites. We applied this model on SRp20 data, an RNA binding protein that might be involved in oncogene activation and obtained interesting insights. </p><p>Together, these two models contribute to the understanding of enhancers and the key role they play in regulating tissue-specific expression patterns during development, as well as provide a better understanding of the diversity of post-transcriptional gene regulation in multiple tissue types.</p> / Dissertation

The global intent of this research was to confirm the validity of the interpolated twitch technique (ITT) to voluntary torque model (first study) and then to utilize this technique in developing definitive criterion measures enhancing the study of training strategies on central fatigue (second study) and velocity specific (third study) voluntary activation (VA) and force production outcomes. The first study assessed the contribution of non-stimulated synergists to the non-linearity of the ITT voluntary torque relationship for elbow flexion contraction. Since multiple synergist stimulation was found not to significantly improve the ITT to voluntary torque relationship, follow-up pilot work was conducted in which the quadriceps femoris was identified as a valid ITT test model muscle. The first training study investigated the effects of high volume, maximum voluntary isometric contraction (MVIC) resistance training of the quadriceps femoris on MVIC levels of force and VA prior to, during, and recovery from a standardized fatigue test protocol. Results showed significant increases in pre-fatigue MVIC VA and force, increases in resistance to early fatigue, but also increased overall rate of fatigue resulting in non-significant changes in total force volume and endurance time. Post-fatigue analysis showed significant training increases in rate and level of recovery for both MVIC VA and force production. The second training study investigated the effects of high resistance ramp versus ballistic type MVIC strength training of the quadriceps femoris on central adaptations in submaximal and maximal levels of MVIC VA and force production. Results showed similar ramp and ballistic group training increases in MVIC force and VA on both ramp and ballistic tests. Ballistic group, ballistic test change scores were significant but not for the ramp test, suggesting ballistic group test-training specificity. The ballistic group showed significant force increases on the submaximal 150ms ballistic test, while the ramp group showed non-significant training changes, further supporting test-training specificity for the ballistic group. Results support utility of the ITT and provide valuable information with regard to training and test-training specificity considerations.

We examine the amalgamated free product of hyperfinite von Neumann algebras. First we describe the amalgamated free product of hyperfinite von Neumann algebras over finite dimensional subalgebras. In this case the result is always the direct sum of a hyperfinite von Neumann algebra and a finite number of interpolated free group factors. We then show that this class is closed under this type of amalgamated free product. After that we allow amalgamation over possibly infinite dimensional multimatrix subalgebras. In this case the product of two hyperfinite von Neumann algebras is the direct sum of a hyperfinite von Neumann algebra and a countable direct sum of interpolated free group factors. As before, we show that this class is closed under amalgamated free products over multimatrix algebras.

No description available.

The purpose of this diploma thesis is to analyze possibilities of non-photorealistic rendering focused on area of image transformation and afterwards design algorithm for video postprocessing to it's cartoon representation. The thesis describes problem analysis, design of the algorithm and it's implementation.

State of the art filter bank structures achieve practically perfect reconstruction with very high computational efficiency. However, the increase in computational requirements due to the need to process increasingly wider band signals is paramount. New filter bank structures that provide extra information about a signal while achieving the same level of required efficiency, and perfect reconstruction properties, need to be developed. In this work a new filter bank structure, the interpolated FIR (IFIR) filter bank is developed. Such a structure combines the concepts of filter banks, and interpolated FIR filters. The filter design procedures for the IFIR filter bank are developed and explained. The resulting structure was compared with the non-maximally-decimated filter bank (NMDFB), achieving the same performance in terms of the number of multiplications required per sample and the overall distortion introduced by the system, when operating with Nyquist prototype filters. In addition, the IFIR filter is tested in both simulated and real communication environments. Performance, in terms of bit-error-rate, was found to not be degraded significantly when using the IFIR filter bank system for transmission and reception of QPSK symbols. / Master of Science

A precipitação é uma das principais variáveis hidrológicas analisadas antes de ser fornecida como dado de entrada aos modelos chuva-vazão. As dificuldades para quantificar o valor da chuva, de modo a representar o evento que verdadeiramente ocorre na bacia hidrográfica, provêm principalmente da alta variabilidade espacial da precipitação, da quantidade de postos distribuídos em uma área, das falhas existentes nas séries pluviométricas (falhas nos aparelhos, no processamento dos dados e por perdas dos dados). Uma das formas de quantificar o valor da chuva em uma determinada área para uso em modelos concentrados chuva-vazão, a partir de uma rede de pluviômetros, pode ser com base na Media Aritmética ou na Mediana, que foi uma das propostas desta pesquisa. Os métodos para se obter estas estatísticas pontuais seriam mediante a utilização dos cálculos diretos aplicados nos postos presentes dentro dos limites da bacia, ou através dos valores dos nós de uma grade regular localizados no interior da bacia, com os valores destes nós estimados por interpolação de medidas pontuais posicionadas no interior e fora dos limites da bacia. Os interpoladores utilizados foram o Inverso do Quadrado da Distância, O Vizinho mais Próximo, O Vizinho Natural e a Triangulação Linear. O método de interpolação mais adequado, para as situações encontradas nas séries de precipitação, foi selecionado com o uso da Técnica da Validação Cruzada, ao qual, o método do Vizinho Natural apresentou melhor desempenho. As séries das precipitações médias e medianas diárias obtidas pelos Cálculos Diretos, e após a espacialização com o Interpolador Vizinho Natural, foram utilizadas como dados de entrada no modelo chuva-vazão concentrado conceitual IPH II, versão WIN_IPH II, em diferentes escalas de bacias hidrográficas embutidas (9426 a 19,5 Km² ). Os resultados das vazões simuladas pelo modelo, após os processos de calibração, com o uso da calibração monoobjetivo SCE-UA, evidenciaram o melhor método, Cálculos Diretos ou Interpolação, e a estatística pontual, Media ou Mediana, que foi mais adequada a partir do desempenho do modelo nas etapas de calibração e de verificação dos parâmetros resultantes. Concluiu-se que além das falhas existentes influenciarem nas chuvas médias e medianas obtidas por cálculos diretos, a mediana não é uma estatística que representa bem o volume de chuva ocorrente na bacia, pois produziu baixo desempenho no modelo e valores de parâmetros sem significado físico para bacias rurais. Observou-se uma melhora nas duas estatísticas pontuais, após o cálculo da chuva média e da mediana com base nos valores dos nós da grade regular no interior da bacia, estimados pelo interpolador Vizinho Natural. O uso das estatísticas pontuais obtidas por este método produziu bom desempenho do modelo nas etapas de calibração e de verificação. A chuva média obtida por interpolação com o Vizinho Natural foi a estatística mais representativa para a maior parte das escalas analisadas. Este trabalho também mostrou a possibilidade de se fazer à transferência de parâmetros do modelo IPH II, com o uso da melhor estatística pontual, de bacias maiores para bacias menores, com o intuito de verificar situações para obtenção de vazões em bacias menores que não possuem medições fluviométricas. / Precipitation is a key hydrological variables analyzed before being supplied as input to rainfall-runoff models. The difficulties to quantify the amount of rain in order to truly represent the event that occurs in the watershed, primarily come from the high spatial variability of precipitation, the number of stations distributed in an area of flaws in the rainfall series (faults in apparatus in data processing and for loss of data). One way to quantify the amount of rainfall in a given area for use in concentrated rainfall-runoff models, from a network of rain gauges, can be based on the arithmetic mean or median, which was one of the purposes of this research. The methods to obtain these statistics would be off by using the direct calculations applied in these posts within the limits of the basin, or through the values of the nodes of a regular grid located inside the basin, we estimated these values by interpolation offs positioned inside and outside the boundaries of the basin. The interpolators used were the Inverse Square Distance, Nearest Neighbor, The Natural Neighbor and Linear Triangulation. The interpolation method most appropriate to the situations found in the precipitation series, was selected by using the Cross Validation technique, which the Natural Neighbor method showed better performance. The series of daily precipitation means and medians obtained by direct calculations, and after the spatial interpolator with Natural Neighbor, were used as input to rainfall-runoff model focused conceptual HPI II, version WIN_IPH II, at different scales of watersheds embedded (9426 to 19.5 km ²). The results of the simulated flows by the model after the calibration procedures, calibration using the SCE-UA monoobjetivo, showed the best method, Direct calculations or interpolation, and timely statistics, mean or median, which was more appropriate to the performance stages of the model calibration and verification of resulting parameters. It was concluded that besides the existing faults influence on rainfall averages and medians obtained by direct calculations, the median is a statistic that well represents the volume of rainfall occurring in the basin, because it produced lower scores in the model and parameter values without physical meaning to rural basins. We observed an improvement in the two point statistics, after calculating the mean and median rainfall based on the values of the nodes of regular grid within the basin, estimated by the Natural Neighbor interpolation. The occasional use of statistics obtained by this method produced good performance of the model in steps of calibration and verification. The average rainfall obtained by interpolation from the Natural Neighbor was statistically more representative for most of the scales analyzed. This work also showed the possibility of making the transfer of model parameters HPI II, using the best statistical point of basins larger for smaller basins, in order to check conditions for obtaining flows in smaller basins that have no measurements gauged.

A precipitação é uma das principais variáveis hidrológicas analisadas antes de ser fornecida como dado de entrada aos modelos chuva-vazão. As dificuldades para quantificar o valor da chuva, de modo a representar o evento que verdadeiramente ocorre na bacia hidrográfica, provêm principalmente da alta variabilidade espacial da precipitação, da quantidade de postos distribuídos em uma área, das falhas existentes nas séries pluviométricas (falhas nos aparelhos, no processamento dos dados e por perdas dos dados). Uma das formas de quantificar o valor da chuva em uma determinada área para uso em modelos concentrados chuva-vazão, a partir de uma rede de pluviômetros, pode ser com base na Media Aritmética ou na Mediana, que foi uma das propostas desta pesquisa. Os métodos para se obter estas estatísticas pontuais seriam mediante a utilização dos cálculos diretos aplicados nos postos presentes dentro dos limites da bacia, ou através dos valores dos nós de uma grade regular localizados no interior da bacia, com os valores destes nós estimados por interpolação de medidas pontuais posicionadas no interior e fora dos limites da bacia. Os interpoladores utilizados foram o Inverso do Quadrado da Distância, O Vizinho mais Próximo, O Vizinho Natural e a Triangulação Linear. O método de interpolação mais adequado, para as situações encontradas nas séries de precipitação, foi selecionado com o uso da Técnica da Validação Cruzada, ao qual, o método do Vizinho Natural apresentou melhor desempenho. As séries das precipitações médias e medianas diárias obtidas pelos Cálculos Diretos, e após a espacialização com o Interpolador Vizinho Natural, foram utilizadas como dados de entrada no modelo chuva-vazão concentrado conceitual IPH II, versão WIN_IPH II, em diferentes escalas de bacias hidrográficas embutidas (9426 a 19,5 Km² ). Os resultados das vazões simuladas pelo modelo, após os processos de calibração, com o uso da calibração monoobjetivo SCE-UA, evidenciaram o melhor método, Cálculos Diretos ou Interpolação, e a estatística pontual, Media ou Mediana, que foi mais adequada a partir do desempenho do modelo nas etapas de calibração e de verificação dos parâmetros resultantes. Concluiu-se que além das falhas existentes influenciarem nas chuvas médias e medianas obtidas por cálculos diretos, a mediana não é uma estatística que representa bem o volume de chuva ocorrente na bacia, pois produziu baixo desempenho no modelo e valores de parâmetros sem significado físico para bacias rurais. Observou-se uma melhora nas duas estatísticas pontuais, após o cálculo da chuva média e da mediana com base nos valores dos nós da grade regular no interior da bacia, estimados pelo interpolador Vizinho Natural. O uso das estatísticas pontuais obtidas por este método produziu bom desempenho do modelo nas etapas de calibração e de verificação. A chuva média obtida por interpolação com o Vizinho Natural foi a estatística mais representativa para a maior parte das escalas analisadas. Este trabalho também mostrou a possibilidade de se fazer à transferência de parâmetros do modelo IPH II, com o uso da melhor estatística pontual, de bacias maiores para bacias menores, com o intuito de verificar situações para obtenção de vazões em bacias menores que não possuem medições fluviométricas. / Precipitation is a key hydrological variables analyzed before being supplied as input to rainfall-runoff models. The difficulties to quantify the amount of rain in order to truly represent the event that occurs in the watershed, primarily come from the high spatial variability of precipitation, the number of stations distributed in an area of flaws in the rainfall series (faults in apparatus in data processing and for loss of data). One way to quantify the amount of rainfall in a given area for use in concentrated rainfall-runoff models, from a network of rain gauges, can be based on the arithmetic mean or median, which was one of the purposes of this research. The methods to obtain these statistics would be off by using the direct calculations applied in these posts within the limits of the basin, or through the values of the nodes of a regular grid located inside the basin, we estimated these values by interpolation offs positioned inside and outside the boundaries of the basin. The interpolators used were the Inverse Square Distance, Nearest Neighbor, The Natural Neighbor and Linear Triangulation. The interpolation method most appropriate to the situations found in the precipitation series, was selected by using the Cross Validation technique, which the Natural Neighbor method showed better performance. The series of daily precipitation means and medians obtained by direct calculations, and after the spatial interpolator with Natural Neighbor, were used as input to rainfall-runoff model focused conceptual HPI II, version WIN_IPH II, at different scales of watersheds embedded (9426 to 19.5 km ²). The results of the simulated flows by the model after the calibration procedures, calibration using the SCE-UA monoobjetivo, showed the best method, Direct calculations or interpolation, and timely statistics, mean or median, which was more appropriate to the performance stages of the model calibration and verification of resulting parameters. It was concluded that besides the existing faults influence on rainfall averages and medians obtained by direct calculations, the median is a statistic that well represents the volume of rainfall occurring in the basin, because it produced lower scores in the model and parameter values without physical meaning to rural basins. We observed an improvement in the two point statistics, after calculating the mean and median rainfall based on the values of the nodes of regular grid within the basin, estimated by the Natural Neighbor interpolation. The occasional use of statistics obtained by this method produced good performance of the model in steps of calibration and verification. The average rainfall obtained by interpolation from the Natural Neighbor was statistically more representative for most of the scales analyzed. This work also showed the possibility of making the transfer of model parameters HPI II, using the best statistical point of basins larger for smaller basins, in order to check conditions for obtaining flows in smaller basins that have no measurements gauged.

A precipitação é uma das principais variáveis hidrológicas analisadas antes de ser fornecida como dado de entrada aos modelos chuva-vazão. As dificuldades para quantificar o valor da chuva, de modo a representar o evento que verdadeiramente ocorre na bacia hidrográfica, provêm principalmente da alta variabilidade espacial da precipitação, da quantidade de postos distribuídos em uma área, das falhas existentes nas séries pluviométricas (falhas nos aparelhos, no processamento dos dados e por perdas dos dados). Uma das formas de quantificar o valor da chuva em uma determinada área para uso em modelos concentrados chuva-vazão, a partir de uma rede de pluviômetros, pode ser com base na Media Aritmética ou na Mediana, que foi uma das propostas desta pesquisa. Os métodos para se obter estas estatísticas pontuais seriam mediante a utilização dos cálculos diretos aplicados nos postos presentes dentro dos limites da bacia, ou através dos valores dos nós de uma grade regular localizados no interior da bacia, com os valores destes nós estimados por interpolação de medidas pontuais posicionadas no interior e fora dos limites da bacia. Os interpoladores utilizados foram o Inverso do Quadrado da Distância, O Vizinho mais Próximo, O Vizinho Natural e a Triangulação Linear. O método de interpolação mais adequado, para as situações encontradas nas séries de precipitação, foi selecionado com o uso da Técnica da Validação Cruzada, ao qual, o método do Vizinho Natural apresentou melhor desempenho. As séries das precipitações médias e medianas diárias obtidas pelos Cálculos Diretos, e após a espacialização com o Interpolador Vizinho Natural, foram utilizadas como dados de entrada no modelo chuva-vazão concentrado conceitual IPH II, versão WIN_IPH II, em diferentes escalas de bacias hidrográficas embutidas (9426 a 19,5 Km² ). Os resultados das vazões simuladas pelo modelo, após os processos de calibração, com o uso da calibração monoobjetivo SCE-UA, evidenciaram o melhor método, Cálculos Diretos ou Interpolação, e a estatística pontual, Media ou Mediana, que foi mais adequada a partir do desempenho do modelo nas etapas de calibração e de verificação dos parâmetros resultantes. Concluiu-se que além das falhas existentes influenciarem nas chuvas médias e medianas obtidas por cálculos diretos, a mediana não é uma estatística que representa bem o volume de chuva ocorrente na bacia, pois produziu baixo desempenho no modelo e valores de parâmetros sem significado físico para bacias rurais. Observou-se uma melhora nas duas estatísticas pontuais, após o cálculo da chuva média e da mediana com base nos valores dos nós da grade regular no interior da bacia, estimados pelo interpolador Vizinho Natural. O uso das estatísticas pontuais obtidas por este método produziu bom desempenho do modelo nas etapas de calibração e de verificação. A chuva média obtida por interpolação com o Vizinho Natural foi a estatística mais representativa para a maior parte das escalas analisadas. Este trabalho também mostrou a possibilidade de se fazer à transferência de parâmetros do modelo IPH II, com o uso da melhor estatística pontual, de bacias maiores para bacias menores, com o intuito de verificar situações para obtenção de vazões em bacias menores que não possuem medições fluviométricas. / Precipitation is a key hydrological variables analyzed before being supplied as input to rainfall-runoff models. The difficulties to quantify the amount of rain in order to truly represent the event that occurs in the watershed, primarily come from the high spatial variability of precipitation, the number of stations distributed in an area of flaws in the rainfall series (faults in apparatus in data processing and for loss of data). One way to quantify the amount of rainfall in a given area for use in concentrated rainfall-runoff models, from a network of rain gauges, can be based on the arithmetic mean or median, which was one of the purposes of this research. The methods to obtain these statistics would be off by using the direct calculations applied in these posts within the limits of the basin, or through the values of the nodes of a regular grid located inside the basin, we estimated these values by interpolation offs positioned inside and outside the boundaries of the basin. The interpolators used were the Inverse Square Distance, Nearest Neighbor, The Natural Neighbor and Linear Triangulation. The interpolation method most appropriate to the situations found in the precipitation series, was selected by using the Cross Validation technique, which the Natural Neighbor method showed better performance. The series of daily precipitation means and medians obtained by direct calculations, and after the spatial interpolator with Natural Neighbor, were used as input to rainfall-runoff model focused conceptual HPI II, version WIN_IPH II, at different scales of watersheds embedded (9426 to 19.5 km ²). The results of the simulated flows by the model after the calibration procedures, calibration using the SCE-UA monoobjetivo, showed the best method, Direct calculations or interpolation, and timely statistics, mean or median, which was more appropriate to the performance stages of the model calibration and verification of resulting parameters. It was concluded that besides the existing faults influence on rainfall averages and medians obtained by direct calculations, the median is a statistic that well represents the volume of rainfall occurring in the basin, because it produced lower scores in the model and parameter values without physical meaning to rural basins. We observed an improvement in the two point statistics, after calculating the mean and median rainfall based on the values of the nodes of regular grid within the basin, estimated by the Natural Neighbor interpolation. The occasional use of statistics obtained by this method produced good performance of the model in steps of calibration and verification. The average rainfall obtained by interpolation from the Natural Neighbor was statistically more representative for most of the scales analyzed. This work also showed the possibility of making the transfer of model parameters HPI II, using the best statistical point of basins larger for smaller basins, in order to check conditions for obtaining flows in smaller basins that have no measurements gauged.

No description available.