Maximum Likelihood Estimation of Mixture Densities for Binned and Truncated Multivariate Data

Binning and truncation of data are common in data analysis and machine learning. This paper addresses the problem of fitting mixture densities to multivariate binned and truncated data. The EM approach proposed by McLachlan and Jones (Biometrics, 44: 2, 571-578, 1988) for the univariate case is generalized to multivariate measurements. The multivariate solution requires the evaluation of multidimensional integrals over each bin at each iteration of the EM procedure. Naive implementation of the procedure can lead to computationally inefficient results. To reduce the computational cost a number of straightforward numerical techniques are proposed. Results on simulated data indicate that the proposed methods can achieve significant computational gains with no loss in the accuracy of the final parameter estimates. Furthermore, experimental results suggest that with a sufficient number of bins and data points it is possible to estimate the true underlying density almost as well as if the data were not binned. The paper concludes with a brief description of an application of this approach to diagnosis of iron deficiency anemia, in the context of binned and truncated bivariate measurements of volume and hemoglobin concentration from an individual's red blood cells.

A mixture model-based approach to the clustering of microarray expression data

Motivation: This paper introduces the software EMMIX-GENE that has been developed for the specific purpose of a model-based approach to the clustering of microarray expression data, in particular, of tissue samples on a very large number of genes. The latter is a nonstandard problem in parametric cluster analysis because the dimension of the feature space (the number of genes) is typically much greater than the number of tissues. A feasible approach is provided by first selecting a subset of the genes relevant for the clustering of the tissue samples by fitting mixtures of t distributions to rank the genes in order of increasing size of the likelihood ratio statistic for the test of one versus two components in the mixture model. The imposition of a threshold on the likelihood ratio statistic used in conjunction with a threshold on the size of a cluster allows the selection of a relevant set of genes. However, even this reduced set of genes will usually be too large for a normal mixture model to be fitted directly to the tissues, and so the use of mixtures of factor analyzers is exploited to reduce effectively the dimension of the feature space of genes. Results: The usefulness of the EMMIX-GENE approach for the clustering of tissue samples is demonstrated on two well-known data sets on colon and leukaemia tissues. For both data sets, relevant subsets of the genes are able to be selected that reveal interesting clusterings of the tissues that are either consistent with the external classification of the tissues or with background and biological knowledge of these sets.

A simulated annealing algorithm for finding consensus sequences

Motivation: A consensus sequence for a family of related sequences is, as the name suggests, a sequence that captures the features common to most members of the family. Consensus sequences are important in various DNA sequencing applications and are a convenient way to characterize a family of molecules. Results: This paper describes a new algorithm for finding a consensus sequence, using the popular optimization method known as simulated annealing. Unlike the conventional approach of finding a consensus sequence by first forming a multiple sequence alignment, this algorithm searches for a sequence that minimises the sum of pairwise distances to each of the input sequences. The resulting consensus sequence can then be used to induce a multiple sequence alignment. The time required by the algorithm scales linearly with the number of input sequences and quadratically with the length of the consensus sequence. We present results demonstrating the high quality of the consensus sequences and alignments produced by the new algorithm. For comparison, we also present similar results obtained using ClustalW. The new algorithm outperforms ClustalW in many cases.

Integrals for continuous-time Markov chains

This paper presents a method of evaluating the expected value of a path integral for a general Markov chain on a countable state space. We illustrate the method with reference to several models, including birth-death processes and the birth, death and catastrophe process. (C) 2002 Elsevier Science Inc. All rights reserved.

Computation of the linear elastic properties of random porous materials with a wide variety of microstructure

A finite-element method is used to study the elastic properties of random three-dimensional porous materials with highly interconnected pores. We show that Young's modulus, E, is practically independent of Poisson's ratio of the solid phase, nu(s), over the entire solid fraction range, and Poisson's ratio, nu, becomes independent of nu(s) as the percolation threshold is approached. We represent this behaviour of nu in a flow diagram. This interesting but approximate behaviour is very similar to the exactly known behaviour in two-dimensional porous materials. In addition, the behaviour of nu versus nu(s) appears to imply that information in the dilute porosity limit can affect behaviour in the percolation threshold limit. We summarize the finite-element results in terms of simple structure-property relations, instead of tables of data, to make it easier to apply the computational results. Without using accurate numerical computations, one is limited to various effective medium theories and rigorous approximations like bounds and expansions. The accuracy of these equations is unknown for general porous media. To verify a particular theory it is important to check that it predicts both isotropic elastic moduli, i.e. prediction of Young's modulus alone is necessary but not sufficient. The subtleties of Poisson's ratio behaviour actually provide a very effective method for showing differences between the theories and demonstrating their ranges of validity. We find that for moderate- to high-porosity materials, none of the analytical theories is accurate and, at present, numerical techniques must be relied upon.

Testing insect movement theory: empirical analysis of pest data routinely collected from agricultural crops

Observations of an insect's movement lead to theory on the insect's flight behaviour and the role of movement in the species' population dynamics. This theory leads to predictions of the way the population changes in time under different conditions. If a hypothesis on movement predicts a specific change in the population, then the hypothesis can be tested against observations of population change. Routine pest monitoring of agricultural crops provides a convenient source of data for studying movement into a region and among fields within a region. Examples of the use of statistical and computational methods for testing hypotheses with such data are presented. The types of questions that can be addressed with these methods and the limitations of pest monitoring data when used for this purpose are discussed. (C) 2002 Elsevier Science B.V. All rights reserved.

Linear ketenimines. Variable structures of C,C-dicyanoketenimines and C,C-bis-sulfonylketenimines

C,C-Dicyanoketenimines 10a-c were generated by flash vacuum thermolysis of ketene NS-acetals 9a-c or by thermal or photochemical decomposition of alpha-azido-,beta-cyanocinnamonitrile 11. In the latter reaction, 3,3-dicyano-2-phenyl-1-azirine 12 is also formed. IR spectroscopy of the keteniminines isolated in Ar matrixes or as neat films, NMR spectroscopy of 10c, and theoretical calculations (B3LYP/6-31G*) demonstrate that these ketenimines have variable geometry, being essentially linear along the CCN-R framework in polar media (neat films and solution), but in the gas phase or Ar matrix they are bent, as is usual for ketenimines. Experiments and calculations agree that a single CN substituent as in 13 is not enough to enforce linearity, and sulfonyl groups are less effective that cyano groups in causing linearity. C,C-Bis(methylsulfonyl)ketenimines 4-5 and a C-cyano-C-(methylsulfonyl)ketenimine 15 are not linear. The compound p-O2NC6H4N=C= C(COOMe)2 previously reported in the literature is probably somewhat linearized along the CCNR moiety. A computational survey (B3LYP/6-31G*) of the inversion barrier at nitrogen indicates that electronegative C-substituents dramatically lower the barrier; this is also true of N-acyl substituents. Increasing polarity causes lower barriers. Although N-alkylbis(methylsulfonyl)ketenimines are not calculated to be linear, the barriers are so low that crystal lattice forces can induce planarity in N-methylbis(methylsulfonyl)ketenimine 3.

Wavelet-based image segment representation

An efficient representation method for arbitrarily shaped image segments is proposed. This method includes a smart way to select wavelet basis to approximate the given image segment, with improved image quality and reduced computational load.

Welding and rheomorphism of phonolitic fallout deposits from the Las Canadas caldera, Tenerife, Canary Islands

The Las Canadas caldera is a nested collapse caldera formed by the successive migration and collapse of shallow magmatic chambers. Among the pyroclastic products of this caldera are phonolitic fallout deposits that crop out in the caldera wall and on the extracaldera slopes. These deposits exhibit an uninterrupted facies gradation from nonwelded to lava-like and record continuous volcanic deposition. Densely welded and lava-like facies result from the extreme attenuation and complete homogenization of juvenile clasts that destroy original clast outlines and any evidence of fallout deposition. Agglutination contributes significantly to the final degree of flattening observed in the welded facies. After deposition, rheomorphic flowage occurs. Emplacement temperatures for one of the welding sequences are calculated from magmatic temperatures and a model of tephra cooling during fallout. Results are 486 degreesC for the nonwelded facies and 740 degreesC for the moderately welded facies. For the same welding sequence, a cooling time between 25 and 54 days is estimated from published experimental and computational data as the possible duration of welding and rheomorphism. Following deposition and agglutination, the lava-like pyroclastic facies had the rheological properties of viscous lavas and flowed down the outer slopes away from the caldera. Some lava-like masses detached from proximal areas to more distal regions. During deposition, the eruptive style evolved from Plinian fallout to fountain-fed spatter deposition. This evolution was accompanied by a decrease in explosive power and a lower height of the eruptive column, which produce higher emplacement temperatures and more effective heat retention of pyroclasts.

Catheter effects in organ perfusion experiments

In a typical isolated organ perfusion experiment, a substance is injected upstream of an organ and then collected at some distance downstream. To reach the organ from the injection site, and then from the organ to the collector, a solute passes through catheters, usually tubes with circular cross-sections. Catheters cause distortion to the concentration-time profile of the perfusion. In this paper, we analyse catheter distribution kinetics from a mathematical point of view, develop the function most suitable for modeling this distribution and successfully apply this function to experimental data. (C) 2002 Academic Press.

Ab initio modelling of basal plane oxidation of graphenes and implications for modelling char combustion

Ab initio calculations have been performed to determine the energetics of oxygen atoms adsorbed onto graphene planes and the possible reaction path extracting carbon atorns in the form of carbon monoxide. Front the energetics it is confirmed that this reaction path will not significantly contribute to the gasification of well ordered carbonaceous chars. Modelling results which explore this limit Lire presented. (C) 2002 Elsevier Science Ltd, All rights reserved.

On the validity of thermogravimetric determination of carbon gasification kinetics

Thermogravimetric analysis has been widely applied in kinetic studies of carbon gasification, with the associated temporal weight change profiles being used to extract kinetic information and to validate gasification models. However the weight change profiles are not always governed by the intrinsic gasification activity because of the effect of chemisorption and its dynamics. In the present work we theoretically determine the criteria under which weight change profiles can be used to determine intrinsic kinetics for CO2 and O2 gasification by examining the region in which the chemisorption dynamics can be assumed pseudo-steady. It is found that the validity of the pseudo-steady assumption depends on the experimental conditions as well as on the initial surface area of carbon. Based on known mechanisms and rate constants an active surface area region is identified within which the steady state assumption is valid and the effect of chemisorption dynamics is negligible. The size of the permissible region is sensitive to the reaction temperature and gas pressure. The results indicate that in some cases the thermogravimetric data should be used with caution in kinetic studies. A large amount of literature on thermogravimetric analyzer determined char gasification kinetics is examined and the importance of chemisorption dynamics for the data assessed.

Structural ordering of coal char during heat treatment and its impact on reactivity

The effect of heat treatment on the structure of an Australian semi-anthracite char was studied in detail in the 850-1150degreesC temperature range using XRD, HRTEM, and electrical resistivity techniques. It was found that the carbon crystallite size in the char does not change significantly during heat treatment in the temperature range studied, for both the raw coal and its ash-free derivative obtained by acid treatment. However, the fraction of the organized carbon in the raw coal chars, determined by XRD, increased with increase of heat treatment time and temperature, while that for the ash-free coal chars remained almost unchanged. This suggests the occurrence of catalytic ordering during heat treatment, supported by the observation that the electrical resistivity of the raw coal chars decreased with heat treatment, while that of the ash-free coal chars did not vary significantly. Further confirmatory evidence was provided by high resolution transmission electron micrographs depicting well-organized carbon layers surrounding iron particles. It is also found that the fraction of organized carbon does not reach unity, but attains an apparent equilibrium value that increases with increase in temperature, providing an apparent heat of ordering of 71.7 kJ mol(-1) in the temperature range studied. Good temperature-independent correlation was found between the electrical resistivity and the organized carbon fraction, indicating that electrical resistivity is indeed structure sensitive. Good correlation was also found between the electrical resistivity and the reactivity of coal char. All these results strongly suggest that the thermal deactivation is the result of a crystallite-perfecting process, which is effectively catalyzed by the inorganic matter in the coal char. Based on kinetic interpretation of the data it is concluded that the process is diffusion controlled, most likely involving transport of iron in the inter-crystallite nanospaces in the temperature range studied. The activation energy of this transport process is found to be very low, at about 11.8 kJ mol(-1), which is corroborated by model-free correlation of the temporal variation of organized carbon fraction as well as electrical resistivity data using the superposition method, and is suggestive of surface transport of iron. (C) 2002 Elsevier Science Ltd. All rights reserved.

Tide-induced water table fluctuations in coastal aquifers bounded by rhythmic shorelines

Previous studies on tidal water table dynamics in unconfined coastal aquifers have focused on the inland propagation of oceanic tides in the cross-shore direction based on the assumption of a straight coastline. Here, two-dimensional analytical solutions are derived to study the effects of rhythmic coastlines on tidal water table fluctuations. The computational results demonstrate that the alongshore variations of the coastline can affect the water table behavior significantly, especially in areas near the centers of the headland and embayment. With the coastline shape effects ignored, traditional analytical solutions may lead to large errors in predicting coastal water table fluctuations or in estimating the aquifer's properties based on these signals. The conditions under which the coastline shape needs to be considered are derived from the new analytical solution.

Finite difference time domain (FDTD) method for modeling the effect of switched gradients on the human body in MRI

Numerical modeling of the eddy currents induced in the human body by the pulsed field gradients in MRI presents a difficult computational problem. It requires an efficient and accurate computational method for high spatial resolution analyses with a relatively low input frequency. In this article, a new technique is described which allows the finite difference time domain (FDTD) method to be efficiently applied over a very large frequency range, including low frequencies. This is not the case in conventional FDTD-based methods. A method of implementing streamline gradients in FDTD is presented, as well as comparative analyses which show that the correct source injection in the FDTD simulation plays a crucial rule in obtaining accurate solutions. In particular, making use of the derivative of the input source waveform is shown to provide distinct benefits in accuracy over direct source injection. In the method, no alterations to the properties of either the source or the transmission media are required. The method is essentially frequency independent and the source injection method has been verified against examples with analytical solutions. Results are presented showing the spatial distribution of gradient-induced electric fields and eddy currents in a complete body model.