Sudden cardiac death represents one of the leading causes of death worldwide, with the majority of these deaths caused by arrhythmias derived from ischæmic events. However, the mechanisms leading from ischæmia to re-entry, arrhythmia and eventual death are poorly understood. Furthermore, variability in the action potential of cardiac tissue, while important in determining arrhythmic risk, is only recently being addressed in computational modelling, with little known about the causes and mechanisms underlying it, nor regarding its evolution in response to pathological conditions such as ischæmia. This dissertation investigates the causes of variability in the repolarisation of the action potential of the rabbit ventricular myocyte, and the response of this variability to ischæmia. The effect of variability in ion channel conductances is investigated by means of a complete search of the parameter space revealed by simultaneous variation in multiple parameters describing ion channel conductances in computational models of the rabbit ventricular action potential. Rabbit data and models are used in this thesis due to the similarities to human data, both in terms of electrophysiology generally, and the response to ischæmia specifically. The response of two different model frameworks is assessed to determine similarities and differences between model frameworks that are designed to reproduce the same system. Those models producing action potential durations that fall within an experimentally derived range at multiple pacing rates are used to define model populations that thus reproduce experimental variability in repolarisation. These model populations are used to investigate the effects of ischæmic conditions on population variability. Variability is measured not only for action potential duration, but also for other biomarkers commonly implicated in the development of re-entry. The work presented in this dissertation is significant for: (1) presenting a comprehensive study of the effect of simultaneous variation in ion channel conductances, with details regarding the interactions between conductances and how these interactions change depending on the pacing rate; (2) detailed examination of the differences between two models of the same system; (3) production of the largest extant populations reproducing experimentally observed variability in action potential duration; (4) the first time model populations have been used to investigate the effects of ischæmia on variability.

This thesis investigates how collective decisions in bird flocks arise from simple rules, the factors that give some birds more influence over a flock's direction, and how travelling as a flock affects spatial learning. I used GPS loggers to track pigeons homing alone and in flocks, and applied mathematical modelling to explore the mechanisms underlying group decisions. Across several experiments, the key results were as follows: Flying home with a more experienced individual not only gave a pigeon an immediate advantage in terms of taking a more direct route, but the followers also learned homing routes just as accurately as pigeons flying alone. This shows that using social cues did not interfere with learning about the landscape during a series of paired flights. Pigeons that were faster during solo homing flights also tended to fly at the front of flocks, where they had more influence over the direction taken. Analysis of momentary interactions during paired flights and simulations of pair trajectories support the conclusion that speed increases the likelihood of leading. A pigeon's solo homing efficiency before flock flights did not correlate with leadership in flocks of ten, but leaders did have more efficient solo tracks when tested after a series of flock flights. A possible explanation is that leaders attended more to the landscape and therefore learned faster. Flocks took straighter routes than pigeons flying alone, as would be expected if they effectively pooled information. In addition, pigeons responded more strongly to the direction of several neighbours, during flock flights, than to a single neighbour during paired flights. This behaviour makes sense adaptively because social information will be more reliable when following several conspecifics compared to one. Through a combination of high-resolution tracking and mathematical modelling, this thesis sheds light on the mechanisms of flocking and its navigational consequences.

Specific recognition of pathogenic molecules by the immune system is mediated by proteins known as antigen receptors. One such component is the antibody. Binding properties of natural and engineered antibodies can be understood by studying the structure of their variable domains, VH and VL. In this thesis we investigate how the two variable domains orientate with respect to one another and therefore influence the geometry of the antigen binding site which is formed between them. We describe a method which fully characterises the VH-VL orientation in a consistent and absolute sense using five angles and a distance. The ABangle method is used to investigate variable domain orientation in structures collected by our database SAbDab. Using the ABangle method we compare VH-VL orientation to the corresponding property in a different component of the immune system, the T-cell receptor (TCR). Despite having similar individual domain structures the variable domain orientations of antibodies and TCRs are found to be distinct. This is found to affect an antibody’s ability to mimic TCR specificity. ABangle's characterisation is used to find determinants of the VH-VL orientation. We identify sequence and structural properties that influence the variable domain pose. A feature based method for predicting VH-VL orientation is presented and assessed. Future directions of this research and its application to the development of antibody therapeutics are described.

"The complex growth patterns of malignant brain tumors can present challenges in developing accurate models. In particular, the computational costs associated with modeling a realistically sized tumor can be prohibitive. The use of high-performance computing (HPC) and novel mathematical techniques can help to overcome this barrier. This paper presents a parallel implementation of a model for the growth of glioma, a form of brain cancer, and discusses how HPC is being used to take a first step toward realistically sized tumor models. Also, consideration is given to the visualization process involved with large-scale computing. Finally, simulation data is presented with a focus on scaling."

A identificação e caracterização geométrica e físico-química de espaços vazios na estrutura tridimensional de proteínas é capaz de agregar informações importantes para guiar o desenho racional de drogas e a caracterização funcional de sítios de ligação e sítios catalíticos. Dessa forma, algumas ferramentas computacionais foram desenvolvidas nas últimas duas décadas, visando efetuar essas caracterizações. Contudo, as ferramentas existentes lidam com uma série de limitações, dais quais merecem destaque a falta de precisão, falta de capacidade de integração em protocolos de larga escala, falta de capacidade de customização e a falta de uma caracterização eletrostática . Tendo em mente estas limitações, desenvolvemos uma nova ferramenta, denominada KV-Finder, com o objetivo de estender as funcionalidades dos programas existentes, fornecendo assim uma caracterização sistemática mais eficiente e mais informativa dos espaços vazios da estrutura tridimensional de proteínas. Através de uma modelagem matricial baseada em um direcionamento realizado pelo usuário, nossa ferramenta identifica e caracteriza espaços vazios em topologias proteicas. O utilitário é capaz de quantificar o volume, a forma, a extensão de sua superfície, os resíduos proteicos que interagem com os espaços vazios e um mapa de cargas parciais da superfície encontrada. Nossa rotina foi integrada com ferramentas gráficas de modelagem molecular, fornecendo uma interação fácil e eficiente com o usuário. A validação de nosso algoritmo foi realizada em um conjunto de proteínas cujos diversos tipos de espaços vazios englobam os mais variados sítios de ligação e sítios catalíticos. O cálculo do volume de cavidades enzimáticas foi efetuado em larga escala, acompanhando a evolução do tamanho de bolsões na superfamília ALDH. Com relação aos outros softwares existentes, nossa ferramenta apresenta uma série de vantagens das quais merecem destaque menor tempo de execução, maior precisão, maior acessibilidade e facilidade de integração com outros programas, além das características únicas de permitir que a busca ocorra em regiões específicas dentro da proteína e de realizar um mapeamento parcial de cargas da superfície encontrada. / The identification and characterization of geometrical and physical-chemical properties in protein vacant spaces aggregates important information for steering rational drug designing and functional characterization of binding and catalytic sites. Therefore, several softwares have been develop during the past two decades in order to perform such characterization. Nevertheless, the existing tools still present a series of limitations such as lack of precision, lack of integrability in large scale protocols, lack of customization capacity and the lack of a proper electrostatic depiction. We developed a new software, dubbed KV-Finder, in order to complement and extend the functionality of existing softwares, providing a systematic and more descriptive portrayal of protein vacant spaces. By employing a user-driven matrix modeling, our tool identifies and characterizes empty spaces in all sorts of protein topologies. The software quantifies the volume, the area and the shape of the surface, the residues that interact with the vacant spaces and a partial charge map of the computed surface. Our routine was integrated with a graphical molecular modeling software, providing the user with a simple and easy-to-use interface. KV-Finder has been validated with a distinct set of proteins and binding sites. The volume computation was carried in large scale, accompanying the evolution of the pocket volume in the ALDH superfamily. Compared with existing software, KV-Finder presents greater precision, greater accessibility and ease of integration in large scale protocols and visualization softwares. Also, the software possesses unique and innovative features such as the ability to segment and subsegment the empty spaces, a electrostatic depiction and a ligand interaction highlight feature.

Possibly the largest problem when working in bioinformatics is the large amount of data to sift through to find useful information. This thesis shows that the use of feature selection (a method of removing irrelevant and redundant information from the dataset) is a useful and even necessary technique to use in these large datasets. This thesis also presents a new method in comparing classes to each other through the use of their features. It also provides a thorough analysis of the use of various feature selection techniques and classifier in different scenarios from bioinformatics. Overall, this thesis shows the importance of the use of feature selection in bioinformatics. / by David Dittman. / Thesis (M.S.C.S.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

SRSF1 is a widely expressed mammalian protein with multiple functions in the regulation of gene expression through processes including transcription, mRNA splicing, and translation. Although much is known of SRSF1 role in alternative splicing of specific genes little is known about its functions as a transcription factor and its global effect on cellular gene expression. We utilized a RNA sequencing (RNA-¬‐Seq) approach to determine the impact of SRSF1 in on cellular gene expression and analyzed both the short term (12 hours) and long term (48 hours) effects of SRSF1 expression in a human cell line. Furthermore, we analyzed and compared the effect of the expression of a naturally occurring deletion mutant of SRSF1 (RRM12) to the full-¬‐length protein. Our analysis reveals that shortly after SRSF1 is over-¬‐expressed the transcription of several histone coding genes is down-¬‐regulated, allowing for a more relaxed chromatin state and efficient transcription by RNA Polymerase II. This effect is reversed at 48 hours. At the same time key genes for the immune pathways are activated, more notably Tumor Necrosis Factor-¬‐Alpha (TNF-¬‐α), suggesting a role for SRSF1 in T cell functions. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2017. / FAU Electronic Theses and Dissertations Collection

In recent years, the collection of genomic data has skyrocketed and databases of genomic data are growing at a faster rate than ever before. Although many computational methods have been developed to interpret these data, they tend to struggle to process the ever increasing file sizes that are being produced and fail to take advantage of the advances in multi-core processors by using parallel processing. In some instances, loss of accuracy has been a necessary trade off to allow faster computation of the data. This thesis discusses one such algorithm that has been developed and how changes were made to allow larger input file sizes and reduce the time required to achieve a result without sacrificing accuracy. An information entropy based algorithm was used as a basis to demonstrate these techniques. The algorithm dissects the distinctive patterns underlying genomic data efficiently requiring no a priori knowledge, and thus is applicable in a variety of biological research applications. This research describes how parallel processing and object-oriented programming techniques were used to process larger files in less time and achieve a more accurate result from the algorithm. Through object oriented techniques, the maximum allowable input file size was significantly increased from 200 mb to 2000 mb. Using parallel processing techniques allowed the program to finish processing data in less than half the time of the sequential version. The accuracy of the algorithm was improved by reducing data loss throughout the algorithm. Finally, adding user-friendly options enabled the program to use requests more effectively and further customize the logic used within the algorithm.

Intratumoural heterogeneity complicates the molecular interpretation of biopsies, as multiple distinct tumour genomes are sampled and analysed at once. Ignoring the presence of these populations can lead to erroneous conclusions, and so a correct analysis must account for the clonal structure of the sample. Several methods to reconstruct tumour clonality from sequencing data have been proposed, spanning methods that either do not consider phylogenetic constraints or posit a perfect phylogeny. Models of the first type are typically latent feature models that can describe the observed data flexibly, but whose results may not be reconcilable with a phylogeny. The second type, instead, generally comprises non-parametric mixture models, with strict assumptions on the tumour’s evolutionary process. The focus of this dissertation is on the development of a phylogenetic latent feature model that can bridge the advantages of these two approaches, allowing deviations from a perfect phylogeny. The work is recounted by three statistical models of increasing complexity. First, I present a non-parametric model based on the Indian Buffet Process prior, and highlight the need for phylogenetic constraints. Second, I develop a finite, phylogenetic extension of the previous model, and show that it can outperform competing methods. Third, I generalise the phylogenetic model to arbitrary copy-number states. Markov chain Monte Carlo algorithms are presented to perform inference. The models are tested on datasets that include synthetic data, controlled biological data, and clinical data. In particular, the copy-number generalisation is applied to longitudinal circulating tumour DNA samples. Liquid biopsies that leverage circulating tumour DNA require sensitive techniques in order to detect mutations at low allele fractions. One method that allows sensitive mutation calling is the amplicon sequencing strategy TAm-Seq. I present bioinformatic tools to improve both the development of TAm-Seq amplicon panels and the analysis of its sequencing data. Finally, an enhancement of this method is presented and shown to detect mutations de novo and in a multiplexed manner at allele fractions less than 0.1%.

There are more neurons in the human brain than seconds in a lifetime. Given this incredible number how can we hope to understand the computations carried out by the full ensemble of neural firing patterns? And neural activity is not the only substrate available for computations. The incredible diversity of function found within biological organisms is matched by an equally rich reservoir available for computation. If we are interested in the metamorphosis of a caterpillar to a butterfly we could explore how DNA expression changes the cell. If we are interested in developing therapeutic drugs we could explore receptors and ion channels. And if we are interested in how humans and other animals interpret incoming streams of sensory information and process them to make moment-by-moment decisions then perhaps we can understand much of this behavior by studying the firing rates of neurons. This is the level and approach we will take in this thesis. Given this diversity of potential reservoirs for computation, combined with limitations in recording technologies, it can be difficult to satisfactorily conclude that we are studying the full set of neural dynamics involved in a particular task. To overcome this limitation, we augment the study of neural activity with the study of artificial recurrent neural networks (RNNs) trained to mimic the behavior of humans and other animals performing experimental tasks. The inputs to the RNN are time-varying signals representing experimental stimuli and we adjust the parameters of the RNN so its time-varying outputs are the desired behavioral responses. In these artificial RNNs we have complete information about the network connectivity and moment-by-moment firing patterns and know, by design, that these are the only computational mechanisms being used to solve the tasks. If the artificial RNN and electrode recordings of real neurons have the same dynamics we can be more confident that we are studying the sufficient set of biological dynamics involved in the task. This is important if we want to make claims about the types of dynamics required, and observed, for various computational tasks, as is the case in Chapter 2 of this thesis. In Chapter 2 we develop tests to identify several classes of neural dynamics. The specific neural dynamic regimes we focus on are interesting because they each have different computational capabilities, including, the ability to keep track of time, or preserve information robustly against the flow of time (working memory). We then apply these tests to electrode recordings from nonhuman primates and artificial RNNs to understand how neural networks are able to simultaneously keep track of time and remember previous experiences in working memory. To accomplish both computational goals the brain is thought to use distinct neural dynamics; stable neural trajectories can be used as a clock to coordinate cognitive activity whereas attractor dynamics provide a stable mechanism for memory storage but all timing information is lost. To identify these neural regimes we decode the passage of time from neural data. Additionally, to encode the passage of time, stabilized neural trajectories can be either high-dimensional as is the case for randomly connected recurrent networks (chaotic reservoir networks) or low-dimensional as is the case for artificial RNNs trained with backpropagation through time. To disambiguate these models we compute the cumulative dimensionality of the neural trajectory as it evolves over time. Recurrent neural networks can also be used to generate hypotheses about neural computation. In Chapter 3 we use RNNs to generate hypotheses about the diverse set of neural response properties seen during spatial navigation, in particular, grid cells, and other spatial correlates, including border cells and band-like cells. The approach we take is 1) pick a task that requires navigation (spatial or mental), 2) create a RNN to solve the task, and 3) adjust the task or constraints on the neural network such that grid cells and other spatial response patterns emerge naturally as the network learns to perform the task. We trained RNNs to perform navigation tasks in 2D arenas based on velocity inputs. We find that grid-like spatial response patterns emerge in trained networks, along with units that exhibit other spatial correlates, including border cells and band-like cells. Surprisingly, the order of the emergence of grid-like and border cells during network training is also consistent with observations from developmental studies. Together, our results suggest that grid cells, border cells and other spatial correlates observed in the Entorhinal Cortex of the mammalian brain may be a natural solution for representing space efficiently given the predominant recurrent connections in the neural circuits. All the tasks we have considered so far in this thesis require memory, but in Chapter 4 we explicitly explore the interactions between multiple memories in a recurrent neural network. Memory is the hallmark of recurrent neural networks, in contrast to standard feedforward neural networks where all signals travel in one direction from inputs to outputs and the network contains no memory of previous experiences. A recurrent neural network, as the name suggests, contains feedback loops giving the network the computational power of memory. In this chapter we train a RNN to perform a human psychophysics experiment and find that in order to reproduce human behavior, noise must be added to the network, causing the RNN to use more stable discrete memories to constrain less stable continuous memories.