Mieloma múltiple: rendimiento del estudio del cariotipo por técnicas convencionales y de hibridación in situ fluorescente. Significado pronóstico de las ganacias de 1 q.

En éste estudio analizamos el rendimiento de las técnicas citogenéticas utilizadas en los protocolos diagnósticos del MM (CC y técnicas de FISH). En primer lugar caracterizamos la serie de pacientes y los estratificamos en grupos de riesgo según los sistemas de estadificación actuales. Después estudiamos el porcentaje de cariotipos patológicos con cada una de las técnicas y en conjunto, encontrando un 40% de cariotipos patológicos por CC y de estos un 11,5% pertenecían a estudios con recuentos de CP ≤20% por citomorfología. La técnica de FISH aumentó hasta un 68% los cariotipos patológicos. También hemos realizado la caracterización de los pacientes con ganancias de 1q y su impacto en la evolución de la enfermedad.

Activity in inferior parietal and medial prefrontal cortex signals the accumulation of evidence in a probability learning task.

In an uncertain environment, probabilities are key to predicting future events and making adaptive choices. However, little is known about how humans learn such probabilities and where and how they are encoded in the brain, especially when they concern more than two outcomes. During functional magnetic resonance imaging (fMRI), young adults learned the probabilities of uncertain stimuli through repetitive sampling. Stimuli represented payoffs and participants had to predict their occurrence to maximize their earnings. Choices indicated loss and risk aversion but unbiased estimation of probabilities. BOLD response in medial prefrontal cortex and angular gyri increased linearly with the probability of the currently observed stimulus, untainted by its value. Connectivity analyses during rest and task revealed that these regions belonged to the default mode network. The activation of past outcomes in memory is evoked as a possible mechanism to explain the engagement of the default mode network in probability learning. A BOLD response relating to value was detected only at decision time, mainly in striatum. It is concluded that activity in inferior parietal and medial prefrontal cortex reflects the amount of evidence accumulated in favor of competing and uncertain outcomes.

Essays in higher moment asset pricing and liquidity risk

This thesis focuses on theoretical asset pricing models and their empirical applications. I aim to investigate the following noteworthy problems: i) if the relationship between asset prices and investors' propensities to gamble and to fear disaster is time varying, ii) if the conflicting evidence for the firm and market level skewness can be explained by downside risk, Hi) if costly learning drives liquidity risk. Moreover, empirical tests support the above assumptions and provide novel findings in asset pricing, investment decisions, and firms' funding liquidity. The first chapter considers a partial equilibrium model where investors have heterogeneous propensities to gamble and fear disaster. Skewness preference represents the desire to gamble, while kurtosis aversion represents fear of extreme returns. Using US data from 1988 to 2012, my model demonstrates that in bad times, risk aversion is higher, more people fear disaster, and fewer people gamble, in contrast to good times. This leads to a new empirical finding: gambling preference has a greater impact on asset prices during market downturns than during booms. The second chapter consists of two essays. The first essay introduces a foramula based on conditional CAPM for decomposing the market skewness. We find that the major market upward and downward movements can be well preadicted by the asymmetric comovement of betas, which is characterized by an indicator called "Systematic Downside Risk" (SDR). We find that SDR can efafectively forecast future stock market movements and we obtain out-of-sample R-squares (compared with a strategy using historical mean) of more than 2.27% with monthly data. The second essay reconciles a well-known empirical fact: aggregating positively skewed firm returns leads to negatively skewed market return. We reconcile this fact through firms' greater response to negative maraket news than positive market news. We also propose several market return predictors, such as downside idiosyncratic skewness. The third chapter studies the funding liquidity risk based on a general equialibrium model which features two agents: one entrepreneur and one external investor. Only the investor needs to acquire information to estimate the unobservable fundamentals driving the economic outputs. The novelty is that information acquisition is more costly in bad times than in good times, i.e. counter-cyclical information cost, as supported by previous empirical evidence. Later we show that liquidity risks are principally driven by costly learning. Résumé Cette thèse présente des modèles théoriques dévaluation des actifs et leurs applications empiriques. Mon objectif est d'étudier les problèmes suivants: la relation entre l'évaluation des actifs et les tendances des investisseurs à parier et à crainadre le désastre varie selon le temps ; les indications contraires pour l'entreprise et l'asymétrie des niveaux de marché peuvent être expliquées par les risques de perte en cas de baisse; l'apprentissage coûteux augmente le risque de liquidité. En outre, des tests empiriques confirment les suppositions ci-dessus et fournissent de nouvelles découvertes en ce qui concerne l'évaluation des actifs, les décisions relatives aux investissements et la liquidité de financement des entreprises. Le premier chapitre examine un modèle d'équilibre où les investisseurs ont des tendances hétérogènes à parier et à craindre le désastre. La préférence asymétrique représente le désir de parier, alors que le kurtosis d'aversion représente la crainte du désastre. En utilisant les données des Etats-Unis de 1988 à 2012, mon modèle démontre que dans les mauvaises périodes, l'aversion du risque est plus grande, plus de gens craignent le désastre et moins de gens parient, conatrairement aux bonnes périodes. Ceci mène à une nouvelle découverte empirique: la préférence relative au pari a un plus grand impact sur les évaluations des actifs durant les ralentissements de marché que durant les booms économiques. Exploitant uniquement cette relation générera un revenu excédentaire annuel de 7,74% qui n'est pas expliqué par les modèles factoriels populaires. Le second chapitre comprend deux essais. Le premier essai introduit une foramule base sur le CAPM conditionnel pour décomposer l'asymétrie du marché. Nous avons découvert que les mouvements de hausses et de baisses majeures du marché peuvent être prédits par les mouvements communs des bêtas. Un inadicateur appelé Systematic Downside Risk, SDR (risque de ralentissement systématique) est créé pour caractériser cette asymétrie dans les mouvements communs des bêtas. Nous avons découvert que le risque de ralentissement systématique peut prévoir les prochains mouvements des marchés boursiers de manière efficace, et nous obtenons des carrés R hors échantillon (comparés avec une stratégie utilisant des moyens historiques) de plus de 2,272% avec des données mensuelles. Un investisseur qui évalue le marché en utilisant le risque de ralentissement systématique aurait obtenu une forte hausse du ratio de 0,206. Le second essai fait cadrer un fait empirique bien connu dans l'asymétrie des niveaux de march et d'entreprise, le total des revenus des entreprises positiveament asymétriques conduit à un revenu de marché négativement asymétrique. Nous décomposons l'asymétrie des revenus du marché au niveau de l'entreprise et faisons cadrer ce fait par une plus grande réaction des entreprises aux nouvelles négatives du marché qu'aux nouvelles positives du marché. Cette décomposition révélé plusieurs variables de revenus de marché efficaces tels que l'asymétrie caractéristique pondérée par la volatilité ainsi que l'asymétrie caractéristique de ralentissement. Le troisième chapitre fournit une nouvelle base théorique pour les problèmes de liquidité qui varient selon le temps au sein d'un environnement de marché incomplet. Nous proposons un modèle d'équilibre général avec deux agents: un entrepreneur et un investisseur externe. Seul l'investisseur a besoin de connaitre le véritable état de l'entreprise, par conséquent, les informations de paiement coutent de l'argent. La nouveauté est que l'acquisition de l'information coute plus cher durant les mauvaises périodes que durant les bonnes périodes, comme cela a été confirmé par de précédentes expériences. Lorsque la récession comamence, l'apprentissage coûteux fait augmenter les primes de liquidité causant un problème d'évaporation de liquidité, comme cela a été aussi confirmé par de précédentes expériences.

Human invariant V alpha 24-J alpha Q TCR supports the development of CD1d-dependent NK1.1+ and NK1.1- T cells in transgenic mice.

A sizable fraction of T cells expressing the NK cell marker NK1.1 (NKT cells) bear a very conserved TCR, characterized by homologous invariant (inv.) TCR V alpha 24-J alpha Q and V alpha 14-J alpha 18 rearrangements in humans and mice, respectively, and are thus defined as inv. NKT cells. Because human inv. NKT cells recognize mouse CD1d in vitro, we wondered whether a human inv. V alpha 24 TCR could be selected in vivo by mouse ligands presented by CD1d, thereby supporting the development of inv. NKT cells in mice. Therefore, we generated transgenic (Tg) mice expressing the human inv. V alpha 24-J alpha Q TCR chain in all T cells. The expression of the human inv. V alpha 24 TCR in TCR C alpha(-/-) mice indeed rescues the development of inv. NKT cells, which home preferentially to the liver and respond to the CD1d-restricted ligand alpha-galactosylceramide (alpha-GalCer). However, unlike inv. NKT cells from non-Tg mice, the majority of NKT cells in V alpha 24 Tg mice display a double-negative phenotype, as well as a significant increase in TCR V beta 7 and a corresponding decrease in TCR V beta 8.2 use. Despite the forced expression of the human CD1d-restricted TCR in C alpha(-/-) mice, staining with mCD1d-alpha-GalCer tetramers reveals that the absolute numbers of peripheral CD1d-dependent T lymphocytes increase at most by 2-fold. This increase is accounted for mainly by an increased fraction of NK1.1(-) T cells that bind CD1d-alpha-GalCer tetramers. These findings indicate that human inv. V alpha 24 TCR supports the development of CD1d-dependent lymphocytes in mice, and argue for a tight homeostatic control on the total number of inv. NKT cells. Thus, human inv. V alpha 24 TCR-expressing mice are a valuable model to study different aspects of the inv. NKT cell subset.

Dynamics of the recruitment of lateral amygdala neurons following successive fear learning to different conditional stimuli

In fear conditioning, an animal learns to associate an unconditioned stimulus (US), such as a shock, and a conditioned stimulus (CS), such as a tone, so that the presentation of the CS alone can trigger conditioned responses. Recent research on the lateral amygdala has shown that following cued fear conditioning, only a subset of higher-excitable neurons are recruited in the memory trace. Their selective deletion after fear conditioning results in a selective erasure of the fearful memory. I hypothesize that the recruitment of highly excitable neurons depends on responsiveness to stimuli, intrinsic excitability and local connectivity. In addition, I hypothesize that neurons recruited for an initial memory also participate in subsequent memories, and that changes in neuronal excitability affect secondary fear learning. To address these hypotheses, I will show that A) a rat can learn to associate two successive short-term fearful memories; B) neuronal populations in the LA are competitively recruited in the memory traces depending on individual neuronal advantages, as well as advantages granted by the local network. By performing two successive cued fear conditioning experiments, I found that rats were able to learn and extinguish the two successive short-term memories, when tested 1 hour after learning for each memory. These rats were equipped with a system of stable extracellular recordings that I developed, which allowed to monitor neuronal activity during fear learning. 233 individual putative pyramidal neurons could modulate their firing rate in response to the conditioned tone (conditioned neurons) and/or non- conditioned tones (generalizing neurons). Out of these recorded putative pyramidal neurons 86 (37%) neurons were conditioned to one or both tones. More precisely, one population of neurons encoded for a shared memory while another group of neurons likely encoded the memories' new features. Notably, in spite of a successful behavioral extinction, the firing rate of those conditioned neurons in response to the conditioned tone remained unchanged throughout memory testing. Furthermore, by analyzing the pre-conditioning characteristics of the conditioned neurons, I determined that it was possible to predict neuronal recruitment based on three factors: 1) initial sensitivity to auditory inputs, with tone-sensitive neurons being more easily recruited than tone- insensitive neurons; 2) baseline excitability levels, with more highly excitable neurons being more likely to become conditioned; and 3) the number of afferent connections received from local neurons, with neurons destined to become conditioned receiving more connections than non-conditioned neurons. - En conditionnement de la peur, un animal apprend à associer un stimulus inconditionnel (SI), tel un choc électrique, et un stimulus conditionné (SC), comme un son, de sorte que la présentation du SC seul suffit pour déclencher des réflexes conditionnés. Des recherches récentes sur l'amygdale latérale (AL) ont montré que, suite au conditionnement à la peur, seul un sous-ensemble de neurones plus excitables sont recrutés pour constituer la trace mnésique. Pour apprendre à associer deux sons au même SI, je fais l'hypothèse que les neurones entrent en compétition afin d'être sélectionnés lors du recrutement pour coder la trace mnésique. Ce recrutement dépendrait d'un part à une activation facilité des neurones ainsi qu'une activation facilité de réseaux de neurones locaux. En outre, je fais l'hypothèse que l'activation de ces réseaux de l'AL, en soi, est suffisante pour induire une mémoire effrayante. Pour répondre à ces hypothèses, je vais montrer que A) selon un processus de mémoire à court terme, un rat peut apprendre à associer deux mémoires effrayantes apprises successivement; B) des populations neuronales dans l'AL sont compétitivement recrutées dans les traces mnésiques en fonction des avantages neuronaux individuels, ainsi que les avantages consentis par le réseau local. En effectuant deux expériences successives de conditionnement à la peur, des rats étaient capables d'apprendre, ainsi que de subir un processus d'extinction, pour les deux souvenirs effrayants. La mesure de l'efficacité du conditionnement à la peur a été effectuée 1 heure après l'apprentissage pour chaque souvenir. Ces rats ont été équipés d'un système d'enregistrements extracellulaires stables que j'ai développé, ce qui a permis de suivre l'activité neuronale pendant l'apprentissage de la peur. 233 neurones pyramidaux individuels pouvaient moduler leur taux d'activité en réponse au son conditionné (neurones conditionnés) et/ou au son non conditionné (neurones généralisant). Sur les 233 neurones pyramidaux putatifs enregistrés 86 (37%) d'entre eux ont été conditionnés à un ou deux tons. Plus précisément, une population de neurones code conjointement pour un souvenir partagé, alors qu'un groupe de neurones différent code pour de nouvelles caractéristiques de nouveaux souvenirs. En particulier, en dépit d'une extinction du comportement réussie, le taux de décharge de ces neurones conditionné en réponse à la tonalité conditionnée est resté inchangée tout au long de la mesure d'apprentissage. En outre, en analysant les caractéristiques de pré-conditionnement des neurones conditionnés, j'ai déterminé qu'il était possible de prévoir le recrutement neuronal basé sur trois facteurs : 1) la sensibilité initiale aux entrées auditives, avec les neurones sensibles aux sons étant plus facilement recrutés que les neurones ne répondant pas aux stimuli auditifs; 2) les niveaux d'excitabilité des neurones, avec les neurones plus facilement excitables étant plus susceptibles d'être conditionnés au son ; et 3) le nombre de connexions reçues, puisque les neurones conditionné reçoivent plus de connexions que les neurones non-conditionnés. Enfin, nous avons constaté qu'il était possible de remplacer de façon satisfaisante le SI lors d'un conditionnement à la peur par des injections bilatérales de bicuculline, un antagoniste des récepteurs de l'acide y-Aminobutirique.

Active Learning of Very-High Resolution Optical Imagery with SVM: Entropy vs Margin Sampling

An active learning method is proposed for the semi-automatic selection of training sets in remote sensing image classification. The method adds iteratively to the current training set the unlabeled pixels for which the prediction of an ensemble of classifiers based on bagged training sets show maximum entropy. This way, the algorithm selects the pixels that are the most uncertain and that will improve the model if added in the training set. The user is asked to label such pixels at each iteration. Experiments using support vector machines (SVM) on an 8 classes QuickBird image show the excellent performances of the methods, that equals accuracies of both a model trained with ten times more pixels and a model whose training set has been built using a state-of-the-art SVM specific active learning method

Machine learning algorithms for spatial data. Case studies: Environmental pollution, natural hazards, renewable resources

This paper presents general problems and approaches for the spatial data analysis using machine learning algorithms. Machine learning is a very powerful approach to adaptive data analysis, modelling and visualisation. The key feature of the machine learning algorithms is that they learn from empirical data and can be used in cases when the modelled environmental phenomena are hidden, nonlinear, noisy and highly variable in space and in time. Most of the machines learning algorithms are universal and adaptive modelling tools developed to solve basic problems of learning from data: classification/pattern recognition, regression/mapping and probability density modelling. In the present report some of the widely used machine learning algorithms, namely artificial neural networks (ANN) of different architectures and Support Vector Machines (SVM), are adapted to the problems of the analysis and modelling of geo-spatial data. Machine learning algorithms have an important advantage over traditional models of spatial statistics when problems are considered in a high dimensional geo-feature spaces, when the dimension of space exceeds 5. Such features are usually generated, for example, from digital elevation models, remote sensing images, etc. An important extension of models concerns considering of real space constrains like geomorphology, networks, and other natural structures. Recent developments in semi-supervised learning can improve modelling of environmental phenomena taking into account on geo-manifolds. An important part of the study deals with the analysis of relevant variables and models' inputs. This problem is approached by using different feature selection/feature extraction nonlinear tools. To demonstrate the application of machine learning algorithms several interesting case studies are considered: digital soil mapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis (avalanches, landslides), assessments of renewable resources (wind fields) with SVM and ANN models, etc. The dimensionality of spaces considered varies from 2 to more than 30. Figures 1, 2, 3 demonstrate some results of the studies and their outputs. Finally, the results of environmental mapping are discussed and compared with traditional models of geostatistics.

The A-B-C of B-B-Q

Whatever the weather, the allure of summer dining alfresco is very appealing. And while eating outdoors can be a real pleasure, far too often the good habits we follow in the kitchen go up in smoke when the barbecue is lit. When planning a barbecue this summer, bear in mind the 6 simple rules outlined in this report, so your friends, family and neighbours go home with memories of a good time - andnot a bug to remember you by!

Adult Learning Services

Evidence Review 4 - Adult learning services Briefing 4 - Adult learning services This pair of documents, commissioned by Public Health England, and written by the UCL Institute of Health Equity, address the role of participation in learning as an adult in improving health. There is evidence that involvement in adult learning has both direct and indirect links with health, for example because it increases employability. There is some evidence that those who are lower down the social gradient benefit most, in health terms, from adult learning. However, there is a gradient both in participation in adult learning and skill level, whereby the more someone would benefit from adult learning, the less likely they are to participate, and the lower their literacy and numeracy skills are likely to be. This is due to a range of barriers, including prohibitively high costs, lack of personal confidence, or lack of availability and access. These papers also show that there are a number of actions local authorities can take to increase access to adult learning, improve quality of provision and increase the extent to which it is delivered and targeted proportionate to need. The full evidence review and a shorter summary briefing are available to download above. This document is part of a series. An overview document which provides an introduction to this and other documents in the series, and links to the other topic areas, is available on the ‘Local Action on health inequalities’ project page. A video of Michael Marmot introducing the work is also available on our videos page.