Regression analysis of spatial data.

Many of the most interesting questions ecologists ask lead to analyses of spatial data. Yet, perhaps confused by the large number of statistical models and fitting methods available, many ecologists seem to believe this is best left to specialists. Here, we describe the issues that need consideration when analysing spatial data and illustrate these using simulation studies. Our comparative analysis involves using methods including generalized least squares, spatial filters, wavelet revised models, conditional autoregressive models and generalized additive mixed models to estimate regression coefficients from synthetic but realistic data sets, including some which violate standard regression assumptions. We assess the performance of each method using two measures and using statistical error rates for model selection. Methods that performed well included generalized least squares family of models and a Bayesian implementation of the conditional auto-regressive model. Ordinary least squares also performed adequately in the absence of model selection, but had poorly controlled Type I error rates and so did not show the improvements in performance under model selection when using the above methods. Removing large-scale spatial trends in the response led to poor performance. These are empirical results; hence extrapolation of these findings to other situations should be performed cautiously. Nevertheless, our simulation-based approach provides much stronger evidence for comparative analysis than assessments based on single or small numbers of data sets, and should be considered a necessary foundation for statements of this type in future.

Detection of Optimal Models in Parameter Space with Support Vector Machines

The paper proposes an approach aimed at detecting optimal model parameter combinations to achieve the most representative description of uncertainty in the model performance. A classification problem is posed to find the regions of good fitting models according to the values of a cost function. Support Vector Machine (SVM) classification in the parameter space is applied to decide if a forward model simulation is to be computed for a particular generated model. SVM is particularly designed to tackle classification problems in high-dimensional space in a non-parametric and non-linear way. SVM decision boundaries determine the regions that are subject to the largest uncertainty in the cost function classification, and, therefore, provide guidelines for further iterative exploration of the model space. The proposed approach is illustrated by a synthetic example of fluid flow through porous media, which features highly variable response due to the parameter values' combination.

Uncertainty Quantification with Support Vector Regression Prediction Models

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. The paper considers a data driven approach in modelling uncertainty in spatial predictions. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic features and describe stochastic variability and non-uniqueness of spatial properties. It is able to capture and preserve key spatial dependencies such as connectivity, which is often difficult to achieve with two-point geostatistical models. Semi-supervised SVR is designed to integrate various kinds of conditioning data and learn dependences from them. A stochastic semi-supervised SVR model is integrated into a Bayesian framework to quantify uncertainty with multiple models fitted to dynamic observations. The developed approach is illustrated with a reservoir case study. The resulting probabilistic production forecasts are described by uncertainty envelopes.

Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Automatic Mallampati Classification Using Active Appearance Models

Difficult tracheal intubation assessment is an important research topic in anesthesia as failed intubations are important causes of mortality in anesthetic practice. The modified Mallampati score is widely used, alone or in conjunction with other criteria, to predict the difficulty of intubation. This work presents an automatic method to assess the modified Mallampati score from an image of a patient with the mouth wide open. For this purpose we propose an active appearance models (AAM) based method and use linear support vector machines (SVM) to select a subset of relevant features obtained using the AAM. This feature selection step proves to be essential as it improves drastically the performance of classification, which is obtained using SVM with RBF kernel and majority voting. We test our method on images of 100 patients undergoing elective surgery and achieve 97.9% accuracy in the leave-one-out crossvalidation test and provide a key element to an automatic difficult intubation assessment system.

Potential evidence of parasite avoidance in an avian malarial vector

Epidemiological studies of malaria or other vector-transmitted diseases often consider vectors as passive actors in the complex life cycle of the parasites, assuming that vector populations are homogeneous and vertebrate hosts are equally susceptible to being infected during their lifetime. However, some studies based on both human and rodent malaria systems found that mosquito vectors preferentially selected infected vertebrate hosts. This subject has been scarcely investigated in avian malaria models and even less in wild animals using natural host-parasite associations. We investigated whether the malaria infection status of wild great tits, Parus major, played a role in host selection by the mosquito vector Culex pipiens. Pairs of infected and uninfected birds were tested in a dual-choice olfactometer to assess their attractiveness to the mosquitoes. Plasmodium-infected birds attracted significantly fewer mosquitoes than the uninfected ones, which suggest that avian malaria parasites alter hosts' odours involved in vector orientation. Reaction time of the mosquitoes, that is, the time taken to select a host, and activation of mosquitoes, defined as the proportion of individuals flying towards one of the hosts, were not affected by the bird's infection status. The importance of these behavioural responses for the vector is discussed in light of recent advances in related or similar model systems.

Uncertainty Quantification Of A Semi-Supervised Support Vector Regression Reservoir Model

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. Recent advances in machine learning offer a novel approach to model spatial distribution of petrophysical properties in complex reservoirs alternative to geostatistics. The approach is based of semisupervised learning, which handles both ?labelled? observed data and ?unlabelled? data, which have no measured value but describe prior knowledge and other relevant data in forms of manifolds in the input space where the modelled property is continuous. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic geological features and describe stochastic variability and non-uniqueness of spatial properties. On the other hand, it is able to capture and preserve key spatial dependencies such as connectivity of high permeability geo-bodies, which is often difficult in contemporary petroleum reservoir studies. Semi-supervised SVR as a data driven algorithm is designed to integrate various kind of conditioning information and learn dependences from it. The semi-supervised SVR model is able to balance signal/noise levels and control the prior belief in available data. In this work, stochastic semi-supervised SVR geomodel is integrated into Bayesian framework to quantify uncertainty of reservoir production with multiple models fitted to past dynamic observations (production history). Multiple history matched models are obtained using stochastic sampling and/or MCMC-based inference algorithms, which evaluate posterior probability distribution. Uncertainty of the model is described by posterior probability of the model parameters that represent key geological properties: spatial correlation size, continuity strength, smoothness/variability of spatial property distribution. The developed approach is illustrated with a fluvial reservoir case. The resulting probabilistic production forecasts are described by uncertainty envelopes. The paper compares the performance of the models with different combinations of unknown parameters and discusses sensitivity issues.

A comparison between joint coordinate system and attitude vector for multi-segment foot kinematics.

The joint angles of multi-segment foot models have been primarily described using two mathematical methods: the joint coordinate system and the attitude vector. This study aimed to determine whether the angles obtained through these two descriptors are comparable, and whether these descriptors have similar sensitivity to experimental errors. Six subjects walked eight times on an instrumented walkway while the joint angles among shank, hindfoot, medial forefoot, and lateral forefoot were measured. The angles obtained using both descriptors and their sensitivity to experimental errors were compared. There was no overall significant difference between the ranges of motion obtained using both descriptors. However, median differences of more than 6° were noticed for the medial-lateral forefoot joint. For all joints and rotation planes, both descriptors provided highly similar angle patterns (median correlation coefficient: R>0.90), except for the medial-lateral forefoot angle in the transverse plane (median R=0.77). The joint coordinate system was significantly more sensitive to anatomical landmarks misplacement errors. However, the absolute differences of sensitivity were small relative to the joints ranges of motion. In conclusion, the angles obtained using these two descriptors were not identical, but were similar for at least the shank-hindfoot and hindfoot-medial forefoot joints. Therefore, the angle comparison across descriptors is possible for these two joints. Comparison should be done more carefully for the medial-lateral forefoot joint. Moreover, despite different sensitivities to experimental errors, the effects of the experimental errors on the angles were small for both descriptors suggesting that both descriptors can be considered for multi-segment foot models.

Personalized modeling for drug concentration prediction using Support Vector Machine

Building a personalized model to describe the drug concentration inside the human body for each patient is highly important to the clinical practice and demanding to the modeling tools. Instead of using traditional explicit methods, in this paper we propose a machine learning approach to describe the relation between the drug concentration and patients' features. Machine learning has been largely applied to analyze data in various domains, but it is still new to personalized medicine, especially dose individualization. We focus mainly on the prediction of the drug concentrations as well as the analysis of different features' influence. Models are built based on Support Vector Machine and the prediction results are compared with the traditional analytical models.

Retinal degeneration progression changes lentiviral vector cell targeting in the retina.

In normal mice, the lentiviral vector (LV) is very efficient to target the RPE cells, but transduces retinal neurons well only during development. In the present study, the tropism of LV has been investigated in the degenerating retina of mice, knowing that the retina structure changes during degeneration. We postulated that the viral transduction would be increased by the alteration of the outer limiting membrane (OLM). Two different LV pseudotypes were tested using the VSVG and the Mokola envelopes, as well as two animal models of retinal degeneration: light-damaged Balb-C and Rhodopsin knockout (Rho-/-) mice. After light damage, the OLM is altered and no significant increase of the number of transduced photoreceptors can be obtained with a LV-VSVG-Rhop-GFP vector. In the Rho-/- mice, an alteration of the OLM was also observed, but the possibility of transducing photoreceptors was decreased, probably by ongoing gliosis. The use of a ubiquitous promoter allows better photoreceptor transduction, suggesting that photoreceptor-specific promoter activity changes during late stages of photoreceptor degeneration. However, the number of targeted photoreceptors remains low. In contrast, LV pseudotyped with the Mokola envelope allows a wide dispersion of the vector into the retina (corresponding to the injection bleb) with preferential targeting of Müller cells, a situation which does not occur in the wild-type retina. Mokola-pseudotyped lentiviral vectors may serve to engineer these glial cells to deliver secreted therapeutic factors to a diseased area of the retina.

An in vivo ultrahigh field 14.1 T (1) H-MRS study on 6-OHDA and α-synuclein-based rat models of Parkinson's disease: GABA as an early disease marker.

The detection of Parkinson's disease (PD) in its preclinical stages prior to outright neurodegeneration is essential to the development of neuroprotective therapies and could reduce the number of misdiagnosed patients. However, early diagnosis is currently hampered by lack of reliable biomarkers. (1) H magnetic resonance spectroscopy (MRS) offers a noninvasive measure of brain metabolite levels that allows the identification of such potential biomarkers. This study aimed at using MRS on an ultrahigh field 14.1 T magnet to explore the striatal metabolic changes occurring in two different rat models of the disease. Rats lesioned by the injection of 6-hydroxydopamine (6-OHDA) in the medial-forebrain bundle were used to model a complete nigrostriatal lesion while a genetic model based on the nigral injection of an adeno-associated viral (AAV) vector coding for the human α-synuclein was used to model a progressive neurodegeneration and dopaminergic neuron dysfunction, thereby replicating conditions closer to early pathological stages of PD. MRS measurements in the striatum of the 6-OHDA rats revealed significant decreases in glutamate and N-acetyl-aspartate levels and a significant increase in GABA level in the ipsilateral hemisphere compared with the contralateral one, while the αSyn overexpressing rats showed a significant increase in the GABA striatal level only. Therefore, we conclude that MRS measurements of striatal GABA levels could allow for the detection of early nigrostriatal defects prior to outright neurodegeneration and, as such, offers great potential as a sensitive biomarker of presymptomatic PD.

Retinal degeneration progression changes lentiviral vector cell targeting in the retina

Ce travail de thèse a été réalisé au sein de l'Unité de Thérapie Génique et Biologie des Cellules Souches de l'Hôpital Jules- Gonin dans le Service d'Ophtalmologie de l'Université de Lausanne. Ce laboratoire recherche des solutions thérapeutiques pour des maladies dégénératives et incurables de la rétine comme les rétinites pigmentaires (RP). Ayant déjà montré certains résultats dans le domaine, la thérapie génique a été notre outil pour ce travail. Cette méthode se base sur le principe de remplacer un gène déficient par sa copie normale, en transportant celle-ci au coeur même du noyau par un vecteur. Il existe à l'heure actuelle différents vecteurs. Un des plus efficaces est un vecteur viral non-réplicatif : le lentivirus, dérivé de HIV-1. Celui-ci a la capacité d'intégrer le génome de la cellule cible, lui conférant ainsi un nouveau matériel génétique. Notre but a été d'établir le tropisme du lentivirus dans une rétine en dégénérescence. Ce lentivirus est connu pour transduire efficacement les cellules de l'épithélium pigmentaire rétinien dans l'oeil adulte sain, ainsi que celles de la neurorétine, mais ce, uniquement durant le développement. On sait aussi que le vecteur lentiviral présente un tropisme différent selon les enveloppes dont il est muni ; par exemple, le lentivirus avec une enveloppe Mokola est connu pour transduire les cellules gliales du système nerveux central. La rétine qui dégénère montre quant à elle des changements de sa structure qui pourraient influencer la diffusion du vecteur et/ou son tropisme. Le postulat de base a été le suivant : chez l'adulte, la transduction des neurones de la rétine via le lentivirus pourrait être facilitée par l'altération de la membrane limitante externe induite par la dégénérescence (meilleure pénétrance du virus). D'un point de vue technique, nous avons utilisé deux types distincts de modèles murins de dégénérescence rétinienne : des souris Balb/C soumises à une dose toxique de lumière et les souris Rhodopsin knockout, animaux génétiquement modifiés. Comme vecteur viral, nous avons employé deux différents pseudotypes de lentivirus (caractérisés par les enveloppes virales) avec différents promoteurs (séquence d'ADN qui initie la transduction et confère la spécificité d'expression d'un gène). En changeant l'enveloppe et le promoteur, nous avons essayé de trouver la meilleure combinaison pour augmenter l'affinité du vecteur vis-à-vis des photorécepteurs d'abord, puis vis-à-vis d'autres cellules de la rétine. Nos résultats ont montré que la membrane limitante externe est effectivement altérée chez les deux modèles de dégénérescence, mais que cette modification ne favorise pas la transduction des photorécepteurs lorsqu'on utilise un vecteur lentiviral contenant une enveloppe VSVG et un promoteur photorécepteur-spécifique ou ubiquitaire. En effet, une forte réaction gliale a été observée. Par contre, en utilisant le lentivirus avec une enveloppe Mokola et un promoteur ubiquitaire, nous avons constaté une très bonne transduction au niveau des cellules de Millier dans la rétine en dégénérescence, phénomène non observé chez les souris sauvages. Ce travail a donc permis de trouver un vecteur viral efficace pour atteindre et transduire les cellules de Miiller, ceci seulement pendant la dégénérescence de la rétine. Ces cellules, une fois transduites, pourraient être utilisées pour sécréter dans la rétine des agents thérapeutiques tels que des facteurs neurotrophiques pour soutenir la survie des photorécepteurs ou des facteurs anti-angiogéniques pour prévenir la néo-vascularisation lors de diabète ou de dégénérescence maculaire liée à l'âge. - In normal mice, the lentiviral vector (LV) is very efficient to target the RPE cells, but transduces retinal neurons well only during development. In the present study, the tropism of LV has been investigated in the degenerating retina of mice, knowing that the retina structure changes during degeneration. We postulated that the viral transduction would be increased by the alteration of the iuter limiting membrane (OLM). Two different LV pseudotypes were tested using the VSVG arid the Mokola envelopes, as well as two animal models of retinal degeneration: light-damaged Balb-C and Rhodopsin knockout (Rho-/-) mice. After light damage, the OLM is altered and no significant increase of the number of transduced photoreceptors can be obtained with a LV-VSVG-Rhop-GFP vector. In the Rho-/- mice, an altération of the OLM was also observed, but the possibility of transducing photoreceptors was decreased, probably by ongoing gliosis. The use of a ubiquitous promoter allows better photoreceptor transduction, suggesting that photoreceptór-specific promoter activity change during late stages of photoreceptor degeneration. However, the number of targeted photoreceptors remains low. In contrast, LV pseudotyped with the tfokola envelope allows a wide dispersion of the ctor into the retina (corresponding to the injection bleb) with preferential targeting of Muller cells, a situation Mc\ does ot occur in the wild- type retina. Mokola-pseudotyped lentiviral vectors may serve to engineer these glial cells to deliver secreted therapeutic factors to a diseased area of the retina.

Monitoring network optimisation for spatial data classification using support vector machines

The paper presents a novel method for monitoring network optimisation, based on a recent machine learning technique known as support vector machine. It is problem-oriented in the sense that it directly answers the question of whether the advised spatial location is important for the classification model. The method can be used to increase the accuracy of classification models by taking a small number of additional measurements. Traditionally, network optimisation is performed by means of the analysis of the kriging variances. The comparison of the method with the traditional approach is presented on a real case study with climate data.

Modelling ecological niches with support vector machines

1. The ecological niche is a fundamental biological concept. Modelling species' niches is central to numerous ecological applications, including predicting species invasions, identifying reservoirs for disease, nature reserve design and forecasting the effects of anthropogenic and natural climate change on species' ranges. 2. A computational analogue of Hutchinson's ecological niche concept (the multidimensional hyperspace of species' environmental requirements) is the support of the distribution of environments in which the species persist. Recently developed machine-learning algorithms can estimate the support of such high-dimensional distributions. We show how support vector machines can be used to map ecological niches using only observations of species presence to train distribution models for 106 species of woody plants and trees in a montane environment using up to nine environmental covariates. 3. We compared the accuracy of three methods that differ in their approaches to reducing model complexity. We tested models with independent observations of both species presence and species absence. We found that the simplest procedure, which uses all available variables and no pre-processing to reduce correlation, was best overall. Ecological niche models based on support vector machines are theoretically superior to models that rely on simulating pseudo-absence data and are comparable in empirical tests. 4. Synthesis and applications. Accurate species distribution models are crucial for effective environmental planning, management and conservation, and for unravelling the role of the environment in human health and welfare. Models based on distribution estimation rather than classification overcome theoretical and practical obstacles that pervade species distribution modelling. In particular, ecological niche models based on machine-learning algorithms for estimating the support of a statistical distribution provide a promising new approach to identifying species' potential distributions and to project changes in these distributions as a result of climate change, land use and landscape alteration.

Rpe65-Gene Transfer Using an Integration-Deficient Lentiviral Vector

Purpose: We previously demonstrated efficient retinal rescue of RPE65 mouse models (Rpe65-/- (Bemelmans et al, 2006) and Rpe65R91W/R91W mice) using a HIV1-derived lentiviral vector encoding for the mouse RPE65 cDNA. In order to optimize a lentiviral vector as an alternative tool for RPE65-derived Leber Congenital Amaurosis clinical trials, we evaluated the efficiency of an integration-deficient lentiviral vector (IDLV) encoding the human RPE65 cDNA to restore retinal function in the Rpe65R91W/R91W mice. Methods: An HIV-1-derived lentiviral vector expressing either the hrGFPII or the human Rpe65 cDNA under the control of a 0.8 kb fragment of the human Rpe65 promoter (R0.8) was produced by transient transfection of 293T cells. A LQ-integrase mutant was used to generate the IDLV vectors. IDLV-R0.8-hRPE65 or hrGFPII were injected subretinally into 1 month-old Rpe65R91W/R91W mice. Functional rescue was assessed by ERG (1 and 3 months post-injection) and cone survival by immunohistology. Results: An increased light sensitivity was detected by scotopic ERG in animals injected with IDLV-R0.8-hRPE65 compared to hrGFPII-treated animals or untreated mice. However the improvement was delayed compared to integration-proficient LV and observed at 3 months but not 1 month post-injection. Immunolabelling of cone markers showed an increased number of cones in the transduced area compared to control groups. Conclusions: The IDLV-R0.8-hRPE65 vectors allow retinal improvement in the Rpe65R91W/R91W mice. Both rod function and cone survival were demonstrated even if there is a delay in the rescue as assessed by scotopic ERG. Integration-deficient vectors minimize insertional mutagenesis and thus are safer candidates for human application. Further experiments using large animals are now needed to validate correct gene transfer and expression of the RPE65 gene as well as tolerance of the vector after subretinal injection before envisaging a clinical trial application.