A principled approach to interactive hierarchical non-linear visualization of high-dimensional data

Hierarchical visualization systems are desirable because a single two-dimensional visualization plot may not be sufficient to capture all of the interesting aspects of complex high-dimensional data sets. We extend an existing locally linear hierarchical visualization system PhiVis [1] in several directions: bf(1) we allow for em non-linear projection manifolds (the basic building block is the Generative Topographic Mapping -- GTM), bf(2) we introduce a general formulation of hierarchical probabilistic models consisting of local probabilistic models organized in a hierarchical tree, bf(3) we describe folding patterns of low-dimensional projection manifold in high-dimensional data space by computing and visualizing the manifold's local directional curvatures. Quantities such as magnification factors [3] and directional curvatures are helpful for understanding the layout of the nonlinear projection manifold in the data space and for further refinement of the hierarchical visualization plot. Like PhiVis, our system is statistically principled and is built interactively in a top-down fashion using the EM algorithm. We demonstrate the visualization system principle of the approach on a complex 12-dimensional data set and mention possible applications in the pharmaceutical industry.

A hierarchical latent variable model for data visualization

Visualization has proven to be a powerful and widely-applicable tool the analysis and interpretation of data. Most visualization algorithms aim to find a projection from the data space down to a two-dimensional visualization space. However, for complex data sets living in a high-dimensional space it is unlikely that a single two-dimensional projection can reveal all of the interesting structure. We therefore introduce a hierarchical visualization algorithm which allows the complete data set to be visualized at the top level, with clusters and sub-clusters of data points visualized at deeper levels. The algorithm is based on a hierarchical mixture of latent variable models, whose parameters are estimated using the expectation-maximization algorithm. We demonstrate the principle of the approach first on a toy data set, and then apply the algorithm to the visualization of a synthetic data set in 12 dimensions obtained from a simulation of multi-phase flows in oil pipelines and to data in 36 dimensions derived from satellite images.

Novel visualization methods for protein data

Visualization of high-dimensional data has always been a challenging task. Here we discuss and propose variants of non-linear data projection methods (Generative Topographic Mapping (GTM) and GTM with simultaneous feature saliency (GTM-FS)) that are adapted to be effective on very high-dimensional data. The adaptations use log space values at certain steps of the Expectation Maximization (EM) algorithm and during the visualization process. We have tested the proposed algorithms by visualizing electrostatic potential data for Major Histocompatibility Complex (MHC) class-I proteins. The experiments show that the variation in the original version of GTM and GTM-FS worked successfully with data of more than 2000 dimensions and we compare the results with other linear/nonlinear projection methods: Principal Component Analysis (PCA), Neuroscale (NSC) and Gaussian Process Latent Variable Model (GPLVM).

Semisupervised learning of hierarchical latent trait models for data visualization

Recently, we have developed the hierarchical Generative Topographic Mapping (HGTM), an interactive method for visualization of large high-dimensional real-valued data sets. In this paper, we propose a more general visualization system by extending HGTM in three ways, which allows the user to visualize a wider range of data sets and better support the model development process. 1) We integrate HGTM with noise models from the exponential family of distributions. The basic building block is the Latent Trait Model (LTM). This enables us to visualize data of inherently discrete nature, e.g., collections of documents, in a hierarchical manner. 2) We give the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode, the user selects "regions of interest," whereas in the automatic mode, an unsupervised minimum message length (MML)-inspired construction of a mixture of LTMs is employed. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. 3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualization plots, since they can highlight the boundaries between data clusters. We illustrate our approach on a toy example and evaluate it on three more complex real data sets. © 2005 IEEE.

Data visualization during the early stages of drug discovery

Multidimensional compound optimization is a new paradigm in the drug discovery process, yielding efficiencies during early stages and reducing attrition in the later stages of drug development. The success of this strategy relies heavily on understanding this multidimensional data and extracting useful information from it. This paper demonstrates how principled visualization algorithms can be used to understand and explore a large data set created in the early stages of drug discovery. The experiments presented are performed on a real-world data set comprising biological activity data and some whole-molecular physicochemical properties. Data visualization is a popular way of presenting complex data in a simpler form. We have applied powerful principled visualization methods, such as generative topographic mapping (GTM) and hierarchical GTM (HGTM), to help the domain experts (screening scientists, chemists, biologists, etc.) understand and draw meaningful decisions. We also benchmark these principled methods against relatively better known visualization approaches, principal component analysis (PCA), Sammon's mapping, and self-organizing maps (SOMs), to demonstrate their enhanced power to help the user visualize the large multidimensional data sets one has to deal with during the early stages of the drug discovery process. The results reported clearly show that the GTM and HGTM algorithms allow the user to cluster active compounds for different targets and understand them better than the benchmarks. An interactive software tool supporting these visualization algorithms was provided to the domain experts. The tool facilitates the domain experts by exploration of the projection obtained from the visualization algorithms providing facilities such as parallel coordinate plots, magnification factors, directional curvatures, and integration with industry standard software. © 2006 American Chemical Society.

GTM: the generative topographic mapping

This thesis describes the Generative Topographic Mapping (GTM) --- a non-linear latent variable model, intended for modelling continuous, intrinsically low-dimensional probability distributions, embedded in high-dimensional spaces. It can be seen as a non-linear form of principal component analysis or factor analysis. It also provides a principled alternative to the self-organizing map --- a widely established neural network model for unsupervised learning --- resolving many of its associated theoretical problems. An important, potential application of the GTM is visualization of high-dimensional data. Since the GTM is non-linear, the relationship between data and its visual representation may be far from trivial, but a better understanding of this relationship can be gained by computing the so-called magnification factor. In essence, the magnification factor relates the distances between data points, as they appear when visualized, to the actual distances between those data points. There are two principal limitations of the basic GTM model. The computational effort required will grow exponentially with the intrinsic dimensionality of the density model. However, if the intended application is visualization, this will typically not be a problem. The other limitation is the inherent structure of the GTM, which makes it most suitable for modelling moderately curved probability distributions of approximately rectangular shape. When the target distribution is very different to that, theaim of maintaining an `interpretable' structure, suitable for visualizing data, may come in conflict with the aim of providing a good density model. The fact that the GTM is a probabilistic model means that results from probability theory and statistics can be used to address problems such as model complexity. Furthermore, this framework provides solid ground for extending the GTM to wider contexts than that of this thesis.

DVMS 1.5: A user manual (the data visualization and modeling system)

The data available during the drug discovery process is vast in amount and diverse in nature. To gain useful information from such data, an effective visualisation tool is required. To provide better visualisation facilities to the domain experts (screening scientist, biologist, chemist, etc.),we developed a software which is based on recently developed principled visualisation algorithms such as Generative Topographic Mapping (GTM) and Hierarchical Generative Topographic Mapping (HGTM). The software also supports conventional visualisation techniques such as Principal Component Analysis, NeuroScale, PhiVis, and Locally Linear Embedding (LLE). The software also provides global and local regression facilities . It supports regression algorithms such as Multilayer Perceptron (MLP), Radial Basis Functions network (RBF), Generalised Linear Models (GLM), Mixture of Experts (MoE), and newly developed Guided Mixture of Experts (GME). This user manual gives an overview of the purpose of the software tool, highlights some of the issues to be taken care while creating a new model, and provides information about how to install & use the tool. The user manual does not require the readers to have familiarity with the algorithms it implements. Basic computing skills are enough to operate the software.

miniDVMS v1.8 : A user manual(The data visualization and modeling system)

Today, the data available to tackle many scientific challenges is vast in quantity and diverse in nature. The exploration of heterogeneous information spaces requires suitable mining algorithms as well as effective visual interfaces. miniDVMS v1.8 provides a flexible visual data mining framework which combines advanced projection algorithms developed in the machine learning domain and visual techniques developed in the information visualisation domain. The advantage of this interface is that the user is directly involved in the data mining process. Principled projection methods, such as generative topographic mapping (GTM) and hierarchical GTM (HGTM), are integrated with powerful visual techniques, such as magnification factors, directional curvatures, parallel coordinates, and user interaction facilities, to provide this integrated visual data mining framework. The software also supports conventional visualisation techniques such as principal component analysis (PCA), Neuroscale, and PhiVis. This user manual gives an overview of the purpose of the software tool, highlights some of the issues to be taken care while creating a new model, and provides information about how to install and use the tool. The user manual does not require the readers to have familiarity with the algorithms it implements. Basic computing skills are enough to operate the software.

Data visualization with simultaneous feature selection

Data visualization algorithms and feature selection techniques are both widely used in bioinformatics but as distinct analytical approaches. Until now there has been no method of measuring feature saliency while training a data visualization model. We derive a generative topographic mapping (GTM) based data visualization approach which estimates feature saliency simultaneously with the training of the visualization model. The approach not only provides a better projection by modeling irrelevant features with a separate noise model but also gives feature saliency values which help the user to assess the significance of each feature. We compare the quality of projection obtained using the new approach with the projections from traditional GTM and self-organizing maps (SOM) algorithms. The results obtained on a synthetic and a real-life chemoinformatics dataset demonstrate that the proposed approach successfully identifies feature significance and provides coherent (compact) projections. © 2006 IEEE.

Computational mechanics of molecular systems:quantifying high-dimensional dynamics by distribution of Poincaré recurrence times

A framework that connects computational mechanics and molecular dynamics has been developed and described. As the key parts of the framework, the problem of symbolising molecular trajectory and the associated interrelation between microscopic phase space variables and macroscopic observables of the molecular system are considered. Following Shalizi and Moore, it is shown that causal states, the constituent parts of the main construct of computational mechanics, the e-machine, define areas of the phase space that are optimal in the sense of transferring information from the micro-variables to the macro-observables. We have demonstrated that, based on the decay of their Poincare´ return times, these areas can be divided into two classes that characterise the separation of the phase space into resonant and chaotic areas. The first class is characterised by predominantly short time returns, typical to quasi-periodic or periodic trajectories. This class includes a countable number of areas corresponding to resonances. The second class includes trajectories with chaotic behaviour characterised by the exponential decay of return times in accordance with the Poincare´ theorem.

High level data fusion

We address the question of how to obtain effective fusion of identification information such that it is robust to the quality of this information. As well as technical issues data fusion is encumbered with a collection of (potentially confusing) practical considerations. These considerations are described during the early chapters in which a framework for data fusion is developed. Following this process of diversification it becomes clear that the original question is not well posed and requires more precise specification. We use the framework to focus on some of the technical issues relevant to the question being addressed. We show that fusion of hard decisions through use of an adaptive version of the maximum a posteriori decision rule yields acceptable performance. Better performance is possible using probability level fusion as long as the probabilities are accurate. Of particular interest is the prevalence of overconfidence and the effect it has on fused performance. The production of accurate probabilities from poor quality data forms the latter part of the thesis. Two approaches are taken. Firstly the probabilities may be moderated at source (either analytically or numerically). Secondly, the probabilities may be transformed at the fusion centre. In each case an improvement in fused performance is demonstrated. We therefore conclude that in order to obtain robust fusion care should be taken to model the probabilities accurately; either at the source or centrally.

Dynamic complexity of chaotic transitions in high-dimensional classical dynamics:Leu-Enkephalin folding

Leu-Enkephalin in explicit water is simulated using classical molecular dynamics. A ß-turn transition is investigated by calculating the topological complexity (in the "computational mechanics" framework [J. P. Crutchfield and K. Young, Phys. Rev. Lett., 63, 105 (1989)]) of the dynamics of both the peptide and the neighbouring water molecules. The complexity of the atomic trajectories of the (relatively short) simulations used in this study reflect the degree of phase space mixing in the system. It is demonstrated that the dynamic complexity of the hydrogen atoms of the peptide and almost all of the hydrogens of the neighbouring waters exhibit a minimum precisely at the moment of the ß-turn transition. This indicates the appearance of simplified periodic patterns in the atomic motion, which could correspond to high-dimensional tori in the phase space. It is hypothesized that this behaviour is the manifestation of the effect described in the approach to molecular transitions by Komatsuzaki and Berry [T. Komatsuzaki and R.S. Berry, Adv. Chem. Phys., 123, 79 (2002)], where a "quasi-regular" dynamics at the transition is suggested. Therefore, for the first time, the less chaotic character of the folding transition in a realistic molecular system is demonstrated. © Springer-Verlag Berlin Heidelberg 2006.

Understanding data collection behaviour of mental health practitioners

Failure to detect patients at risk of attempting suicide can result in tragic consequences. Identifying risks earlier and more accurately helps prevent serious incidents occurring and is the objective of the GRiST clinical decision support system (CDSS). One of the problems it faces is high variability in the type and quantity of data submitted for patients, who are assessed in multiple contexts along the care pathway. Although GRiST identifies up to 138 patient cues to collect, only about half of them are relevant for any one patient and their roles may not be for risk evaluation but more for risk management. This paper explores the data collection behaviour of clinicians using GRiST to see whether it can elucidate which variables are important for risk evaluations and when. The GRiST CDSS is based on a cognitive model of human expertise manifested by a sophisticated hierarchical knowledge structure or tree. This structure is used by the GRiST interface to provide top-down controlled access to the patient data. Our research explores relationships between the answers given to these higher-level 'branch' questions to see whether they can help direct assessors to the most important data, depending on the patient profile and assessment context. The outcome is a model for dynamic data collection driven by the knowledge hierarchy. It has potential for improving other clinical decision support systems operating in domains with high dimensional data that are only partially collected and in a variety of combinations.

Improved data visualisation through multiple dissimilarity modelling

Popular dimension reduction and visualisation algorithms rely on the assumption that input dissimilarities are typically Euclidean, for instance Metric Multidimensional Scaling, t-distributed Stochastic Neighbour Embedding and the Gaussian Process Latent Variable Model. It is well known that this assumption does not hold for most datasets and often high-dimensional data sits upon a manifold of unknown global geometry. We present a method for improving the manifold charting process, coupled with Elastic MDS, such that we no longer assume that the manifold is Euclidean, or of any particular structure. We draw on the benefits of different dissimilarity measures allowing for the relative responsibilities, under a linear combination, to drive the visualisation process.