Made available in DSpace on 2014-10-09T12:43:30Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:09:43Z (GMT). No. of bitstreams: 1 06636.pdf: 4715244 bytes, checksum: 8df295514eae9493e07c3d6352c4bf32 (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

This multidisciplinary thesis is concerned with the prediction of drug formulations for the reduction of drug toxicity. Both scientific and computational approaches are utilised to make original contributions to the field of predictive toxicology. The first part of this thesis provides a detailed scientific discussion on all aspects of drug formulation and toxicity. Discussions are focused around the principal mechanisms of drug toxicity and how drug toxicity is studied and reported in the literature. Furthermore, a review of the current technologies available for formulating drugs for toxicity reduction is provided. Examples of studies reported in the literature that have used these technologies to reduce drug toxicity are also reported. The thesis also provides an overview of the computational approaches currently employed in the field of in silico predictive toxicology. This overview focuses on the machine learning approaches used to build predictive QSAR classification models, with examples discovered from the literature provided. Two methodologies have been developed as part of the main work of this thesis. The first is focused on use of directed bipartite graphs and Venn diagrams for the visualisation and extraction of drug-vehicle relationships from large un-curated datasets which show changes in the patterns of toxicity. These relationships can be rapidly extracted and visualised using the methodology proposed in chapter 4. The second methodology proposed, involves mining large datasets for the extraction of drug-vehicle toxicity data. The methodology uses an area-under-the-curve principle to make pairwise comparisons of vehicles which are classified according to the toxicity protection they offer, from which predictive classification models based on random forests and decisions trees are built. The results of this methodology are reported in chapter 6.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22359号 / 医博第4600号 / 新制||医||1042(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 長尾 美紀, 教授 滝田 順子, 教授 河本 宏 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

The thesis deals with the comparison of conventional and heuristic methods in data mining used for binary classification. In the theoretical part, four different models are described. Model classification is demonstrated on simple examples. In the practical part, models are compared on real data. This part also consists of data cleaning, outliers removal, two different transformations and dimension reduction. In the last part methods used to quality testing of models are described.

Ertapenem is an antibiotic commonly used to treat a broad spectrum of infections and is part of a broader class of antibiotics called carbapenems. Unlike other carbapenems, ertapenem has a longer half-life and thus only has to be administered once a day. Previously, a physiologically-based pharmacokinetic (PBPK) model was developed to investigate the uptake, distribution, and elimination of ertapenem following a single one gram dose in normal height, normal weight males. Due to the absorption properties of ertapenem, the amount of fat in the body can influence how the drug binds, how quickly the drug passes through the body, and thus how effective the drug might be. Thus, we have revised the model so that it is applicable to males and females of differing body mass index (BMI). Simulations were performed to consider the distribution of the antibiotic in males and females with varying body mass indexes. These results could help to determine if there is a need for altered dosing regimens in the future.

No description available.

The goal of this theses is to compare existing bankruptcy prediction models with its new modification unique for this work, which could perform better than its competition. Proposed model is logit-based and consists of the combination of variables used in Altman´s and Ohlson´s models. The final model is estimated for medium sized companies in EU which aren´t publicly traded. This model achieved prediction accuracy of 97,1% (97.4% for healthy and 91.1% for bankrupt compa-nies) on its original dataset. As expected, when verified on new dataset, the accu-racy dropped but still reaches 97.1% (99.3% for healthy and 37.7% for bankrupt companies). The model is compared with its competition (original and modified version of Ohlson´s and partially Altman´s models) and it is shown that it has higher prediction accuracy.

Since the last decade, an increasing number of proteins have been shown to be capable of undergoing reversible liquid-liquid phase separation (LLPS) in response to an external stimulus, and the resulting protein-rich phase (coacervate) is considered as one of the main components of membrane-less organelles. Most of these proteins are intrinsically disordered proteins (IDPs) or contain intrinsically disordered regions. More importantly, LLPS often plays an important role in cellular signaling and development of cells and tissues. However, the molecular mechanisms underlying LLPS of proteins remain poorly understood. Elastin-like proteins (ELPs), a class of IDPs derived from the hydrophobic domains of tropoelastin, are known to undergo LLPS reversibly above a concentration-dependent transition temperature (TT), allowing ELPs to be a promising thermo-responsive drug delivery vector for treating cancer. Previous studies have suggested that, as temperature increases, ELPs experience an increased propensity for type II beta-turns. Our hypothesis is that the interaction is initiated at the beta-turn positions. In this work, integrative approaches including experimental and computational methods were employed to study the early stages of ELP phase separation. Using nuclear magnetic resonance spectroscopy (NMR), and paramagnetic relaxation enhancement (PRE), we have characterized structural properties of self-association in several ELPs. NMR chemical shifts suggest that ELPs adopt a beta-turn conformation even at temperatures below the TT. The intermolecular PRE reveals there is a stronger interaction between the higher beta-turn propensity regions. Building on this observation, a series of structural ensembles were generated for ELP incorporating differing amounts of beta-turn bias, from 1% to 90%. To mimic the early stages of the phase change, two monomers were paired, assuming preferential interaction at beta-turn regions. Following dimerization, the ensemble-averaged hydrodynamic properties were calculated for each degree of beta-turn bias, and results were compared with analytical ultracentrifugation (AUC) experiments at various temperatures. The ensemble calculation reveals that accessible surface area changes dramatically as oligomers are formed from monomers with a high beta-turn content. Together, these observations suggest a model where ELP self-association is initiated at beta-turn positions, where the driving force of phase separation is solvent exclusion due to changes in the hydrophobic accessible surface area.

This multidisciplinary thesis is concerned with the prediction of drug formulations for the reduction of drug toxicity. Both scientific and computational approaches are utilised to make original contributions to the field of predictive toxicology. The first part of this thesis provides a detailed scientific discussion on all aspects of drug formulation and toxicity. Discussions are focused around the principal mechanisms of drug toxicity and how drug toxicity is studied and reported in the literature. Furthermore, a review of the current technologies available for formulating drugs for toxicity reduction is provided. Examples of studies reported in the literature that have used these technologies to reduce drug toxicity are also reported. The thesis also provides an overview of the computational approaches currently employed in the field of in silico predictive toxicology. This overview focuses on the machine learning approaches used to build predictive QSAR classification models, with examples discovered from the literature provided. Two methodologies have been developed as part of the main work of this thesis. The first is focused on use of directed bipartite graphs and Venn diagrams for the visualisation and extraction of drug-vehicle relationships from large un-curated datasets which show changes in the patterns of toxicity. These relationships can be rapidly extracted and visualised using the methodology proposed in chapter 4. The second methodology proposed, involves mining large datasets for the extraction of drug-vehicle toxicity data. The methodology uses an area-under-the-curve principle to make pairwise comparisons of vehicles which are classified according to the toxicity protection they offer, from which predictive classification models based on random forests and decisions trees are built. The results of this methodology are reported in chapter 6.

No description available.