Clustering strategies for object databases

When object databases arrived on the scene some ten years ago, they provided database capabilities for previously neglected, complex applications, such as CAD, but were burdened with one inherent teething problem, poor performance. Physical database design is one tool that can provide performance improvements and it is the general area of concern for this thesis. Clustering is one fruitful design technique which can provide improvements in performance. However, clustering in object databases has not been explored in depth and so has not been truly exploited. Further, clustering, although a physical concern, can be determined from the logical model. The object model is richer than previous models, notably the relational model, and so it is anticipated that the opportunities with respect to clustering are greater. This thesis provides a thorough analysis of object clustering strategies with a view to highlighting any links between the object logical and physical model and improving performance. This is achieved by considering all possible types of object logical model construct and the implementation of those constructs in terms of theoretical clusterings strategies to produce actual clustering arrangements. This analysis results in a greater understanding of object clustering strategies, aiding designers in the development process and providing some valuable rules of thumb to support the design process.

Higher-order co-occurrences for exploratory point pattern analysis and decision tree clustering on spatial data

Analyzing geographical patterns by collocating events, objects or their attributes has a long history in surveillance and monitoring, and is particularly applied in environmental contexts, such as ecology or epidemiology. The identification of patterns or structures at some scales can be addressed using spatial statistics, particularly marked point processes methodologies. Classification and regression trees are also related to this goal of finding "patterns" by deducing the hierarchy of influence of variables on a dependent outcome. Such variable selection methods have been applied to spatial data, but, often without explicitly acknowledging the spatial dependence. Many methods routinely used in exploratory point pattern analysis are2nd-order statistics, used in a univariate context, though there is also a wide literature on modelling methods for multivariate point pattern processes. This paper proposes an exploratory approach for multivariate spatial data using higher-order statistics built from co-occurrences of events or marks given by the point processes. A spatial entropy measure, derived from these multinomial distributions of co-occurrences at a given order, constitutes the basis of the proposed exploratory methods. © 2010 Elsevier Ltd.

Clustering of cerebral cortical lesions in patients with corticobasal degeneration

Clustering of ballooned neurons (BN) and tau positive neurons with inclusion bodies (tau+ neurons) was studied in the upper and lower laminae of the frontal, parietal and temporal cortex in 12 patients with corticobasal degeneration (CBD). In a significant proportion of brain areas examined, BN and tau+ neurons exhibited clustering with a regular distribution of clusters parallel to the pia mater. A regular pattern of clustering of BN and tau+ neurons was observed equally frequently in all cortical areas examined and in the upper and lower laminae. No significant correlations were observed between the cluster sizes of BN or tau+ neurons in the upper compared with the lower cortex or between the cluster sizes of BN and tau+ neurons. The results suggest that BN and tau+ neurons in CBD exhibit the same type of spatial pattern as lesions in Alzheimer's disease, Lewy body dementia and Pick's disease. The regular periodicity of the cerebral cortical lesions is consistent with the degeneration of the cortico-cortical projections in CBD.

Clustering of Pick bodies in patients with Pick's disease

Clustering of Pick bodies (PB) was studied in the frontal and temporal lobe in 10 cases of Pick's disease (PD). Pick bodies exhibited clustering in 47/50 (94%) brain areas analysed. In 20/50 (40%) brain areas, PB were present in a single large cluster ≤ 6400 μm in diameter, in 27/50 (54%) PB occurred in smaller clusters (200-3200 μm in diameter) which exhibited a regular periodicity relative to the tissue boundary, in 1/50 (2%) there was a regular distribution of individual PB and in 2/50 (4%), PB were randomly distributed. Mean cluster size of the PB was greater in the dentate gyrus compared with the inferior temporal gyrus and lateral occipitotemporal gyrus. Mean cluster size of PB in a brain region was positively correlated with the mean density of PB. Hence, PB exhibit essentially the same spatial patterns as senile plaques and neurofibrillary tangles in Alzheimer's disease (AD) and Lewy bodies in Dementia with Lewy bodies (DLB).

Dementia with Lewy bodies:Clustering of Lewy bodies in human patients

Clustering of Lewy bodies (LB) was studied in four regions of the medial temporal lobe in 12 cases of dementia with LB (DLB). LB exhibited clustering in 67/70 (96%) brain areas analysed. In 34/70 (49%) analyses, LB were present in a single large cluster ≤6400 μm in diameter, in 33/70 (47%) LB occurred in smaller clusters 200-3200 μm in diameter which exhibited a regular periodicity relative to the tissue boundary and in 3/70 (4%), LB were randomly distributed. A regular pattern of LB clusters was observed equally frequently in the cortex and hippocampus, in upper and lower cortical laminae and in 'pure' cases of DLB with negligible Alzheimer's disease (AD) pathology compared with cases of AD with DLB. In cortical regions, there was no significant correlation between LB cluster size in the upper and lower cortical laminae. The regular periodicity of LB clusters suggests that LB develop in relation to the cells of origin of specific cortico-cortical and cortico-hippocampal projections.

Is the clustering of β-amyloid (Aβ) deposits in the frontal cortex of Alzheimer patients determined by blood vessels?

The clustering pattern of diffuse, primitive and classic β-amyloid (Aβ) deposits was studied in the upper laminae of the frontal cortex of 9 patients with sporadic Alzheimer's disease (AD). Aβ stained tissue was counterstained with collagen type IV antiserum to determine whether the clusters of Aβ deposits were related to blood vessels. In all patients, Aβ deposits and blood vessels were clustered, with in many patients, a regular periodicity of clusters along the cortex parallel to the pia. The classic Aβ deposit clusters coincided with those of the larger blood vessels in all patients and with clusters of smaller blood vessels in 4 patients. Diffuse deposit clusters were related to blood vessels in 3 patients. Primitive deposit clusters were either unrelated to or negatively correlated with the blood vessels in six patients. Hence, Aβ deposit subtypes differ in their relationship to blood vessels. The data suggest a direct and specific role for the larger blood vessels in the formation of amyloid cores in AD. © 1995.

Is the clustering of neurofibrillary tangles in Alzheimer's patients related to the cells of origin of specific cortico-cortical projections?

The spatial pattern of cellular neurofibrillary tangles (NFT) was studied in the supra- and infragranular layers of various cortical regions in cases of Alzheimer's disease (AD). The objective was to test the hypothesis that NFT formation was associated with the cells of origin of specific cortico-cortical projections. The novel feature of the study was that pattern analysis enabled the dimension and spacing of NFT clusters along the cortical ribbon to be estimated. In the majority of brain regions studied, NFT occurred in clusters of neurons which were regularly spaced along the cortical strip. This pattern is consistent with the predicted distribution of the cells of origin of specific cortico-cortico projections. Mean NFT cluster size varied from 250 to > 12800 microns in different cortical tissues suggesting either variation in the size of the cell clusters or a dynamic process in the development of NFT in relation to these cell clusters. The formation of NFT in cell clusters which may give rise to the feed-forward and feed-back cortico-cortical projections suggests a possible route of spread of NFT pathology in AD between cortical regions and from the cortex to subcortical areas.

Changes in the clustering patterns of beta-amyloid (Abeta) with age in Down's syndrome

The spatial distribution patterns of the diffuse, primitive, and classic beta-amyloid (Abeta) deposits were studied in areas of the medial temporal lobe in 12 cases of Down's Syndrome (DS) 35 to 67 years of age. Large clusters of diffuse deposits were present in the youngest patients; cluster size then declined with patient age but increased again in the oldest patients. By contrast, the cluster sizes of the primitive and classic deposits increased with age to a maximum in patients 45 to 55 and 60 years of age respectively and declined in size in the oldest patients. In the parahippocampal gyrus (PHG), the clusters of the primitive deposits were most highly clustered in cases of intermediate age. The data suggest a developmental sequence in DS in which Abeta is deposited initially in the form of large clusters of diffuse deposits that are then gradually replaced by clusters of primitive and classic deposits. The oldest patients were an exception to this sequence in that the pattern of clustering resembled that of the youngest patients.

Clustering patterns of neurofibrillary tangles in Alzheimer’s disease

Clustering of cellular neurofibrillary tangles (NFT) was studied in the cerebral cortex and hippocampus in cases of Alzheimer’s disease (AD) using a regression method. The objective of the study was to test the hypothesis that clustering of NFTs reflects the degeneration of the cortico-cortical pathways. In 25/38 (66%) of analyses of individual brain areas, a significant peak to trough and peak to peak distance was obtained suggesting that the clusters of NFTs were regularly distributed in bands parallel to the tissue boundary. In analyses of cortical tissues with regularly distributed clusters, peak to peak distance was between 1000 and 1600 microns in 13/24 (54%) of analyses, >1600 microns in 10/24 (42%) and <1000 microns in 1/24 (4%) of analyses. A regular distribution of NFT clusters was less evident in the CA sectors of the hippocampus than in the cortex. Hence, in a significant proportion of brain areas, the spacing of NFT clusters along the cerebral cortex was consistent with the predicted distribution of the cells of origin of specific cortico-cortical projections. However, in many brain regions, the sizes of the NFT clusters were larger than predicted which may be attributable to the spread of NFTs to adjacent groups of cells as the disease progresses.

Spatially orthogonal chemical functionalization of a hierarchical pore network for catalytic cascade reactions

The chemical functionality within porous architectures dictates their performance as heterogeneous catalysts; however, synthetic routes to control the spatial distribution of individual functions within porous solids are limited. Here we report the fabrication of spatially orthogonal bifunctional porous catalysts, through the stepwise template removal and chemical functionalization of an interconnected silica framework. Selective removal of polystyrene nanosphere templates from a lyotropic liquid crystal-templated silica sol–gel matrix, followed by extraction of the liquid crystal template, affords a hierarchical macroporous–mesoporous architecture. Decoupling of the individual template extractions allows independent functionalization of macropore and mesopore networks on the basis of chemical and/or size specificity. Spatial compartmentalization of, and directed molecular transport between, chemical functionalities affords control over the reaction sequence in catalytic cascades; herein illustrated by the Pd/Pt-catalysed oxidation of cinnamyl alcohol to cinnamic acid. We anticipate that our methodology will prompt further design of multifunctional materials comprising spatially compartmentalized functions.

Higher-order co-occurrences for exploratory point pattern analysis and decision tree clustering on spatial data

Analyzing geographical patterns by collocating events, objects or their attributes has a long history in surveillance and monitoring, and is particularly applied in environmental contexts, such as ecology or epidemiology. The identification of patterns or structures at some scales can be addressed using spatial statistics, particularly marked point processes methodologies. Classification and regression trees are also related to this goal of finding "patterns" by deducing the hierarchy of influence of variables on a dependent outcome. Such variable selection methods have been applied to spatial data, but, often without explicitly acknowledging the spatial dependence. Many methods routinely used in exploratory point pattern analysis are2nd-order statistics, used in a univariate context, though there is also a wide literature on modelling methods for multivariate point pattern processes. This paper proposes an exploratory approach for multivariate spatial data using higher-order statistics built from co-occurrences of events or marks given by the point processes. A spatial entropy measure, derived from these multinomial distributions of co-occurrences at a given order, constitutes the basis of the proposed exploratory methods. © 2010 Elsevier Ltd.

Sensitivity of peptide conformational dynamics on clustering of a classical molecular dynamics trajectory

We investigate the sensitivity of a Markov model with states and transition probabilities obtained from clustering a molecular dynamics trajectory. We have examined a 500 ns molecular dynamics trajectory of the peptide valine-proline-alanine-leucine in explicit water. The sensitivity is quantified by varying the boundaries of the clusters and investigating the resulting variation in transition probabilities and the average transition time between states. In this way, we represent the effect of clustering using different clustering algorithms. It is found that in terms of the investigated quantities, the peptide dynamics described by the Markov model is sensitive to the clustering; in particular, the average transition times are found to vary up to 46%. Moreover, inclusion of nonphysical sparsely populated clusters can lead to serious errors of up to 814%. In the investigation, the time step used in the transition matrix is determined by the minimum time scale on which the system behaves approximately Markovian. This time step is found to be about 100 ps. It is concluded that the description of peptide dynamics with transition matrices should be performed with care, and that using standard clustering algorithms to obtain states and transition probabilities may not always produce reliable results.

The development of a multi-function computer-assisted instruction system using a hierarchical database structure as the lesson compendium.

Computer-Based Learning systems of one sort or another have been in existence for almost 20 years, but they have yet to achieve real credibility within Commerce, Industry or Education. A variety of reasons could be postulated for this, typically: - cost - complexity - inefficiency - inflexibility - tedium Obviously different systems deserve different levels and types of criticism, but it still remains true that Computer-Based Learning (CBL) is falling significantly short of its potential. Experience of a small, but highly successful CBL system within a large, geographically distributed industry (the National Coal Board) prompted an investigation into currently available packages, the original intention being to purchase the most suitable software and run it on existing computer hardware, alongside existing software systems. It became apparent that none of the available CBL packages were suitable, and a decision was taken to develop an in-house Computer-Assisted Instruction system according to the following criteria: - cheap to run; - easy to author course material; - easy to use; - requires no computing knowledge to use (as either an author or student) ; - efficient in the use of computer resources; - has a comprehensive range of facilities at all levels. This thesis describes the initial investigation, resultant observations and the design, development and implementation of the SCHOOL system. One of the principal characteristics c£ SCHOOL is that it uses a hierarchical database structure for the storage of course material - thereby providing inherently a great deal of the power, flexibility and efficiency originally required. Trials using the SCHOOL system on IBM 303X series equipment are also detailed, along with proposed and current development work on what is essentially an operational CBL system within a large-scale Industrial environment.

Characterizing the hierarchical structures of bioactive sol-gel silicate glass and hybrid scaffolds for bone regeneration

Bone is the second most widely transplanted tissue after blood. Synthetic alternatives are needed that can reduce the need for transplants and regenerate bone by acting as active temporary templates for bone growth. Bioactive glasses are one of the most promising bone replacement/regeneration materials because they bond to existing bone, are degradable and stimulate new bone growth by the action of their dissolution products on cells. Sol-gel-derived bioactive glasses can be foamed to produce interconnected macropores suitable for tissue ingrowth, particularly cell migration and vascularization and cell penetration. The scaffolds fulfil many of the criteria of an ideal synthetic bone graft, but are not suitable for all bone defect sites because they are brittle. One strategy for improving toughness of the scaffolds without losing their other beneficial properties is to synthesize inorganic/organic hybrids. These hybrids have polymers introduced into the sol-gel process so that the organic and inorganic components interact at the molecular level, providing control over mechanical properties and degradation rates. However, a full understanding of how each feature or property of the glass and hybrid scaffolds affects cellular response is needed to optimize the materials and ensure long-term success and clinical products. This review focuses on the techniques that have been developed for characterizing the hierarchical structures of sol-gel glasses and hybrids, from atomicscale amorphous networks, through the covalent bonding between components in hybrids and nanoporosity, to quantifying open macroporous networks of the scaffolds. Methods for non-destructive in situ monitoring of degradation and bioactivity mechanisms of the materials are also included. © 2012 The Royal Society.

Non-linear hierarchical visualisation

This thesis applies a hierarchical latent trait model system to a large quantity of data. The motivation for it was lack of viable approaches to analyse High Throughput Screening datasets which maybe include thousands of data points with high dimensions. High Throughput Screening (HTS) is an important tool in the pharmaceutical industry for discovering leads which can be optimised and further developed into candidate drugs. Since the development of new robotic technologies, the ability to test the activities of compounds has considerably increased in recent years. Traditional methods, looking at tables and graphical plots for analysing relationships between measured activities and the structure of compounds, have not been feasible when facing a large HTS dataset. Instead, data visualisation provides a method for analysing such large datasets, especially with high dimensions. So far, a few visualisation techniques for drug design have been developed, but most of them just cope with several properties of compounds at one time. We believe that a latent variable model (LTM) with a non-linear mapping from the latent space to the data space is a preferred choice for visualising a complex high-dimensional data set. As a type of latent variable model, the latent trait model can deal with either continuous data or discrete data, which makes it particularly useful in this domain. In addition, with the aid of differential geometry, we can imagine the distribution of data from magnification factor and curvature plots. Rather than obtaining the useful information just from a single plot, a hierarchical LTM arranges a set of LTMs and their corresponding plots in a tree structure. We model the whole data set with a LTM at the top level, which is broken down into clusters at deeper levels of t.he hierarchy. In this manner, the refined visualisation plots can be displayed in deeper levels and sub-clusters may be found. Hierarchy of LTMs is trained using expectation-maximisation (EM) algorithm to maximise its likelihood with respect to the data sample. Training proceeds interactively in a recursive fashion (top-down). The user subjectively identifies interesting regions on the visualisation plot that they would like to model in a greater detail. At each stage of hierarchical LTM construction, the EM algorithm alternates between the E- and M-step. Another problem that can occur when visualising a large data set is that there may be significant overlaps of data clusters. It is very difficult for the user to judge where centres of regions of interest should be put. We address this problem by employing the minimum message length technique, which can help the user to decide the optimal structure of the model. In this thesis we also demonstrate the applicability of the hierarchy of latent trait models in the field of document data mining.