Treating cancer is difficult as the disease is complex and drug responses often depend on the patient's characteristics. Precision medicine aims to solve this by selecting individualized treatments. Since this involves the analysis of large datasets, machine learning can be used to make the drug selection process more efficient. Traditionally, such models utilize bulk gene expression data. However, this potentially masks information from small cell populations and fails to address tumor heterogeneity. Therefore, this thesis applies data deconvolution methods to bulk gene expression data and estimates the corresponding cell type-specific gene expression profiles. This "increases" the resolution of the input data for the drug response prediction. A hold-out dataset, LODOCV and LOCOCV were used for the evaluation of this approach. Furthermore, all results are compared against a baseline model, which was trained on bulk data. Overall, the accuracy of the cell type-specific model did not show an improvement compared to the bulk model. It also prioritizes information from bulk samples, which makes the additional data unnecessary. The robustness of the cell type-specific model is slightly lower than that of the bulk model. Note, that these outcomes are not necessarily due to a flaw in the underlying concept, but may be connected to poor deconvolution results as the same reference matrix was used for the deconvolution of all bulk samples regardless of the cancer type or disease.

Obsessive-compulsive disorder (OCD), with a prevalence of 1-2 %, frequently leads a chronic course. Persons with OCD are often reluctant to seek help and, if they do, their OCD is often missed. This is unfortunate, since active treatment may substantially improve social function and quality of life. Serotonin reuptake inhibitors (SRIs) have welldocumented efficacy in OCD, but delayed response may be problematic. Methods to predict response have been lacking. Because SRIs are effective, pathophysiological research on OCD has focussed on serotonin. However, no clear aberrations of serotonin have been found, thus other mechanisms ought to be involved. Our aims were to facilitate clinical detection and assessment of OCD, to search for biochemical correlates of response and side-effects in SRI treatment of OCD and to identify any possible involvement of oxytocin in the pathophysiology of OCD. In study I, we tested in 402 psychiatric out-patients the psychometric properties of a concise rating scale, “Brief Obsessive Compulsive Scale” (BOCS). BOCS was shown to be easy to use and have excellent discriminant validity in relation to other common psychiatric diagnoses. Studies II-V were based on 36 OCD patients from a randomised controlled trial of paroxetine, clomipramine or placebo. In study II, contrary to expectation, we found that the change (decrease) of serotonin in whole blood was most pronounced in non-responders to SRI. This is likely to reflect inflammatory influence on platelet turnover rather than serotonergic processes within the central nervous system. In studies IV-V, we found relations between changes of oxytocin in plasma and the anti-obsessive response, and between oxytocin and the SRI related delay of orgasm, respectively. In both cases, the relation to central oxytocinergic mechanisms is unclear. In males, delayed orgasm predicted anti-obsessive response.

Preface - Biotechnology is generally considered to be the wave of the future. To facilitate accurate and rapid development of medication and treatments, it is critical that we are able to simulate the human body. One section of this complex model would be the human energy system. Pharmaceutical companies are currently pouring vast amounts of capital into research regarding general simulation of cellular structures, protein structures and bodily processes. Their aim is to develop treatments and medication for major diseases. Some of these diseases are epidemics like cancer, cardiovascular diseases, stress, obesity, etc. One of the most important causes of these diseases is poor blood glucose control. Current management methods for insulin dependent diabetes are limited to trial and error systems: clearly ineffective and prone to errors. It is critical that better management systems be developed, to ease the diabetic epidemic. The blood glucose control system is one of the major systems in the body, as we are in constant need of energy to facilitate the optimum functioning of the human body. This study makes use of a developed simulation model for the human energy system to ease the management of Diabetes mellitus, which is a malfunction of the human energy system. This dissertation is presented in two parts: The first part discusses the human energy simulation model, and the verification thereof, while the second presents possible applications of this model to ease the management of Diabetes. The human energy system simulation model - This section discusses the development and verification of the model. It also touches on the causes, and current methods, of managing diabetes, as well as the functioning of the human energy system. The human energy model is approached with the conservation of energy in mind. A top down model is developed, using data from independent studies to verify the model. Application of human energy simulation model - The human energy simulation model is of little use if the intended audience cannot use it: people suffering from malfunctioning energy systems. These include people having trouble with obesity, diabetes, cardiovascular disease, etc. To facilitate this, we need to provide a variety of products useable by this group of people. We propose a variety of ways in which the model can be used: Cellular phone applications, Personal digital assistants (PDAs) applications, as well as computer software. By making use of current technology, we generate a basic proof-of-concept application to demonstrate the intended functionality. / MIng (Mechanical Engineering) North-West University, Potchefstroom Campus, 2004

Preface - Biotechnology is generally considered to be the wave of the future. To facilitate accurate and rapid development of medication and treatments, it is critical that we are able to simulate the human body. One section of this complex model would be the human energy system. Pharmaceutical companies are currently pouring vast amounts of capital into research regarding general simulation of cellular structures, protein structures and bodily processes. Their aim is to develop treatments and medication for major diseases. Some of these diseases are epidemics like cancer, cardiovascular diseases, stress, obesity, etc. One of the most important causes of these diseases is poor blood glucose control. Current management methods for insulin dependent diabetes are limited to trial and error systems: clearly ineffective and prone to errors. It is critical that better management systems be developed, to ease the diabetic epidemic. The blood glucose control system is one of the major systems in the body, as we are in constant need of energy to facilitate the optimum functioning of the human body. This study makes use of a developed simulation model for the human energy system to ease the management of Diabetes mellitus, which is a malfunction of the human energy system. This dissertation is presented in two parts: The first part discusses the human energy simulation model, and the verification thereof, while the second presents possible applications of this model to ease the management of Diabetes. The human energy system simulation model - This section discusses the development and verification of the model. It also touches on the causes, and current methods, of managing diabetes, as well as the functioning of the human energy system. The human energy model is approached with the conservation of energy in mind. A top down model is developed, using data from independent studies to verify the model. Application of human energy simulation model - The human energy simulation model is of little use if the intended audience cannot use it: people suffering from malfunctioning energy systems. These include people having trouble with obesity, diabetes, cardiovascular disease, etc. To facilitate this, we need to provide a variety of products useable by this group of people. We propose a variety of ways in which the model can be used: Cellular phone applications, Personal digital assistants (PDAs) applications, as well as computer software. By making use of current technology, we generate a basic proof-of-concept application to demonstrate the intended functionality. / MIng (Mechanical Engineering) North-West University, Potchefstroom Campus, 2004

No description available.

Preface - Biotechnology is generally accepted to be the next economical wave of the future. In order to attain the many benefits associated with this growing industry simulation modelling techniques have to be implemented successfully. One of the simulations that ne' ed to be performed is that of the human energy system. Pharmaceutical companies are currently pouring vast amounts of capital into research regarding simulation of bodily processes. Their aim is to develop cures, treatments, medication, etc. for major diseases. These diseases include epidemics like diabetes, cancer, cardiovascular diseases, obesity, stress, hypertension, etc. One of the most important driving forces behind these diseases is poor blood sugar control. The blood glucose system is one of the major subsystems of the complete human energy system. In this study a simulation model and procedure for simulating blood glucose response due to various external influences on the human body is presented. The study is presented in two parts. The first is the development of a novel concept for quantifying glucose energy flow into, within and out of the human energy system. The new quantification unit is called ets (equivalent teaspoons sugar). The second part of the study is the implementation of the ets concept in order to develop the simulation model. Development of the ets concept - In the first part of the study the ets concept, used for predicting glycaemic response, is developed and presented. The two current methods for predicting glycaemic response due to ingestion of food are discussed, namely carbohydrate counting and the glycaemic index. Furthermore, it is shown that it is currently incorrectly assumed that 100% of the chemical energy contained in food is available to the human energy system after consumption. The ets concept is derived to provide a better measure of available energy from food. In order to verify the ets concept, two links with ets are investigated. These are the links with insulin response prediction as well as with endurance energy expenditure. It is shown that with both these links linear relationships provide a good approximation of empirical data. It is also shown that individualised characterisation of different people is only dependent on a single measurable variable for each link. Lastly, two novel applications of the ets concept are considered. The first is a new method to use the ets values associated with food and energy expenditure in order to calculate both short-acting and long-acting insulin dosages for Type 1 diabetics. The second application entails a new quantification method for describing the effects of stress and illness in terms of ets. Development of the blood glucose simulation model - The second part of the study presents a literature study regarding human physiology, the development for the blood glucose simulation model as well as a verification study of the simulation model. Firstly, a brief overview is given for the need and motivation behind simulation is given. A discussion on the implementation of the techniques for construction of the model is also shown. The procedure for solving the model is then outlined. During the literature study regarding human physiology two detailed schematic layouts are presented and discussed. The first layout involves the complex flow pathways of energy through the human energy system. The second layout presents a detailed discussion on the control system involved with the glucose energy pathway. Following the literature review the model for predicting glycaemic response is proposed. The design of the component models used for the simulations of the internal processes are developed in detail as well as the control strategies implemented for the control system of the simulation model. Lastly, the simulation model is applied for glycaemic response prediction of actual test subjects and the quality of the predictions are evaluated. The verification of the model and the procedure is performed by comparing simulated results to measured data. Two evaluations were considered, namely long-term and short-term trials. The quality of both are determined according to certain evaluation criteria and it is found that the model is more than 70% accurate for long-term simulations and more than 80% accurate for short-term simulations. Conclusion - In conclusion, it is shown that simplified simulation of the human energy system is not only possible but also relatively accurate. However, in order to accomplish the simulations a simple quantification method is required and this is provided by the ets concept developed in the first part of this study. Some recommendations are also made for future research regarding both the ets concept and the simulation model. Finally, as an initial endeavour the simulation model and the ets concept proposed in this study may provide the necessary edge for groundbreaking biotechnological discoveries. / PhD (Mechanical Engineering) North-West University, Potchefstroom Campus, 2003

Preface - Biotechnology is generally accepted to be the next economical wave of the future. In order to attain the many benefits associated with this growing industry simulation modelling techniques have to be implemented successfully. One of the simulations that ne' ed to be performed is that of the human energy system. Pharmaceutical companies are currently pouring vast amounts of capital into research regarding simulation of bodily processes. Their aim is to develop cures, treatments, medication, etc. for major diseases. These diseases include epidemics like diabetes, cancer, cardiovascular diseases, obesity, stress, hypertension, etc. One of the most important driving forces behind these diseases is poor blood sugar control. The blood glucose system is one of the major subsystems of the complete human energy system. In this study a simulation model and procedure for simulating blood glucose response due to various external influences on the human body is presented. The study is presented in two parts. The first is the development of a novel concept for quantifying glucose energy flow into, within and out of the human energy system. The new quantification unit is called ets (equivalent teaspoons sugar). The second part of the study is the implementation of the ets concept in order to develop the simulation model. Development of the ets concept - In the first part of the study the ets concept, used for predicting glycaemic response, is developed and presented. The two current methods for predicting glycaemic response due to ingestion of food are discussed, namely carbohydrate counting and the glycaemic index. Furthermore, it is shown that it is currently incorrectly assumed that 100% of the chemical energy contained in food is available to the human energy system after consumption. The ets concept is derived to provide a better measure of available energy from food. In order to verify the ets concept, two links with ets are investigated. These are the links with insulin response prediction as well as with endurance energy expenditure. It is shown that with both these links linear relationships provide a good approximation of empirical data. It is also shown that individualised characterisation of different people is only dependent on a single measurable variable for each link. Lastly, two novel applications of the ets concept are considered. The first is a new method to use the ets values associated with food and energy expenditure in order to calculate both short-acting and long-acting insulin dosages for Type 1 diabetics. The second application entails a new quantification method for describing the effects of stress and illness in terms of ets. Development of the blood glucose simulation model - The second part of the study presents a literature study regarding human physiology, the development for the blood glucose simulation model as well as a verification study of the simulation model. Firstly, a brief overview is given for the need and motivation behind simulation is given. A discussion on the implementation of the techniques for construction of the model is also shown. The procedure for solving the model is then outlined. During the literature study regarding human physiology two detailed schematic layouts are presented and discussed. The first layout involves the complex flow pathways of energy through the human energy system. The second layout presents a detailed discussion on the control system involved with the glucose energy pathway. Following the literature review the model for predicting glycaemic response is proposed. The design of the component models used for the simulations of the internal processes are developed in detail as well as the control strategies implemented for the control system of the simulation model. Lastly, the simulation model is applied for glycaemic response prediction of actual test subjects and the quality of the predictions are evaluated. The verification of the model and the procedure is performed by comparing simulated results to measured data. Two evaluations were considered, namely long-term and short-term trials. The quality of both are determined according to certain evaluation criteria and it is found that the model is more than 70% accurate for long-term simulations and more than 80% accurate for short-term simulations. Conclusion - In conclusion, it is shown that simplified simulation of the human energy system is not only possible but also relatively accurate. However, in order to accomplish the simulations a simple quantification method is required and this is provided by the ets concept developed in the first part of this study. Some recommendations are also made for future research regarding both the ets concept and the simulation model. Finally, as an initial endeavour the simulation model and the ets concept proposed in this study may provide the necessary edge for groundbreaking biotechnological discoveries. / PhD (Mechanical Engineering) North-West University, Potchefstroom Campus, 2003

[pt] Prever com precisão a resposta a drogas para uma dada amostra baseado em características moleculares pode ajudar a otimizar o desenvolvimento de drogas e explicar mecanismos por trás das respostas aos tratamentos. Nessa dissertação, dois estudos de caso foram gerados, cada um aplicando diferentes dados genômicos para a previsão de resposta a drogas. O estudo de caso 1 avaliou dados de perfis de metilação de DNA como um tipo de característica molecular que se sabe ser responsável por causar tumorigênese e modular a resposta a tratamentos. Usando perfis de metilação de 987 linhagens celulares do genoma completo na base de dados Genomics of Drug Sensitivity in Cancer (GDSC), utilizamos algoritmos de aprendizado de máquina para avaliar o potencial preditivo de respostas citotóxicas para oito drogas contra o câncer. Nós comparamos a performance de cinco algoritmos de classificação e quatro algoritmos de regressão representando metodologias diversas, incluindo abordagens tree-, probability-, kernel-, ensemble- e distance-based. Aplicando sub-amostragem artificial em graus variados, essa pesquisa procura avaliar se o treinamento baseado em resultados relativamente extremos geraria melhoria no desempenho. Ao utilizar algoritmos de classificação e de regressão para prever respostas discretas ou contínuas, respectivamente, nós observamos consistentemente excelente desempenho na predição quando os conjuntos de treinamento e teste consistiam em dados de linhagens celulares. Algoritmos de classificação apresentaram melhor desempenho quando nós treinamos os modelos utilizando linhagens celulares com valores de resposta a drogas relativamente extremos, obtendo valores de area-under-the-receiver-operating-characteristic-curve de até 0,97. Os algoritmos de regressão tiveram melhor desempenho quando treinamos os modelos utilizado o intervalo completo de valores de resposta às drogas, apesar da dependência das métricas de desempenho utilizadas. O estudo de caso 2 avaliou dados de RNA-seq, dados estes comumente utilizados no estudo da eficácia de drogas. Aplicando uma abordagem de aprendizado semi-supervisionado, essa pesquisa busca avaliar o impacto da combinação de dados rotulados e não-rotulados para melhorar a predição do modelo. Usando dados rotulados de RNA-seq do genoma completo de uma média de 125 amostras de tumor AML rotuladas da base de dados Beat AML (separados por tipos de droga) e 151 amostras de tumor AML não-rotuladas na base de dados The Cancer Genome Atlas (TCGA), utilizamos uma estrutura de modelo semi-supervisionado para prever respostas citotóxicas para quatro drogas contra câncer. Modelos semi-supervisionados foram gerados, avaliando várias combinações de parâmetros e foram comparados com os algoritmos supervisionados de classificação. / [en] Accurately predicting drug responses for a given sample based on molecular features may help to optimize drug-development pipelines and explain mechanisms behind treatment responses. In this dissertation, two case studies were generated, each applying different genomic data to predict drug response. Case study 1 evaluated DNA methylation profile data as one type of molecular feature that is known to drive tumorigenesis and modulate treatment responses. Using genome-wide, DNA methylation profiles from 987 cell lines in the Genomics of Drug Sensitivity in Cancer (GDSC) database, we used machine-learning algorithms to evaluate the potential to predict cytotoxic responses for eight anti-cancer drugs. We compared the performance of five classification algorithms and four regression algorithms representing diverse methodologies, including tree-, probability-, kernel-, ensemble- and distance-based approaches. By applying artificial subsampling in varying degrees, this research aims to understand whether training based on relatively extreme outcomes would yield improved performance. When using classification or regression algorithms to predict discrete or continuous responses, respectively, we consistently observed excellent predictive performance when the training and test sets consisted of cell-line data. Classification algorithms performed best when we trained the models using cell lines with relatively extreme drug-response values, attaining area-under-the-receiver-operating-characteristic-curve values as high as 0.97. The regression algorithms performed best when we trained the models using the full range of drug-response values, although this depended on the performance metrics we used. Case study 2 evaluated RNA-seq data as one of the most popular molecular data used to study drug efficacy. By applying a semi-supervised learning approach, this research aimed to understand the impact of combining labeled and unlabeled data to improve model prediction. Using genome-wide RNA-seq labeled data from an average of 125 AML tumor samples in the Beat AML database (varying by drug type) and 151 unlabeled AML tumor samples in The Cancer Genome Atlas (TCGA) database, we used a semi-supervised model structure to predict cytotoxic responses for four anti-cancer drugs. Semi-supervised models were generated, while assessing several parameter combinations and were compared against supervised classification algorithms.