Chancellery Research - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 24
-
ItemNo Preview AvailableSparSNP: Fast and memory-efficient analysis of all SNPs for phenotype prediction(BMC, 2012-05-10)BACKGROUND: A central goal of genomics is to predict phenotypic variation from genetic variation. Fitting predictive models to genome-wide and whole genome single nucleotide polymorphism (SNP) profiles allows us to estimate the predictive power of the SNPs and potentially develop diagnostic models for disease. However, many current datasets cannot be analysed with standard tools due to their large size. RESULTS: We introduce SparSNP, a tool for fitting lasso linear models for massive SNP datasets quickly and with very low memory requirements. In analysis on a large celiac disease case/control dataset, we show that SparSNP runs substantially faster than four other state-of-the-art tools for fitting large scale penalised models. SparSNP was one of only two tools that could successfully fit models to the entire celiac disease dataset, and it did so with superior performance. Compared with the other tools, the models generated by SparSNP had better than or equal to predictive performance in cross-validation. CONCLUSIONS: Genomic datasets are rapidly increasing in size, rendering existing approaches to model fitting impractical due to their prohibitive time or memory requirements. This study shows that SparSNP is an essential addition to the genomic analysis toolkit.SparSNP is available at http://www.genomics.csse.unimelb.edu.au/SparSNP.
-
ItemRobust Observation Detection for Single Object Tracking: Deterministic and Probabilistic Patch-Based Approaches(MDPI, 2012-11)In video analytics, robust observation detection is very important as the content of the videos varies a lot, especially for tracking implementation. Contrary to the image processing field, the problems of blurring, moderate deformation, low illumination surroundings, illumination change and homogenous texture are normally encountered in video analytics. Patch-Based Observation Detection (PBOD) is developed to improve detection robustness to complex scenes by fusing both feature- and template-based recognition methods. While we believe that feature-based detectors are more distinctive,however, for finding the matching between the frames are best achieved by a collection of points as in template-based detectors. Two methods of PBOD-the deterministic and probabilistic approaches-have been tested to find the best mode of detection. Both algorithms start by building comparison vectors at each detected points of interest. The vectors are matched to build candidate patches based on their respective coordination. For the deterministic method, patch matching is done in 2-level test where threshold-based position and size smoothing are applied to the patch with the highest correlation value. Forthe second approach, patch matching is done probabilistically by modelling the histograms of the patches by Poisson distributions for both RGB and HSV colour models. Then,maximum likelihood is applied for position smoothing while a Bayesian approach is appliedfor size smoothing. The result showed that probabilistic PBOD outperforms the deterministic approach with average distance error of 10.03% compared with 21.03%. This algorithm is best implemented as a complement to other simpler detection methods due to heavy processing requirement.
-
ItemRobust Foreground Detection: A Fusion of Masked Grey World, Probabilistic Gradient Information and Extended Conditional Random Field Approach(MDPI, 2012-05)Foreground detection has been used extensively in many applications such as people counting, traffic monitoring and face recognition. However, most of the existing detectors can only work under limited conditions. This happens because of the inability of the detector to distinguish foreground and background pixels, especially in complex situations. Our aim is to improve the robustness of foreground detection under sudden and gradual illumination change, colour similarity issue, moving background and shadow noise. Since it is hard to achieve robustness using a single model, we have combined several methods into an integrated system. The masked grey world algorithm is introduced to handle sudden illumination change. Colour co-occurrence modelling is then fused with the probabilistic edge-based background modelling. Colour co-occurrence modelling is good in filtering moving background and robust to gradual illumination change, while an edge-based modelling is used for solving a colour similarity problem. Finally, an extended conditional random field approach is used to filter out shadow and afterimage noise. Simulation results show that our algorithm performs better compared to the existing methods, which makes it suitable for higher-level applications.
-
ItemGenerating Adaptive Behaviour within a Memory-Prediction Framework(PUBLIC LIBRARY SCIENCE, 2012-01-17)The Memory-Prediction Framework (MPF) and its Hierarchical-Temporal Memory implementation (HTM) have been widely applied to unsupervised learning problems, for both classification and prediction. To date, there has been no attempt to incorporate MPF/HTM in reinforcement learning or other adaptive systems; that is, to use knowledge embodied within the hierarchy to control a system, or to generate behaviour for an agent. This problem is interesting because the human neocortex is believed to play a vital role in the generation of behaviour, and the MPF is a model of the human neocortex.We propose some simple and biologically-plausible enhancements to the Memory-Prediction Framework. These cause it to explore and interact with an external world, while trying to maximize a continuous, time-varying reward function. All behaviour is generated and controlled within the MPF hierarchy. The hierarchy develops from a random initial configuration by interaction with the world and reinforcement learning only. Among other demonstrations, we show that a 2-node hierarchy can learn to successfully play "rocks, paper, scissors" against a predictable opponent.
-
ItemNo Preview AvailableCardio-renal and metabolic adaptations during pregnancy in female rats born small: implications for maternal health and second generation fetal growth(WILEY, 2012-02)Intrauterine growth restriction caused by uteroplacental insufficiency increases risk of cardiovascular and metabolic disease in offspring. Cardio-renal and metabolic responses to pregnancy are critical determinants of immediate and long-term maternal health. However, no studies to date have investigated the renal and metabolic adaptations in growth restricted offspring when they in turn become pregnant. We hypothesised that the physiological challenge of pregnancy in growth restricted females exacerbates disease outcome and compromises next generation fetal growth. Uteroplacental insufficiency was induced by bilateral uterine vessel ligation (Restricted) or sham surgery (Control) on day 18 of gestation in WKY rats and F1 female offspring birth and postnatal body weights were recorded. F1 Control and Restricted females were mated at 4 months and blood pressure, renal and metabolic parameters were measured in late pregnancy and F2 fetal and placental weights recorded. Age-matched non-pregnant Control and Restricted F1 females were also studied. F1 Restricted females were born 10-15% lighter than Controls. Basal insulin secretion and pancreatic β-cell mass were reduced in non-pregnant Restricted females but restored in pregnancy. Pregnant Restricted females, however, showed impaired glucose tolerance and compensatory glomerular hypertrophy, with a nephron deficit but normal renal function and blood pressure. F2 fetuses from Restricted mothers exposed to physiological measures during pregnancy were lighter than Controls highlighting additive adverse effects when mothers born small experience stress during pregnancy. Female rats born small exhibit mostly normal cardio-renal adaptations but altered glucose control during late pregnancy making them vulnerable to lifestyle challenges.
-
ItemNo Preview AvailableExercise early in life in rats born small does not normalize reductions in skeletal muscle PGC-1α in adulthood(AMER PHYSIOLOGICAL SOC, 2012-05)We have previously shown that 4 wk of exercise training early in life normalizes the otherwise greatly reduced pancreatic β-cell mass in adult male rats born small. The aim of the current study was to determine whether a similar normalization in adulthood of reduced skeletal muscle mitochondrial biogenesis markers and alterations in skeletal muscle lipids of growth-restricted male rats occurs following early exercise training. Bilateral uterine vessel ligation performed on day 18 of gestation resulted in Restricted offspring born small (P < 0.05) compared with both sham-operated Controls and a sham-operated Reduced litter group. Offspring remained sedentary or underwent treadmill running from 5-9 (early exercise) or 20-24 (later exercise) wk of age. At 24 wk of age, Restricted and Reduced litter offspring had lower (P < 0.05) skeletal muscle peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α) protein expression compared with Control offspring. Early exercise training had the expected effect of increasing skeletal muscle markers of mitochondrial biogenesis, but, at this early age (9 wk), there was no deficit in Restricted and Reduced litter skeletal muscle mitochondrial biogenesis. Unlike our previous observations in pancreatic β-cell mass, there was no "reprogramming" effect of early exercise on adult skeletal muscle such that PGC-1α was lower in adult Restricted and Reduced litter offspring irrespective of exercise training. Later exercise training increased mitochondrial biogenesis in all groups. In conclusion, although the response to exercise training remains intact, early exercise training in rats born small does not have a reprogramming effect to prevent deficits in skeletal muscle markers of mitochondrial biogenesis in adulthood.
-
ItemT-cell autoreactivity to citrullinated autoantigenic peptides in rheumatoid arthritis patients carrying HLA-DRB1 shared epitope alleles(BIOMED CENTRAL LTD, 2012)INTRODUCTION: Anti-citrullinated peptide antibodies are found in rheumatoid arthritis (RA) patients with HLA-DRβ chains encoding the shared epitope (SE) sequence. Citrullination increases self-antigen immunogenicity, through increased binding affinity to SE-containing HLA-DR molecules. To characterise T-cell autoreactivity towards citrullinated self-epitopes, we profiled responses of SE+ healthy controls and RA patients to citrullinated and unmodified epitopes of four autoantigens. METHODS: We compared T-cell proliferative and cytokine responses to citrullinated and native type II collagen 1,237 to 1,249, vimentin 66 to 78, aggrecan 84 to 103 and fibrinogen 79 to 91 in six SE+ healthy controls and in 21 RA patients with varying disease duration. Cytokine-producing cells were stained after incubation with peptide in the presence of Brefeldin-A. RESULTS: Although proliferative responses were low, IL-6, IL-17 and TNF were secreted by CD4+ T cells of SE+ RA patients and healthy controls, as well as IFNγ and IL-10 secreted by RA patients, in response to citrullinated peptides. Of the epitopes tested, citrullinated aggrecan was most immunogenic. Patients with early RA were more likely to produce IL-6 in response to no epitope or to citrullinated aggrecan, while patients with longstanding RA were more likely to produce IL-6 to more than one epitope. Cytokine-producing CD4+ T cells included the CD45RO+ and CD45RO- and the CD28+ and CD28- subsets in RA patients. CONCLUSION: Proinflammatory cytokines were produced by CD4+ T cells in SE+ individuals in response to citrullinated self-epitopes, of which citrullinated aggrecan was most immunogenic. Our data suggest that the T-cell response to citrullinated self-epitopes matures and diversifies with development of RA.
-
ItemExosomes: vehicles for the transfer of toxic proteins associated with neurodegenerative diseases?(FRONTIERS MEDIA SA, 2012)Exosomes are small membranous vesicles secreted by a number of cell types including neurons and can be isolated from conditioned cell media or bodily fluids such as urine and plasma. Exosome biogenesis involves the inward budding of endosomes to form multivesicular bodies (MVB). When fused with the plasma membrane, the MVB releases the vesicles into the extracellular environment as exosomes. Proposed functions of these vesicles include roles in cell-cell signaling, removal of unwanted proteins, and the transfer of pathogens between cells. One such pathogen which exploits this pathway is the prion, the infectious particle responsible for the transmissible neurodegenerative diseases such as Creutzfeldt-Jakob disease (CJD) of humans or bovine spongiform encephalopathy (BSE) of cattle. Similarly, exosomes are also involved in the processing of the amyloid precursor protein (APP) which is associated with Alzheimer's disease. Exosomes have been shown to contain full-length APP and several distinct proteolytically cleaved products of APP, including Aβ. In addition, these fragments can be modulated using inhibitors of the proteases involved in APP cleavage. These observations provide further evidence for a novel pathway in which PrP and APP fragments are released from cells. Other proteins such as superoxide dismutase I and alpha-synuclein (involved in amyotrophic lateral sclerosis and Parkinson's disease, respectively) are also found associated with exosomes. This review will focus on the role of exosomes in neurodegenerative disorders and discuss the potential of these vesicles for the spread of neurotoxicity, therapeutics, and diagnostics for these diseases.
-
ItemNon-linear separation of pressure, velocity and wave intensity into forward and backward components(SPRINGER HEIDELBERG, 2012-06)Separating pressure, flow/velocity and wave intensity signals into forward and backward components provide insights about arterial wave propagation and reflection. A linear wave separation is normally used, but ignores the pressure-dependence of wave speed. While a non-linear separation could incorporate this pressure-dependence, no such method exists for wave intensity decomposition. Moreover, although linear separation errors for pressure (5-10 %) have been quantified previously, errors for velocity and wave intensity have not. Accordingly, we describe a non-linear wave separation technique based on the method of characteristics. Data from a computer model suggest that the percentage linear separation errors for velocity and wave intensity are approximately one-half and twice that for pressure, respectively. Although comparable to measurement uncertainty in many instances, linear separation errors may become more significant: (1) if wave speed varies substantially over the cardiac cycle, e.g. if pulse pressure or vessel compliance is high, (2) if the degree of wave reflection in the arterial system is large, or (3) if the constant wave speed used for the linear separation is closer to the minimum or maximum pressure-dependent value rather than the mean. Consideration of linear separation errors may therefore be important in some physiological settings.
-
ItemNo Preview AvailablePhotoinitiated Alkyne-Azide Click and Radical Cross-Linking Reactions for the Patterning of PEG Hydrogels(AMER CHEMICAL SOC, 2012-03)The photolithographical patterning of hydrogels based solely on the surface immobilization and cross-linking of alkyne-functionalized poly(ethylene glycol) (PEG-tetraalkyne) is described. Photogenerated radicals as well as UV absorption by a copper chelating ligand result in the photochemical redox reduction of Cu(II) to Cu(I). This catalyzes the alkyne-azide click reaction to graft the hydrogels onto an azide-functionalized plasma polymer (N(3)PP) film. The photogenerated radicals were also able to abstract hydrogen atoms from PEG-tetraalkyne to form poly(α-alkoxy) radicals. These radicals can initiate cross-linking by addition to the alkynes and intermolecular recombination to form the PEG hydrogels. Spatially controlling the two photoinitiated reactions by UV exposure through a photomask leads to surface patterned hydrogels, with thicknesses that were tunable from tens to several hundreds of nanometers. The patterned PEG hydrogels (ca. 60 μm wide lines) were capable of resisting the attachment of L929 mouse fibroblast cells, resulting in surfaces with spatially controlled cell attachment. The patterned hydrogel surface also demonstrated spatially resolved chemical functionality, as postsynthetic modification of the hydrogels was successfully carried out with azide-functionalized fluorescent dyes via subsequent alkyne-azide click reactions.