Consistently replicating locus linked to migraine on 10q22-q23

Here, we present the results of two genome-wide scans in two diverse populations in which a consistent use of recently introduced migraine-phenotyping methods detects and replicates a locus on 10q22-q23, with an additional independent replication. No genetic variants have been convincingly established in migraine, and although several loci have been reported, none of them has been consistently replicated. We employed the three known migraine-phenotyping methods (clinical end diagnosis, latent-class analysis, and trait-component analysis) with robust multiple testing correction in a large sample set of 1675 individuals from 210 migraine families from Finland and Australia. Genome-wide multipoint linkage analysis that used the Kong and Cox exponential model in Finns detected a locus on 10q22-q23 with highly significant evidence of linkage (LOD 7.68 at 103 cM in female-specific analysis). The Australian sample showed a LOD score of 3.50 at the same locus (100 cM), as did the independent Finnish replication study (LOD score 2.41, at 102 cM). In addition, four previously reported loci on 8q21, 14q21, 18q12, and Xp21 were also replicated. A shared-segment analysis of 10q22-q23 linked Finnish families identified a 1.6-9.5 cM segment, centered on 101 cM, which shows in-family homology in 95% of affected Finns. This region was further studied with 1323 SNPs. Although no significant association was observed, four regions warranting follow-up studies were identified. These results support the use of symptomology-based phenotyping in migraine and suggest that the 10q22-q23 locus probably contains one or more migraine susceptibility variants.

Land leasing, land degradation and agricultural productivity in Finland

This study analysed whether the land tenure insecurity problem has led to a decline in long-term land improvements (liming and phosphorus fertilization) under the Common Agricultural Policy (CAP) and Nordic production conditions in European Union (EU) countries such as Finland. The results suggests that under traditional cash lease contracts, which are encouraged by the existing land leasing regulations and agricultural subsidy programs, the land tenure insecurity problem on leased land reduces land improvements that have a long pay-back period. In particular, soil pH was found to be significantly lower on land cultivated under a lease contract compared to land owned by the farmers themselves. The results also indicate that land improvements could not be reversed by land markets, because land owners would otherwise have carried out land improvements even if not farming by themselves. To reveal the causality between land tenure and land improvements, the dynamic optimisation problem was solved by a stochastic dynamic programming routine with known parameters for one-period returns and transition equations. The model parameters represented Finnish soil quality and production conditions. The decision rules were solved for alternative likelihood scenarios over the continuation of the fixed-term lease contract. The results suggest that as the probability of non-renewal of the lease contract increases, farmers quickly reduce investments in irreversible land improvements and, thereafter, yields gradually decline. The simulations highlighted the observed trends of a decline in land improvements on land parcels that are cultivated under lease contracts. Land tenure has resulted in the neglect of land improvement in Finland. This study aimed to analyze whether these challenges could be resolved by a tax policy that encourages land sales. Using Finnish data, real estate tax and a temporal relaxation on the taxation of capital gains showed some potential for the restructuring of land ownership. Potential sellers who could not be revealed by traditional logit models were identified with the latent class approach. Those landowners with an intention to sell even without a policy change were sensitive to temporal relaxation in the taxation of capital gains. In the long term, productivity and especially productivity growth are necessary conditions for the survival of farms and the food industry in Finland. Technical progress was found to drive the increase in productivity. The scale had only a moderate effect and for the whole study period (1976–2006) the effect was close to zero. Total factor productivity (TFP) increased, depending on the model, by 0.6–1.7% per year. The results demonstrated that the increase in productivity was hindered by the policy changes introduced in 1995. It is also evidenced that the increase in land leasing is connected to these policy changes. Land institutions and land tenure questions are essential in agricultural and rural policies on all levels, from local to international. Land ownership and land titles are commonly tied to fundamental political, economic and social questions. A fair resolution calls for innovative and new solutions both on national and international levels. However, this seems to be a problem when considering the application of EU regulations to member states inheriting divergent landownership structures and farming cultures. The contribution of this study is in describing the consequences of fitting EU agricultural policy to Finnish agricultural land tenure conditions and heritage.

Latent variable models for neural data analysis

The brain is perhaps the most complex system to have ever been subjected to rigorous scientific investigation. The scale is staggering: over 10^11 neurons, each making an average of 10^3 synapses, with computation occurring on scales ranging from a single dendritic spine, to an entire cortical area. Slowly, we are beginning to acquire experimental tools that can gather the massive amounts of data needed to characterize this system. However, to understand and interpret these data will also require substantial strides in inferential and statistical techniques. This dissertation attempts to meet this need, extending and applying the modern tools of latent variable modeling to problems in neural data analysis.

It is divided into two parts. The first begins with an exposition of the general techniques of latent variable modeling. A new, extremely general, optimization algorithm is proposed - called Relaxation Expectation Maximization (REM) - that may be used to learn the optimal parameter values of arbitrary latent variable models. This algorithm appears to alleviate the common problem of convergence to local, sub-optimal, likelihood maxima. REM leads to a natural framework for model size selection; in combination with standard model selection techniques the quality of fits may be further improved, while the appropriate model size is automatically and efficiently determined. Next, a new latent variable model, the mixture of sparse hidden Markov models, is introduced, and approximate inference and learning algorithms are derived for it. This model is applied in the second part of the thesis.

The second part brings the technology of part I to bear on two important problems in experimental neuroscience. The first is known as spike sorting; this is the problem of separating the spikes from different neurons embedded within an extracellular recording. The dissertation offers the first thorough statistical analysis of this problem, which then yields the first powerful probabilistic solution. The second problem addressed is that of characterizing the distribution of spike trains recorded from the same neuron under identical experimental conditions. A latent variable model is proposed. Inference and learning in this model leads to new principled algorithms for smoothing and clustering of spike data.

Desenvolvimento e validação de modelo preditivo e avaliação de testes de diagnóstico por classe latente para o parasitismo por L. chagasi em cães atendidos no Hospital Veterinário Universitário da UFPI, Teresina

A leishmaniose visceral americana (LVA) é uma doença em expansão no Brasil, para a qual se dispõem de poucas, e aparentemente ineficientes, estratégias de controle. Um dos grandes problemas para a contenção da leishmaniose visceral americana é a falta de um método acurado de identificação dos cães infectados, considerados os principais reservatórios da doença no meio urbano. Neste sentido, a caracterização de marcadores clínico-laboratoriais da infecção neste reservatório e a avaliação mais adequada do desempenho de testes para diagnóstico da infecção podem contribuir para aumentar a efetividade das estratégias de controle da LVA. Com isso, o presente estudo tem dois objetivos principais: (1) desenvolver e validar um modelo de predição para o parasitismo por Leishmania chagasi em cães, baseado em resultados de testes sorológicos e sinais clínicos e (2) avaliar a sensibilidade e especificidade de critérios clínicos, sorológicos e parasitológicos para detecção de infecção canina por L. chagasi mediante análise de classe latente. O primeiro objetivo foi desenvolvido a partir de estudo em que foram obtidos dados de exames clínico, sorológico e parasitológico de todos os cães, suspeitos ou não de LVA, atendidos no Hospital Veterinário Universitário da Universidade Federal do Piauí (HVU-UFPI), em Teresina, nos anos de 2003 e 2004, totalizando 1412 animais. Modelos de regressão logística foram construídos com os animais atendidos em 2003 com a finalidade de desenvolver um modelo preditivo para o parasitismo com base nos sinais clínicos e resultados de sorologia por Imunofluorescência Indireta (IFI). Este modelo foi validado nos cães atendidos no hospital em 2004. Para a avaliação da área abaixo da curva ROC (auROC), sensibilidade, especificidade, valores preditivos positivo (VPP), valores preditivos negativo (VPN) e acurácia global, foram criados três modelos: um somente baseado nas variáveis clínicas, outro considerando somente o resultado sorológico e um último considerando conjuntamente a clínica e a sorologia. Dentre os três, o último modelo apresentou o melhor desempenho (auROC=90,1%, sensibilidade=82,4%, especificidade=81,6%, VPP=73,4%, VPN=88,2% e acurácia global=81,9%). Conclui-se que o uso de modelos preditivos baseados em critérios clínicos e sorológicos para o diagnóstico da leishmaniose visceral canina pode ser de utilidade no processo de avaliação da infecção canina, promovendo maior agilidade na contenção destes animais com a finalidade de reduzir os níveis de transmissão. O segundo objetivo foi desenvolvido por meio de um estudo transversal com 715 cães de idade entre 1 mês e 13 anos, com raça variada avaliados por clínicos veterinários no HVU-UFPI, no período de janeiro a dezembro de 2003. As sensibilidades e especificidades de critérios clínicos, sorológicos e parasitológicos para detecção de infecção canina por Leishmania chagasi foram estimadas por meio de análise de classe latente, considerando quatro modelos de testes e diferentes pontos de corte. As melhores sensibilidades estimadas para os critérios clínico, sorológico e parasitológico foram de 60%, 95% e 66%, respectivamente. Já as melhores especificidades estimadas para os critérios clínico, sorológico e parasitológico foram de 77%, 90% e 100%, respectivamente. Conclui-se que o uso do exame parasitológico como padrão-ouro para validação de testes diagnósticos não é apropriado e que os indicadores de acurácia dos testes avaliados são insuficientes e não justificam que eles sejam usados isoladamente para diagnóstico da infecção com a finalidade de controle da doença.

A Bayesian method for identification of stock mixtures from molecular marker data

Molecular markers have been demonstrated to be useful for the estimation of stock mixture proportions where the origin of individuals is determined from baseline samples. Bayesian statistical methods are widely recognized as providing a preferable strategy for such analyses. In general, Bayesian estimation is based on standard latent class models using data augmentation through Markov chain Monte Carlo techniques. In this study, we introduce a novel approach based on recent developments in the estimation of genetic population structure. Our strategy combines analytical integration with stochastic optimization to identify stock mixtures. An important enhancement over previous methods is the possibility of appropriately handling data where only partial baseline sample information is available. We address the potential use of nonmolecular, auxiliary biological information in our Bayesian model.

Learning shape priors for single view reconstruction

In this paper, we aim to reconstruct free-from 3D models from a single view by learning the prior knowledge of a specific class of objects. Instead of heuristically proposing specific regularities and defining parametric models as previous research, our shape prior is learned directly from existing 3D models under a framework based on the Gaussian Process Latent Variable Model (GPLVM). The major contributions of the paper include: 1) a probabilistic framework for prior-based reconstruction we propose, which requires no heuristic of the object, and can be easily generalized to handle various categories of 3D objects, and 2) an attempt at automatic reconstruction of more complex 3D shapes, like human bodies, from 2D silhouettes only. Qualitative and quantitative experimental results on both synthetic and real data demonstrate the efficacy of our new approach. ©2009 IEEE.

Gaussian mixture modeling with Gaussian process latent variable models

Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets. © 2010 Springer-Verlag.

Cross-Domain Multitask Learning with Latent Probit Models

Learning multiple tasks across heterogeneous domains is a challenging problem since the feature space may not be the same for different tasks. We assume the data in multiple tasks are generated from a latent common domain via sparse domain transforms and propose a latent probit model (LPM) to jointly learn the domain transforms, and the shared probit classifier in the common domain. To learn meaningful task relatedness and avoid over-fitting in classification, we introduce sparsity in the domain transforms matrices, as well as in the common classifier. We derive theoretical bounds for the estimation error of the classifier in terms of the sparsity of domain transforms. An expectation-maximization algorithm is derived for learning the LPM. The effectiveness of the approach is demonstrated on several real datasets.

It's not that I don't care, I just don't care very much: confounding between attribute non-attendance and taste heterogeneity

With the growing interest in the topic of attribute non-attendance, there is now widespread use of latent class (LC) structures aimed at capturing such behaviour, across a number of different fields. Specifically, these studies rely on a confirmatory LC model, using two separate values for each coefficient, one of which is fixed to zero while the other is estimated, and then use the obtained class probabilities as an indication of the degree of attribute non-attendance. In the present paper, we argue that this approach is in fact misguided, and that the results are likely to be affected by confounding with regular taste heterogeneity. We contrast the confirmatory model with an exploratory LC structure in which the values in both classes are estimated. We also put forward a combined latent class mixed logit model (LC-MMNL) which allows jointly for attribute non-attendance and for continuous taste heterogeneity. Across three separate case studies, the exploratory LC model clearly rejects the confirmatory LC approach and suggests that rates of non-attendance may be much lower than what is suggested by the standard model, or even zero. The combined LC-MMNL model similarly produces significant improvements in model fit, along with substantial reductions in the implied rate of attribute non-attendance, in some cases even eliminating the phenomena across the sample population. Our results thus call for a reappraisal of the large body of recent work that has implied high rates of attribute non-attendance for some attributes. Finally, we also highlight a number of general issues with attribute non-attendance, in particular relating to the computation of willingness to pay measures.

Évolution des profils d'état de santé et utilisation des services sociosanitaires chez les personnes âgées

Réalisé en cotutelle avec l'Université de Paris-Sud

La comorbidité de l'épuisement professionnel et de la consommation de substances psychoactives.

L’objectif principal du présent mémoire réside dans l’exploration des liens concomitants existant entre les trois dimensions de l’épuisement professionnel telles que définies par Maslach (1981) et la consommation de substances psychoactives, plus précisément au niveau de la consommation épisodique excessive et hebdomadaire à risque d’alcool et de la consommation de médicaments psychotropes. À partir d’un échantillon composé de 1809 travailleurs provenant de 83 milieux de travail québécois, des profils-types correspondant à des formes particulières de comorbidité de santé mentale au travail sont identifiés grâce à la méthode d’analyse en classes latentes. Ainsi quatre profils-types sont dégagés: un premier regroupant les individus dits «sains», dont les scores aux différentes dimensions de l’épuisement professionnel sont faibles et dont la consommation de substances psychoactives est modérée; deux autres correspondant à des formes intermédiaires de risques; et un quatrième rassemblant des travailleurs dits «fragiles» dont les scores pour chacune des dimensions de l’épuisement professionnel se situent dans le quintile le plus élevé et dont les probabilités de consommation de substances psychoactives sont grandes. De plus, cette recherche s’est penchée sur l’identification de facteurs de risque et de protection associés à chacun des profils-types. À cet effet, les résultats des analyses corroborent la plupart des associations retrouvées au sein de la littérature quant aux facteurs du travail (composantes des modèles du stress professionnel de Karasek et Theorell (1990) ainsi que de Siegrist (1990)), hors travail (statut matrimonial, obligations parentales, revenu du ménage) et certaines caractéristiques individuelles (âge et genre). De faibles récompenses et un fort degré de surinvestissement de la part de l’individu se révèlent être des facteurs de risque particulièrement significatifs pour les formes intermédiaires et à risque de comorbidité de la santé mentale au travail. Dans une moindre mesure, une faible utilisation des compétences, des demandes psychologiques élevées, un soutien social inadéquat et le jeune âge expliquent une part de la variation observée entre les différents profils-types. Enfin, les résultats soutiennent une conceptualisation tridimensionnelle de l’épuisement professionnel.

Prédire le professionnalisme : une exploration des modèles de la famille de Rasch et de la régression par classes latentes (Étude PERSONA)

L’étude cherche à provoquer la convergence des regards sur des enjeux méthodologiques fondamentaux, soit les enjeux de mesure, de décision et d’impact inhérents à toute démarche de sélection académique. À cet effet, elle explorera la capacité de prédiction de certaines variables non cognitives envers la compétence de professionnalisme observée chez les étudiants du doctorat professionnel de premier cycle en pharmacie. La sélection des candidats au sein des programmes académiques en santé repose en grande partie sur une évaluation de la capacité cognitive des étudiants. Tenant compte du virage compétence pris par la majorité des programmes en santé, la pertinence et la validité des critères traditionnels de sélection sont remises en question. La présente étude propose de valider l’utilisation des échelles de mesure de la personnalité, des valeurs et de l’autodétermination pour guider l’optimalité et l’équité des décisions de sélection. Les enjeux de mesure de ces variables seront abordés principalement par la modélisation dichotomique et polytomique de Rasch. L’application de la méthode des strates permettra, par la suite, de répondre aux enjeux de décision en procédant à une différenciation et un classement des étudiants. Puis, les enjeux d’impact seront, à leur tour, explorés par le modèle de régression par classes latentes. L’étude démontre notamment que le recours à la modélisation a permis une différenciation précise des étudiants. Cependant, la violation de certaines conditions d’application des modèles et la faible différenciation établie entre les étudiants sur la base des critères de professionnalisme, rendent l’évaluation de la capacité de prédiction de la personnalité, des valeurs et de l’autodétermination hasardeuse. À cet effet, les modèles identifiés par les analyses de régression par classes latentes s’avèrent peu concluants. Les classes latentes ainsi identifiées ne présentent pas de distinctions marquées et utiles à la sélection. Bien que les diverses procédures de modélisation proposées présentent des avantages intéressants pour une utilisation en contexte de sélection académique, des recherches additionnelles sur la qualité des critères de professionnalisme et sur la qualité des échelles de mesure des variables non cognitives demeurent nécessaires.

Bayesian parameter estimation for latent Markov random fields and social networks

Undirected graphical models are widely used in statistics, physics and machine vision. However Bayesian parameter estimation for undirected models is extremely challenging, since evaluation of the posterior typically involves the calculation of an intractable normalising constant. This problem has received much attention, but very little of this has focussed on the important practical case where the data consists of noisy or incomplete observations of the underlying hidden structure. This paper specifically addresses this problem, comparing two alternative methodologies. In the first of these approaches particle Markov chain Monte Carlo (Andrieu et al., 2010) is used to efficiently explore the parameter space, combined with the exchange algorithm (Murray et al., 2006) for avoiding the calculation of the intractable normalising constant (a proof showing that this combination targets the correct distribution in found in a supplementary appendix online). This approach is compared with approximate Bayesian computation (Pritchard et al., 1999). Applications to estimating the parameters of Ising models and exponential random graphs from noisy data are presented. Each algorithm used in the paper targets an approximation to the true posterior due to the use of MCMC to simulate from the latent graphical model, in lieu of being able to do this exactly in general. The supplementary appendix also describes the nature of the resulting approximation.

Typology of alcohol mixed with energy drink consumers: motivations for use

Background: Previous research on alcohol mixed with energy drinks (AmED) has shown that use is typically driven by hedonistic, social, functional, and intoxication-related motives, with differential associations with alcohol-related harm across these constructs. There has been no research looking at whether there are subgroups of consumers based on patterns of motivations. Consequently, the aims were to determine the typology of motivations for AmED use among a community sample and to identify correlates of subgroup membership. In addition, we aimed to determine whether this structure of motivations applied to a university student sample. Methods: Data were used from an Australian community sample (n = 731) and an Australian university student sample (n = 594) who were identified as AmED consumers when completing an online survey about their alcohol and ED use. Participants reported their level of agreement with 14 motivations for AmED use; latent classes of AmED consumers were identified based on patterns of motivation endorsement using latent class analysis. Results: A 4-class model was selected using data from the community sample: (i) taste consumers (31%): endorsed pleasurable taste; (ii) energy-seeking consumers (24%): endorsed functional and taste motives; (iii) hedonistic consumers (33%): endorse pleasure and sensation-seeking motives, as well as functional and taste motives; and (iv) intoxication-related consumers (12%): endorsed motives related to feeling in control of intoxication, as well as hedonistic, functional, and taste motives. The consumer subgroups typically did not differ on demographics, other drug use, alcohol and ED use, and AmED risk taking. The patterns of motivations for the 4-class model were similar for the university student sample. Conclusions: This study indicated the existence of 4 subgroups of AmED consumers based on their patterns of motivations for AmED use consistently structured across the community and university student sample. These findings lend support to the growing conceptualization of AmED consumers as a heterogeneous group in regard to motivations for use, with a hierarchical and cumulative class order in regard to the number of types of motivation for AmED use. Prospective research may endeavor to link session-specific motives and outcomes, as it is apparent that primary consumption motives may be fluid between sessions.

Subtyping based on readiness and confidence: the identification of help-seeking profiles for gamblers accessing web-based counselling

BACKGROUND AND AIMS: Problem gamblers are not a homogeneous group and recent data suggest that subtyping can improve treatment outcomes. This study administered three readiness rulers and aimed to identify subtypes of gamblers accessing a national web-based counselling service based on these rulers. METHODS: Participants were 1204 gamblers (99.4% problem gamblers) who accessed a single session of web-based counselling in Australia. Measures included three readiness rulers (importance, readiness and confidence to resist an urge to gamble), demographics and the Problem Gambling Severity Index (PGSI). RESULTS: Gamblers reported high importance of change [mean = 9.2, standard deviation (SD) = 1.51] and readiness to change (mean = 8.86, SD = 1.84), but lower confidence to resist an urge to gamble (mean = 3.93, SD = 2.44) compared with importance and readiness. The statistical fit indices of a latent class analysis identified a four-class model. Subtype 1 was characterized by a very high readiness to change and very low confidence to resist an urge to gamble (n = 662, 55.0%) and subtype 2 reported high readiness and low confidence (n = 358, 29.7%). Subtype 3 reported moderate ratings on all three rulers (n = 139, 11.6%) and subtype 4 reported high importance of change but low readiness and confidence (n = 45, 3.7%). A multinomial logistic regression indicated that subtypes differed by gender (P < 0.001), age (P = 0.01), gambling activity (P < 0.05), preferred mode of gambling (P < 0.001) and PGSI score (P < 0.001). CONCLUSIONS: Problem gamblers in Australia who seek web-based counselling comprise four distinct subgroups based on self-reported levels of readiness to change, confidence to resist the urge to gamble and importance of change.