Automated tracking of cells across timelapse microscopy image sequences typically employs complex segmentation routines and/or bio-staining of the tracking objective. Often accurate identification of a cell's morphology is not of interest and the accurate segmentation of cells in pursuit of non-morphological parameters is complex and time consuming. This thesis explores the potential of internalized quantum dot nanoparticles as alternative, bio- and photo-stable optical markers for tracking the motions of cells through time. CdTe/ZnS core-shell quantum dots act as nodes in moving light display networks within A549, epithelial, lung cancer cells over a 40 hour time period. These quantum dot fluorescence sources are identified and interpreted using simplistic algorithms to find consistent, non-subjective centroids that represent cell centre locations. The presented tracking protocols yield an approximate 91% success rate over 24 hours and 78% over the full 40 hours. The nanoparticle moving light displays also provide simultaneous collection of cell motility data, resolution of mitotic traversal dynamics and identification of familial relationships enabling the construction of multi-parameter lineage trees. This principle is then developed further through inclusion of 3 different coloured quantum dots to create cell specific colour barcodes and reduce the number of time points necessary to successfully track cells through time. The tracking software and identification of parameters without detailed morphological knowledge is also demonstrated through automated extraction of DOX accumulation profiles and Cobalt agglomeration accruement statistics from two separate toxicology assays without the need for cell segmentation.

Thesis (DTech(Biomedical Technology))--Cape Technikon, Cape Town, 2000 / The diagnosis of breast cancer is not possible using currently available serological detection of cancer markers as these lack adequate sensitivity or specificity. This study investigates the prevalence and significance of anti-p53 antibody and c-erbB-2 protein in the post-surgical sera of South African breast cancer patients and correlates these features with the clinicopathological characteristics of breast cancer. Further, this study investigates the possibilityofimproving prognostic sensitivityby combining the two subject markers to monitor each patient. Further, this study will provide the opportunity to investigate lNhether only certain types of breast cancer can elicit an immunological response and at what stage and grade of tumour antibodies are present in the postoperative serum. The study also establishes a foundation for determining in South Africa lNhether there is a genetic influence in the response to p53 mutation and INhther this response is higher in the indigenous African women compared to other South African women. The purpose of the study is to determine if the resulting findings can be used to enhance our ability to diagnose breast cancer and to identify node-negative breast cancer patients at high risk for early disease recurrence and or death, for 1Nh0m adjuvant therapy is unequivocally justified. The study accrued 92 South African breast cancer patients who were essentially women of colour 62 [67%] indigenous African women and 20 [22%] Caucasian of Indian descent, 6 [6%J of mixed [ColouredJ background and only 4 [4%J Caucasian of White descent. A predominantly indigenous African populationwas chosen becausethey are the group most likely to benefitfrom an easily repeatable, affordable serological cancer marker.

Abstract Extraintestinal manifestations in inflammatory bowel disease (IBD) have been described with varying frequencies. The aim of this study was to estimate the prevalence of pancreatic duct abnormalities, exocrine and endocrine dysfunction, elevated pancreatic enzymes, hepato-biliary disease, coexisting cholangiographic and pancreatographic duct changes, and elevated serum levels of fibrosis markers in IBD, and to correlate the findings with clinical, endoscopic and histologic variables. From a local patient register, 237 patients were randomly selected and studied. Of these, 170 had ulcerative colitis (UC), 46 had Crohn's disease (CD), and 21 had indeterminate colitis (IC). A detailed history was obtained from medical records and in a face-to-face interview. The patients were screened with a para-aminobenzoic acid (PABA) test and, for pancreatic enzymes, liver function tests, serum aminoterminal propeptide of type III procollagen (PIIINP), and laminin. Further pancreatic evaluation included endoscopic retrograde cholangiopancreato-graphy (ERCP), ultrasound (US), secretin test, and glucagon C-peptide test. Further hepatobiliary evaluation consisted of ERCP, US, and liver biopsy. In IBD, the prevalence rates of pancreatic duct abnormalities and exocrine dysfunction were 8% and 4%, respectively. Parallel impairment of exocrine and endocrine functions was shown. Acute idiopathic pancreatitis may complicate IBD. About 7-17% presented with elevated pancreatic enzymes. Enzyme elevation was associated with extensive and histologically active disease and, in some cases, with primary sclerosing cholangitis (PSC). Abnormal liver test results were commoner in patients with CD than in patients with UC (30% versus 11%). The prevalence of PSC in IBD was 11%, which is higher than previously reported (3.7-7.5%). PSC was commoner in patients with CD than in patients with UC (17.4% versus 7.6%). About half of the PSC patients had concomitant pancreatic duct changes, and the prevalence of concurrent cholangiographic and pancreatographic duct changes in IBD was 4.6%. Both serum PIIINP and laminin were increased in IBD patients. This was not only seen in patients with hepatobiliary disease and PSC, but also in patients with pancreatic disease. In conclusion, pancreatic and hepatobiliary complications in IBD occur with high and similar frequencies in all IBD categories and are associated with each other. They are not clearly associated with the clinical course of IBD.

Abstract The most threatening outcome of interstitial lung diseases is death caused by progressive pulmonary fibrosis characterised by increased collagen deposition, although the clinical course is highly variable. The aim of this study was to evaluate the role of procollagen I and III propeptides in estimating collagen metabolism and its relationship to disease activity and prognosis in patients with sarcoidosis, fibrosing alveolitis and asbestos-related lung diseases. The study included 160 patients. The levels of procollagen I carboxyterminal propeptide (PICP) and procollagen III aminoterminal propeptide (PIIINP) in serum, bronchoalveolar lavage fluid (BALF) and epithelial lining fluid (ELF) were assessed from 137 patients employing human antigens. There were 60 patients with sarcoidosis, 18 with fibrosing alveolitis and 5 with asbestosis and 17 controls. Thirty-seven patients had been exposed to asbestos, but did not show parenchymal involvement. Twenty-five of them had pleural plaques, while 12 had normal chest radiographs. Immunohistochemical stainings for procollagen I aminoterminal (PINP) and III aminoterminal propeptide were carried out on open lung biopsies of the remaining 23 of the 160 patients, of whom 13 had sarcoidosis and 10 fibrosing alveolitis. Antibodies to these procollagen peptides react with the aminoterminal domains of the corresponding propeptides intracellularly and with the respective pN-collagen in collagen fibres in the extracellular space. Procollagen III aminoterminal propeptide was elevated in the sera of the patients with sarcoidosis and fibrosing alveolitis, but not in the asbestosis or asbestos-exposed patients as compared to the controls. The level of PIIINP in BALF was highest in sarcoidosis and second highest in fibrosing alveolitis, but hardly detectable in the other groups. BALF-PICP was higher in the patients with fibrosing alveolitis, sarcoidosis and asbestosis than in the controls. PIIINP in BALF correlated with BALF-PICP, serum angiotensin-converting enzyme (S-ACE), interleukin 2-receptor, BALF-albumin and BALF-lymphocytes and BALF-PICP had a significant correlation with BALF-albumin and BALF-lymphocytes in sarcoidosis. BALF/ELF-PICP had an inverse correlation with the specific diffusion coefficient (DLCO/VA) in fibrosing alveolitis. Both PIIINP and PICP were higher in ELF than in serum in sarcoidosis and fibrosing alveolitis and PICP was higher in ELF compared to serum in asbestosis, suggesting active local synthesis in the lower respiratory tract. The levels of PIIINP in BALF were significantly elevated in sarcoidosis patients with parenchymal involvement compared to those without. Detectable PIIINP in BALF also predicted a poor outcome in fibrosing alveolitis. BALF-PIIINP reflected the disease activity based on chest radiographs in sarcoidosis and a poor prognosis in fibrosing alveolitis, whereas BALF-PICP marked the development of fibrosis. In lung biopsy specimens, type I and III pN-collagens were increased in fibrosing alveolitis and sarcoidosis. Type I pN-collagen was expressed in areas with damaged or deficient alveolar epithelium. Type III pN-collagen was present underneath regenerative, metaplastic alveolar and bronchiolar type epithelium and was accumulated both in the loose, newly formed fibrosis and in the denser old fibrosis. Type I procollagen was present in intracellular spots in newly formed fibrosis. In sarcoidosis, type I procollagen was present intracellularly in granulomas, whereas type III pN-collagen was expressed extracellularly around granulomas.

The increasing demand for recombinant therapeutic proteins coupled with advances in technologies allow research to develop approaches to improve the efficiency, yield, and quality of biopharmaceutical products from CHO cells. CHO DG44 cells used in this study were engineered to express erythropoeitin (EPO) as the model recombinant protein in a DHFR-based selection system. From a series of CHO-DG44 cell lines derived from a polyclonal population, one cell line expressed a notable change in growth phenotype during prolonged culture (10 weeks). This cell line (IJ4) exhibited prolonged growth, reached a greater density, and delayed cell death. The change in growth was reflected in an increased total yield of EPO, whilst the specific productivity of cell line IJ4 remained similar. The increased total yield of EPO presents a desirable goal for production and hence detailed ‘omics studies were performed to identify factors associated with better cell growth and survivability. Two different ‘omics analyses were performed (microarray transcriptomic and GC-MS metabolic profiling) to identify potential target genes and key metabolites associated with changes in growth profile. The -omics analyses identified a subset of genes (MMP20, PLA1A, POSTN, SLC46A3, and TOP2A), and a metabolic marker (farnesal) strongly associated with changes in cell growth and nutrient uptake. The use of complementary ‘omics approaches to identify molecular markers has allowed an integrated model to be built, which explains how CHO cell phenotype can adapt to long-term culture, and this defines molecular approaches for cell line screening and engineering.

No description available.

This study analyzes the economic impact of a recent gene marker innovation for seedstock selection in beef cattle. Gene markers are being developed for many beef cattle attributes; this study focused on the tenderness quality of beef using two categories: tender and tough. The study begins by describing conventional procedures for seedstock selection, the science which underlies selection by gene markers and other non-genetic procedures currently being used to improve beef tenderness. After describing the commercialization of the gene marker innovation, a stylized model of a beef supply chain is constructed. The supply chain consists of a representative consumer, a producer/processor group and a monopolist supplier of the patented technology. Welfare changes resulting from the adoption of the innovation were simulated using four sets of demand elasticity data from literatures. An important focus of this research is determining how the economic surplus from the innovation will be shared by consumers, producers and the gene marker monopolist. The consumer and gene marker monopolist benefit from the technology unless the marginal and fixed cost variables (not estimated in this study) of the monopolist, are excessively high. Producer surplus was simulated as positive with three of the four elasticity data sets. The share of surplus capture by producers is generally low relative to the gains captured by consumers and the gene marker monopolist. Comparative static analysis reveal that the benefit from the innovation varies across breeds, being higher for breeds in which the favorable form of the marker gene is more likely to be present. Despite the apparent benefits of the innovation for beef supply chain participants, reported interviews with industry scientists reveal that markers should not be viewed as a replacement for conventional selection techniques. Indeed, selecting seedstock on the basis of a small number of available markers is not likely to produce the benefits that are currently being promised by life science companies. Consequently, this study recommends that the innovation be incorporated into existing seedstock selection practices. Much more analysis is needed to understand the full economic impact of gene markers for beef tenderness and for other beef quality attributes. / Land and Food Systems, Faculty of / Graduate

M.Sc. / The health of aquatic ecosystems can be severely compromised by a variety of pollutants, such as heavy metals, which are related to anthropogenic activities. Increased recognition is given to the use of organisms, especially molluscs, in the biomonitoring of aquatic ecosystems. This promising approach complements the interpretation of the physico-chemical measurements of water quality. The bioaccumulation of pollutants as well as the resultant biological responses (biomarkers) in organisms can be used in assessing the spatial and temporal trends of chronically polluted environments. The aim of this study was to determine if selected freshwater gastropods (Melanoides tuberculata, Physa acuta, Helisoma duryi and Lymnaea columella) can be used as suitable biomonitors or indicators of water quality. Interspecies differences in metal bioaccumulation and biomarker responses were determined in order to select the most suitable biomonitor/indicator organism to be used. The bioaccumulation of metals was related to the biomarker responses of the organisms. This study was divided into an active biomonitoring (ABM) study and a laboratory exposure study. In the first study, the organisms, M. tuberculata and L. columella were chronically (two, four and six week period) exposed to field conditions. Metal bioaccumulation as well as the biomarker techniques, DNA damage, catalase (CAT) activity, reduced glutathione (GSH) content and cellular energy allocation (CEA) were measured in the organisms. These general biomarkers of exposure and effect, on the biochemical and cellular levels of biological organisation can give a rapid and sensitive assessment of organism health. The second study consisted of exposing the gastropods, M. tuberculata, P. acuta, H. duryi and L. columella to sub-lethal zinc and cadmium concentrations. The uptake and depuration kinetics of these metals were determined in M. tuberculata and H. duryi for a six hour and 48 hour period, respectively. The bioaccumulation of Zn and Cd as well as the biomarkers, DNA damage, CAT activity, GSH content and CEA were measured in all the species, after a two week exposure period. / Prof. J.H.J. van Vuren

The theme of this dissertation is to construct heterogeneous biomarker networks using graphical models for understanding disease progression and prognosis. Biomarkers may organize into networks of connected regions. Substantial heterogeneity in networks between individuals and subgroups of individuals is observed. The strengths of network connections may vary across subjects depending on subject-specific covariates (e.g., genetic variants, age). In addition, the connectivities between biomarkers, as subject-specific network features, have been found to predict disease clinical outcomes. Thus, it is important to accurately identify biomarker network structure and estimate the strength of connections. Graphical models have been extensively used to construct complex networks. However, the estimated networks are at the population level, not accounting for subjects’ covariates. More flexible covariate-dependent graphical models are needed to capture the heterogeneity in subjects and further create new network features to improve prediction of disease clinical outcomes and stratify subjects into clinically meaningful groups. A large number of parameters are required in covariate-dependent graphical models. Regularization needs to be imposed to handle the high-dimensional parameter space. Furthermore, personalized clinical symptom networks can be constructed to investigate co-occurrence of clinical symptoms. When there are multiple biomarker modalities, the estimation of a target biomarker network can be improved by incorporating prior network information from the external modality. This dissertation contains four parts to achieve these goals: (1) An efficient l0-norm feature selection method based on augmented and penalized minimization to tackle the high-dimensional parameter space involved in covariate-dependent graphical models; (2) A two-stage approach to identify disease-associated biomarker network features; (3) An application to construct personalized symptom networks; (4) A node-wise biomarker graphical model to leverage the shared mechanism between multi-modality data when external modality data is available. In the first part of the dissertation, we propose a two-stage procedure to regularize l0-norm as close as possible and solve it by a highly efficient and simple computational algorithm. Advances in high-throughput technologies in genomics and imaging yield unprecedentedly large numbers of prognostic biomarkers. To accommodate the scale of biomarkers and study their association with disease outcomes, penalized regression is often used to identify important biomarkers. The ideal variable selection procedure would search for the best subset of predictors, which is equivalent to imposing an l0-penalty on the regression coefficients. Since this optimization is a non-deterministic polynomial-time hard (NP-hard) problem that does not scale with number of biomarkers, alternative methods mostly place smooth penalties on the regression parameters, which lead to computationally feasible optimization problems. However, empirical studies and theoretical analyses show that convex approximation of l0-norm (e.g., l1) does not outperform their l0 counterpart. The progress for l0-norm feature selection is relatively slower, where the main methods are greedy algorithms such as stepwise regression or orthogonal matching pursuit. Penalized regression based on regularizing l0-norm remains much less explored in the literature. In this work, inspired by the recently popular augmenting and data splitting algorithms including alternating direction method of multipliers, we propose a two-stage procedure for l0-penalty variable selection, referred to as augmented penalized minimization-L0 (APM-L0). APM-L0 targets l0-norm as closely as possible while keeping computation tractable, efficient, and simple, which is achieved by iterating between a convex regularized regression and a simple hard-thresholding estimation. The procedure can be viewed as arising from regularized optimization with truncated l1 norm. Thus, we propose to treat regularization parameter and thresholding parameter as tuning parameters and select based on cross-validation. A one-step coordinate descent algorithm is used in the first stage to significantly improve computational efficiency. Through extensive simulation studies and real data application, we demonstrate superior performance of the proposed method in terms of selection accuracy and computational speed as compared to existing methods. The proposed APM-L0 procedure is implemented in the R-package APML0. In the second part of the dissertation, we develop a two-stage method to estimate biomarker networks that account for heterogeneity among subjects and evaluate the network’s association with disease clinical outcome. In the first stage, we propose a conditional Gaussian graphical model with mean and precision matrix depending on covariates to obtain subject- or subgroup-specific networks. In the second stage, we evaluate the clinical utility of network measures (connection strengths) estimated from the first stage. The second stage analysis provides the relative predictive power of between-region network measures on clinical impairment in the context of regional biomarkers and existing disease risk factors. We assess the performance of the proposed method by extensive simulation studies and application to a Huntington’s disease (HD) study to investigate the effect of HD causal gene on the rate of change in motor symptom through affecting brain subcortical and cortical grey matter atrophy connections. We show that cortical network connections and subcortical volumes, but not subcortical connections are identified to be predictive of clinical motor function deterioration. We validate these findings in an independent HD study. Lastly, highly similar patterns seen in the grey matter connections and a previous white matter connectivity study suggest a shared biological mechanism for HD and support the hypothesis that white matter loss is a direct result of neuronal loss as opposed to the loss of myelin or dysmyelination. In the third part of the dissertation, we apply the methodology to construct heterogeneous cross-sectional symptom networks. The co-occurrence of symptoms may result from the direct interactions between these symptoms and the symptoms can be treated as a system. In addition, subject-specific risk factors (e.g., genetic variants, age) can also exert external influence on the system. In this work, we develop a covariate-dependent conditional Gaussian graphical model to obtain personalized symptom networks. The strengths of network connections are modeled as a function of covariates to capture the heterogeneity among individuals and subgroups of individuals. We assess the performance of the proposed method by simulation studies and an application to a Huntington’s disease study to investigate the networks of symptoms in different domains (motor, cognitive, psychiatric) and identify the important brain imaging biomarkers associated with the connections. We show that the symptoms in the same domain interact more often with each other than across domains. We validate the findings using subjects’ measurements from follow-up visits. In the fourth part of the dissertation, we propose an integrative learning approach to improve the estimation of subject-specific networks of target modality when external modality data is available. The biomarker networks measured by different modalities of data (e.g., structural magnetic resonance imaging (sMRI), diffusion tensor imaging (DTI)) may share the same true underlying biological mechanism. In this work, we propose a node-wise biomarker graphical model to leverage the shared mechanism between multi-modality data to provide a more reliable estimation of the target modality network and account for the heterogeneity in networks due to differences between subjects and networks of external modality. Latent variables are introduced to represent the shared unobserved biological network and the information from the external modality is incorporated to model the distribution of the underlying biological network. An approximation approach is used to calculate the posterior expectations of latent variables to reduce time. The performance of the proposed method is demonstrated by extensive simulation studies and an application to construct gray matter brain atrophy network of Huntington’s disease by using sMRI data and DTI data. The estimated network measures are shown to be meaningful for predicting follow-up clinical outcomes in terms of patient stratification and prediction. Lastly, we conclude the dissertation with comments on limitations and extensions.

Miniaturization, integration and intelligence are the developing trends for sensor,especially for biosensors. The development of microelectronics technology is a powerful engine to full this objective. It is well known that the microelectronic fabrication process in proven technology for fabrication of integrated circuits. Advances in the field of micro-electronics and micro-mechanical devices combined with medical science have led to the development of numerous analytical devices in monitoring of a wide range of analytes. The unique properties of nanoscale materials offer excellent prospects for interfacing biological recognition events with electronic signal transduction and for designing a new generation of bio-electronic devices exhibiting novel functions. Biosensor development has the potential to meet the need for rapid, sensitive, and specic detection of pathogenic bacteria from natural sources. This work focuses on development of one such electrochemical biosensor platform and discusses dierent aspects related to the design of biosensor and biodetection systems. A new transducer for bio sensor applications based on 3-dimensional, comb structured interdigitated electrode arrays was chosen mainly for two reasons. Firstly, this geometry allows the monitoring of both resistivity and dielectric constant of solution, thus making interdigitated electrodes more versatile tools than other kind of transducers. Second, they present short electric eld penetration depths, which make them more sensitive to changes occurring close to their surface (20 - 100 nm above the surface). This fact enables the monitoring of local changes in the vicinity of interest. Binding of analyte molecules to the chemically modied transducer surface induces important changes in the conductivity between the electrodes. Interdigitated electrodes have been employed to detect the presence of Anti-Transglutaminase (TG) antibodies, that are established biomarkers for Celiac disease which is due to gluten allergy. The biosensor was optimized for specific and sensitive detection of this biomarker. The sensor showed a sensitivity down to picomolar(pM) concentration of the biomarker. Gold nanoparticles were further used for signal enhancement so as to bring the sensor performance closer to Enzyme linked immunosorbant assay (ELISA).