Mapping of Environmental Data Using Kernel-Based Methods

Recently, kernel-based Machine Learning methods have gained great popularity in many data analysis and data mining fields: pattern recognition, biocomputing, speech and vision, engineering, remote sensing etc. The paper describes the use of kernel methods to approach the processing of large datasets from environmental monitoring networks. Several typical problems of the environmental sciences and their solutions provided by kernel-based methods are considered: classification of categorical data (soil type classification), mapping of environmental and pollution continuous information (pollution of soil by radionuclides), mapping with auxiliary information (climatic data from Aral Sea region). The promising developments, such as automatic emergency hot spot detection and monitoring network optimization are discussed as well.

Comparison of Energy Minimization Methods for 3-D Brain Tissue Classification

This paper presents 3-D brain tissue classificationschemes using three recent promising energy minimizationmethods for Markov random fields: graph cuts, loopybelief propagation and tree-reweighted message passing.The classification is performed using the well knownfinite Gaussian mixture Markov Random Field model.Results from the above methods are compared with widelyused iterative conditional modes algorithm. Theevaluation is performed on a dataset containing simulatedT1-weighted MR brain volumes with varying noise andintensity non-uniformities. The comparisons are performedin terms of energies as well as based on ground truthsegmentations, using various quantitative metrics.

Fracture network characterization using hydrological and geophysical data

Les décisions de gestion des eaux souterraines doivent souvent être justiffées par des modèles quantitatifs d'aquifères qui tiennent compte de l'hétérogénéité des propriétés hydrauliques. Les aquifères fracturés sont parmi les plus hétérogènes et très difficiles à étudier. Dans ceux-ci, les fractures connectées, d'ouverture millimètrique, peuvent agir comme conducteurs hydrauliques et donc créer des écoulements très localisés. Le manque général d'informations sur la distribution spatiale des fractures limite la possibilité de construire des modèles quantitatifs de flux et de transport. Les données qui conditionnent les modèles sont généralement spatialement limitées, bruitées et elles ne représentent que des mesures indirectes de propriétés physiques. Ces limitations aux données peuvent être en partie surmontées en combinant différents types de données, telles que les données hydrologiques et de radar à pénétration de sol plus commun ément appelé géoradar. L'utilisation du géoradar en forage est un outil prometteur pour identiffer les fractures individuelles jusqu'à quelques dizaines de mètres dans la formation. Dans cette thèse, je développe des approches pour combiner le géoradar avec les données hydrologiques affn d'améliorer la caractérisation des aquifères fracturés. Des investigations hydrologiques intensives ont déjà été réalisées à partir de trois forage adjacents dans un aquifère cristallin en Bretagne (France). Néanmoins, la dimension des fractures et la géométrie 3-D des fractures conductives restaient mal connue. Affn d'améliorer la caractérisation du réseau de fractures je propose dans un premier temps un traitement géoradar avancé qui permet l'imagerie des fractures individuellement. Les résultats montrent que les fractures perméables précédemment identiffées dans les forages peuvent être caractérisées géométriquement loin du forage et que les fractures qui ne croisent pas les forages peuvent aussi être identiffées. Les résultats d'une deuxième étude montrent que les données géoradar peuvent suivre le transport d'un traceur salin. Ainsi, les fractures qui font partie du réseau conductif et connecté qui dominent l'écoulement et le transport local sont identiffées. C'est la première fois que le transport d'un traceur salin a pu être imagé sur une dizaines de mètres dans des fractures individuelles. Une troisième étude conffrme ces résultats par des expériences répétées et des essais de traçage supplémentaires dans différentes parties du réseau local. En outre, la combinaison des données de surveillance hydrologique et géoradar fournit la preuve que les variations temporelles d'amplitude des signaux géoradar peuvent nous informer sur les changements relatifs de concentrations de traceurs dans la formation. Par conséquent, les données géoradar et hydrologiques sont complémentaires. Je propose ensuite une approche d'inversion stochastique pour générer des modèles 3-D de fractures discrètes qui sont conditionnés à toutes les données disponibles en respectant leurs incertitudes. La génération stochastique des modèles conditionnés par géoradar est capable de reproduire les connexions hydrauliques observées et leur contribution aux écoulements. L'ensemble des modèles conditionnés fournit des estimations quantitatives des dimensions et de l'organisation spatiale des fractures hydrauliquement importantes. Cette thèse montre clairement que l'imagerie géoradar est un outil utile pour caractériser les fractures. La combinaison de mesures géoradar avec des données hydrologiques permet de conditionner avec succès le réseau de fractures et de fournir des modèles quantitatifs. Les approches présentées peuvent être appliquées dans d'autres types de formations rocheuses fracturées où la roche est électriquement résistive.

Value and depreciation of mineral resources over the very long run: An empirical contrast of different methods

The paper contrasts empirically the results of alternative methods for estimating thevalue and the depreciation of mineral resources. The historical data of Mexico andVenezuela, covering the period 1920s-1980s, is used to contrast the results of severalmethods. These are the present value, the net price method, the user cost method andthe imputed income method. The paper establishes that the net price and the user costare not competing methods as such, but alternative adjustments to different scenariosof closed and open economies. The results prove that the biases of the methods, ascommonly described in the theoretical literature, only hold under the most restrictedscenario of constant rents over time. It is argued that the difference between what isexpected to happen and what actually did happen is for the most part due to a missingvariable, namely technological change. This is an important caveat to therecommendations made based on these models.

Exact and approximate stepdown methods for multiple hypothesis testing

Consider the problem of testing k hypotheses simultaneously. In this paper,we discuss finite and large sample theory of stepdown methods that providecontrol of the familywise error rate (FWE). In order to improve upon theBonferroni method or Holm's (1979) stepdown method, Westfall and Young(1993) make eective use of resampling to construct stepdown methods thatimplicitly estimate the dependence structure of the test statistics. However,their methods depend on an assumption called subset pivotality. The goalof this paper is to construct general stepdown methods that do not requiresuch an assumption. In order to accomplish this, we take a close look atwhat makes stepdown procedures work, and a key component is a monotonicityrequirement of critical values. By imposing such monotonicity on estimatedcritical values (which is not an assumption on the model but an assumptionon the method), it is demonstrated that the problem of constructing a validmultiple test procedure which controls the FWE can be reduced to the problemof contructing a single test which controls the usual probability of a Type 1error. This reduction allows us to draw upon an enormous resamplingliterature as a general means of test contruction.

Comparison of different methods for thin section EM analysis of Mycobacterium smegmatis.

Bacteria are generally difficult specimens to prepare for conventional resin section electron microscopy and mycobacteria, with their thick and complex cell envelope layers being especially prone to artefacts. Here we made a systematic comparison of different methods for preparing Mycobacterium smegmatis for thin section electron microscopy analysis. These methods were: (1) conventional preparation by fixatives and epoxy resins at ambient temperature. (2) Tokuyasu cryo-section of chemically fixed bacteria. (3) rapid freezing followed by freeze substitution and embedding in epoxy resin at room temperature or (4) combined with Lowicryl HM20 embedding and ultraviolet (UV) polymerization at low temperature and (5) CEMOVIS, or cryo electron microscopy of vitreous sections. The best preservation of bacteria was obtained with the cryo electron microscopy of vitreous sections method, as expected, especially with respect to the preservation of the cell envelope and lipid bodies. By comparison with cryo electron microscopy of vitreous sections both the conventional and Tokuyasu methods produced different, undesirable artefacts. The two different types of freeze-substitution protocols showed variable preservation of the cell envelope but gave acceptable preservation of the cytoplasm, but not lipid bodies, and bacterial DNA. In conclusion although cryo electron microscopy of vitreous sections must be considered the 'gold standard' among sectioning methods for electron microscopy, because it avoids solvents and stains, the use of optimally prepared freeze substitution also offers some advantages for ultrastructural analysis of bacteria.

Methods for integration site distribution analyses in animal cell genomes.

The question of where retroviral DNA becomes integrated in chromosomes is important for understanding (i) the mechanisms of viral growth, (ii) devising new anti-retroviral therapy, (iii) understanding how genomes evolve, and (iv) developing safer methods for gene therapy. With the completion of genome sequences for many organisms, it has become possible to study integration targeting by cloning and sequencing large numbers of host-virus DNA junctions, then mapping the host DNA segments back onto the genomic sequence. This allows statistical analysis of the distribution of integration sites relative to the myriad types of genomic features that are also being mapped onto the sequence scaffold. Here we present methods for recovering and analyzing integration site sequences.

Isolating DNA from sexual assault cases: a comparison of standard methods with a nuclease-based approach.

AbstractText BACKGROUND: Profiling sperm DNA present on vaginal swabs taken from rape victims often contributes to identifying and incarcerating rapists. Large amounts of the victim's epithelial cells contaminate the sperm present on swabs, however, and complicate this process. The standard method for obtaining relatively pure sperm DNA from a vaginal swab is to digest the epithelial cells with Proteinase K in order to solubilize the victim's DNA, and to then physically separate the soluble DNA from the intact sperm by pelleting the sperm, removing the victim's fraction, and repeatedly washing the sperm pellet. An alternative approach that does not require washing steps is to digest with Proteinase K, pellet the sperm, remove the victim's fraction, and then digest the residual victim's DNA with a nuclease. METHODS: The nuclease approach has been commercialized in a product, the Erase Sperm Isolation Kit (PTC Labs, Columbia, MO, USA), and five crime laboratories have tested it on semen-spiked female buccal swabs in a direct comparison with their standard methods. Comparisons have also been performed on timed post-coital vaginal swabs and evidence collected from sexual assault cases. RESULTS: For the semen-spiked buccal swabs, Erase outperformed the standard methods in all five laboratories and in most cases was able to provide a clean male profile from buccal swabs spiked with only 1,500 sperm. The vaginal swabs taken after consensual sex and the evidence collected from rape victims showed a similar pattern of Erase providing superior profiles. CONCLUSIONS: In all samples tested, STR profiles of the male DNA fractions obtained with Erase were as good as or better than those obtained using the standard methods.

Predição da hidroquímica da água (qualidade de água) usando sondagens electromagnéticas transientes (TEM)

This work aims the applicability of the Transient electromagnetic method at an arid and semiarid environmental condition in the Santiago Island – Cape Verde. Some seashore areas of this island show an increasing salt contamination of the groundwater. The main objective of present work is to relate this water-quality condition with parameters taken from the transient sounding’s data. In this context, transient soundings have been acquired from 2005 through 2009, at several chosen valleys near the sea, in a mean rate of one field campaign each year. The first phase of this work was the understanding of the geophysical method details, problems and applicability, as the chosen and acquired equipment was the first one to be permanently available to the Portuguese geosciences community. This first phase was also accomplished with field tests. Interpretation of the transient sounding’s data curves were done by application of 1-D inversion methods already developed and published, as also with quasi 2-D and quasi 3-D inversion algorithms, where applicability was feasible. This was the second phase. The 2-D and 3-D approximation results are satisfactory and promising; although a higher spatial sounding’s density should certainly allow for better results. At phase three, these results have been compared against the available lithologic, hydrologic and hydrochemical data, in the context of Santiago’s island settings. The analyses of these merged data showed that two distinct origins for the observed inland groundwater salinity are possible; seashore shallow mixing with contemporary seawater and mixing with a deep and older salty layer from up flow groundwater. Relations between the electric resistivity and the salt water content distribution were found for the surveyed areas. To this environment condition, the electromagnetic transient method proved to be a reliable and powerful technique. The groundwater quality can be accessed beyond the few available watershed points, which have an uneven distribution.

Dynamic graphics of parametrically linked multivariate methods used in compositional data analysis

Many multivariate methods that are apparently distinct can be linked by introducing oneor more parameters in their definition. Methods that can be linked in this way arecorrespondence analysis, unweighted or weighted logratio analysis (the latter alsoknown as "spectral mapping"), nonsymmetric correspondence analysis, principalcomponent analysis (with and without logarithmic transformation of the data) andmultidimensional scaling. In this presentation I will show how several of thesemethods, which are frequently used in compositional data analysis, may be linkedthrough parametrizations such as power transformations, linear transformations andconvex linear combinations. Since the methods of interest here all lead to visual mapsof data, a "movie" can be made where where the linking parameter is allowed to vary insmall steps: the results are recalculated "frame by frame" and one can see the smoothchange from one method to another. Several of these "movies" will be shown, giving adeeper insight into the similarities and differences between these methods.

Genome-Wide Prediction Methods in Highly Diverse and Heterozygous Species: Proof-of-Concept through Simulation in Grapevine.

Nowadays, genome-wide association studies (GWAS) and genomic selection (GS) methods which use genome-wide marker data for phenotype prediction are of much potential interest in plant breeding. However, to our knowledge, no studies have been performed yet on the predictive ability of these methods for structured traits when using training populations with high levels of genetic diversity. Such an example of a highly heterozygous, perennial species is grapevine. The present study compares the accuracy of models based on GWAS or GS alone, or in combination, for predicting simple or complex traits, linked or not with population structure. In order to explore the relevance of these methods in this context, we performed simulations using approx 90,000 SNPs on a population of 3,000 individuals structured into three groups and corresponding to published diversity grapevine data. To estimate the parameters of the prediction models, we defined four training populations of 1,000 individuals, corresponding to these three groups and a core collection. Finally, to estimate the accuracy of the models, we also simulated four breeding populations of 200 individuals. Although prediction accuracy was low when breeding populations were too distant from the training populations, high accuracy levels were obtained using the sole core-collection as training population. The highest prediction accuracy was obtained (up to 0.9) using the combined GWAS-GS model. We thus recommend using the combined prediction model and a core-collection as training population for grapevine breeding or for other important economic crops with the same characteristics.

The application of NMR and MS methods for detection of adulteration of wine, fruit juices, and olive oil. A review

This review covers two important techniques, high resolution nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS), used to characterize food products and detect possible adulteration of wine, fruit juices, and olive oil, all important products of the Mediterranean Basin. Emphasis is placed on the complementary use of SNIF-NMR (site-specific natural isotopic fractionation nuclear magnetic resonance) and IRMS (isotope-ratio mass spectrometry) in association with chemometric methods for detecting the adulteration.

Variance reduction methods for simulation of densities on Wiener space

We develop a general error analysis framework for the Monte Carlo simulationof densities for functionals in Wiener space. We also study variancereduction methods with the help of Malliavin derivatives. For this, wegive some general heuristic principles which are applied to diffusionprocesses. A comparison with kernel density estimates is made.