Detecting genetic and nutritional lung cancer risk factors related to folate metabolism using Bayesian generalized linear models

Complex diseases, such as cancer, are caused by various genetic and environmental factors, and their interactions. Joint analysis of these factors and their interactions would increase the power to detect risk factors but is statistically. Bayesian generalized linear models using student-t prior distributions on coefficients, is a novel method to simultaneously analyze genetic factors, environmental factors, and interactions. I performed simulation studies using three different disease models and demonstrated that the variable selection performance of Bayesian generalized linear models is comparable to that of Bayesian stochastic search variable selection, an improved method for variable selection when compared to standard methods. I further evaluated the variable selection performance of Bayesian generalized linear models using different numbers of candidate covariates and different sample sizes, and provided a guideline for required sample size to achieve a high power of variable selection using Bayesian generalize linear models, considering different scales of number of candidate covariates. ^ Polymorphisms in folate metabolism genes and nutritional factors have been previously associated with lung cancer risk. In this study, I simultaneously analyzed 115 tag SNPs in folate metabolism genes, 14 nutritional factors, and all possible genetic-nutritional interactions from 1239 lung cancer cases and 1692 controls using Bayesian generalized linear models stratified by never, former, and current smoking status. SNPs in MTRR were significantly associated with lung cancer risk across never, former, and current smokers. In never smokers, three SNPs in TYMS and three gene-nutrient interactions, including an interaction between SHMT1 and vitamin B12, an interaction between MTRR and total fat intake, and an interaction between MTR and alcohol use, were also identified as associated with lung cancer risk. These lung cancer risk factors are worthy of further investigation.^

Automatic compile-time parallelization of logic programs for restricted, goal-level, independent and-parallelism.

A framework for the automatic parallelization of (constraint) logic programs is proposed and proved correct. Intuitively, the parallelization process replaces conjunctions of literals with parallel expressions. Such expressions trigger at run-time the exploitation of restricted, goal-level, independent and-parallelism. The parallelization process performs two steps. The first one builds a conditional dependency graph (which can be implified using compile-time analysis information), while the second transforms the resulting graph into linear conditional expressions, the parallel expressions of the &-Prolog language. Several heuristic algorithms for the latter ("annotation") process are proposed and proved correct. Algorithms are also given which determine if there is any loss of parallelism in the linearization process with respect to a proposed notion of maximal parallelism. Finally, a system is presented which implements the proposed approach. The performance of the different annotation algorithms is compared experimentally in this system by studying the time spent in parallelization and the effectiveness of the results in terms of speedups.

Non-failure analysis for logic programs

We provide a method whereby, given mode and (upper approximation) type information, we can detect procedures and goals that can be guaranteed to not fail (i.e., to produce at least one solution or not termínate). The technique is based on an intuitively very simple notion, that of a (set of) tests "covering" the type of a set of variables. We show that the problem of determining a covering is undecidable in general, and give decidability and complexity results for the Herbrand and linear arithmetic constraint systems. We give sound algorithms for determining covering that are precise and efiicient in practice. Based on this information, we show how to identify goals and procedures that can be guaranteed to not fail at runtime. Applications of such non-failure information include programming error detection, program transiormations and parallel execution optimization, avoiding speculative parallelism and estimating lower bounds on the computational costs of goals, which can be used for granularity control. Finally, we report on an implementation of our method and show that better results are obtained than with previously proposed approaches.

From eventual to atomic and locally atomic CC programs: A concurrent semantics

We present a concurrent semantics (i.e. a semantics where concurrency is explicitely represented) for CC programs with atomic tells. This allows to derive concurrency, dependency, and nondeterminism information for such languages. The ability to treat failure information puts CLP programs also in the range of applicability of our semantics: although such programs are not concurrent, the concurrency information derived in the semantics may be interpreted as possible parallelism, thus allowing to safely parallelize those computation steps which appear to be concurrent in the net. Dually, the dependency information may also be interpreted as necessary sequentialization, thus possibly exploiting it to schedule CC programs. The fact that the semantical structure contains dependency information suggests a new tell operation, which checks for consistency only the constraints it depends on, achieving a reasonable trade-off between efficiency and atomicity.

The CDG, UDG, and MEL methods for automatic compile-time parallelization of logic programs for independent and-parallelism

There has been significant interest in parallel execution models for logic programs which exploit Independent And-Parallelism (IAP). In these models, it is necessary to determine which goals are independent and therefore eligible for parallel execution and which goals have to wait for which others during execution. Although this can be done at run-time, it can imply a very heavy overhead. In this paper, we present three algorithms for automatic compiletime parallelization of logic programs using IAP. This is done by converting a clause into a graph-based computational form and then transforming this graph into linear expressions based on &-Prolog, a language for IAP. We also present an algorithm which, given a clause, determines if there is any loss of parallelism due to linearization, for the case in which only unconditional parallelism is desired. Finally, the performance of these annotation algorithms is discussed for some benchmark programs.

Linear Bayes policy for learning in contextual-bandits

Machine and Statistical Learning techniques are used in almost all online advertisement systems. The problem of discovering which content is more demanded (e.g. receive more clicks) can be modeled as a multi-armed bandit problem. Contextual bandits (i.e., bandits with covariates, side information or associative reinforcement learning) associate, to each specific content, several features that define the “context” in which it appears (e.g. user, web page, time, region). This problem can be studied in the stochastic/statistical setting by means of the conditional probability paradigm using the Bayes’ theorem. However, for very large contextual information and/or real-time constraints, the exact calculation of the Bayes’ rule is computationally infeasible. In this article, we present a method that is able to handle large contextual information for learning in contextual-bandits problems. This method was tested in the Challenge on Yahoo! dataset at ICML2012’s Workshop “new Challenges for Exploration & Exploitation 3”, obtaining the second place. Its basic exploration policy is deterministic in the sense that for the same input data (as a time-series) the same results are obtained. We address the deterministic exploration vs. exploitation issue, explaining the way in which the proposed method deterministically finds an effective dynamic trade-off based solely in the input-data, in contrast to other methods that use a random number generator.

Estudio de las dinámicas de pérdida de peso corporal en los programas de ejercicio físico y nutrición = Study of the dynamics of body weight loss in exercise and nutritional programs

Introducción. La obesidad puede definirse como una enfermedad metabólica crónica de origen multifactorial, lo que provoca trastornos o problemas físicos y psicológicos a la persona, con patologías asociadas que limitan la esperanza de vida y deterioran la calidad de la misma, siendo determinante para sus áreas sociales y laborales. Este trastorno metabólico crónico se caracteriza por una acumulación excesiva de energía en el cuerpo en forma de grasa, lo que lleva a un aumento de peso con respecto al valor esperado por sexo, edad y altura. La gestión y el tratamiento de la obesidad tienen objetivos más amplios que la pérdida de peso e incluyen la reducción del riesgo y la mejora de la salud. Estos pueden ser alcanzados por la pérdida modesta de peso (es decir, 10.5% del peso corporal inicial), la mejora del contenido nutricional de la dieta y un modesto incremento en la actividad física y condición física. La dieta es uno de los métodos más populares para perder peso corporal. El ejercicio es otra alternativa para perder peso corporal. El aumento de ejercicio provoca un desequilibrio cuando se mantiene la ingesta calórica. También tiene ventajas, como la mejora del tono muscular, la capacidad cardiovascular, fuerza y flexibilidad, aumenta el metabolismo basal y mejora el sistema inmunológico. Objetivos. El objetivo de esta tesis es contribuir en un estudio de intervención para aclarar la evolución del peso corporal durante una intervención de dieta y ejercicio. Para ello, se evaluaron los efectos de la edad, sexo, índice de masa corporal inicial y el tipo de tratamiento en las tendencias de pérdida de peso. Otro objetivo de la tesis era crear un modelo de regresión lineal múltiple capaz de predecir la pérdida de peso corporal después del periodo de intervención. Y, por último, determinar el efecto sobre la composición corporal (peso corporal, índice de masa corporal, la masa grasa, y la masa libre de grasa) de las diferentes intervenciones basadas en ejercicios (fuerza, resistencia, resistencia combinada con fuerza, y las recomendaciones de actividad física (grupo control)) en combinación con dieta de adultos con sobrepeso y obesidad, después de la intervención, así como los cambios de la composición corporal 3 años más tarde. Diseño de la investigación. Los datos empleados en el análisis de esta tesis son parte del proyecto “Programas de Nutrición y Actividad Física para el tratamiento de la obesidad” (PRONAF). El proyecto PRONAF es un estudio clínico sobre programas de nutrición y actividad física para el sobrepeso y la obesidad, desarrollado en España durante varios años de intervención. Fue diseñado, en parte, para comparar diferentes tipos de intervención, con el objetivo de evaluar su impacto en las dinámicas de pérdida de peso, en personas con sobrepeso y obesidad. Como diseño experimental, el estudio se basó en una restricción calórica, a la que, en algunos casos, se le añadió un protocolo de entrenamiento (fuerza, resistencia, o combinado, en igualdad de volumen e intensidad). Las principales variables para la investigación que comprende esta tesis fueron: el peso corporal y la composición corporal (masa grasa y masa libre de grasa). Conclusiones. En esta tesis, para los programas de pérdida de peso en personas con sobrepeso y obesidad con un 25-30% de la restricción calórica, el peso corporal se redujo significativamente en ambos sexos, sin tener en cuenta la edad y el tipo de tratamiento seguido. Según los resultados del estudio, la pérdida de peso realizada por un individuo (hombre o mujer) durante los seis meses puede ser representada por cualquiera de las cinco funciones (lineal, potencial, exponencial, logarítmica y cuadrática) en ambos sexos, siendo la cuadrática la que tiende a representarlo mejor. Además, se puede concluir que la pérdida de peso corporal se ve afectada por el índice de masa corporal inicial y el sexo, siendo mayor para las personas obesas que para las de sobrepeso, que muestran diferencias entre sexos sólo en la condición de sobrepeso. Además, es posible calcular el peso corporal final de cualquier participante involucrado en una intervención utilizando la metodología del proyecto PRONAF sólo conociendo sus variables iniciales de composición corporal. Además, los cuatro tipos de tratamientos tuvieron resultados similares en cambios en la composición corporal al final del período de intervención, con la única excepción de la masa libre de grasa, siendo los grupos de entrenamiento los que la mantuvieron durante la restricción calórica. Por otro lado, sólo el grupo combinado logra mantener la reducción de la masa grasa (%) 3 años después del final de la intervención. ABSTRACT Introduction. Obesity can be defined as a chronic metabolic disease from a multifactorial origin, which leads to physical and psychological impacts to the person, with associated pathologies that limit the life expectancy and deteriorate the quality of it, being determinant for the social and labor areas of the person. This chronic metabolic disorder is characterized by an excessive accumulation of energy in the body as fat, leading to increased weight relative to the value expected by sex, age and height. The management and treatment of obesity have wider objectives than weight loss alone and include risk reduction and health improvement. These may be achieved by modest weight loss (i.e. 5–10% of initial body weight), improved nutritional content of the diet and modest increases in physical activity and fitness. Weight loss through diet is one of the most popular approaches to lose body weight. Exercise is another alternative to lose body weight. The increase of exercise causes an imbalance when the caloric intake is maintained. It also has advantages such as improved muscle tone, cardiovascular fitness, strength and flexibility, increases the basal metabolism and improves immune system. Objectives. The aim of this thesis is to contribute with an interventional study to clarify the evolution of the body weight during a diet and exercise intervention. For this, the effects of age, sex, initial body mass index and type of treatment on weight loss tendencies were evaluated. Another objective of the thesis was to create a multiple linear regression model able to predict the body weight loss after the intervention period. And, finally, to determine the effect upon body composition (body weight, body mass index, fat mass, and fat-free mass of different exercise-based interventions (strength, endurance, combined endurance and strength, and physical activity recommendations group (control group)) combined with diet in overweight and obese adults, after intervention as well as body composition changes 3 years later. Research Design. The data used in the analysis of this thesis are part of the project "Programs of Nutrition and Physical Activity for the treatment of obesity" (PRONAF). The PRONAF project is a clinical trial program about nutrition and physical activity for overweight and obesity, developed in Spain for several years of intervention. It was designed, in part, to compare different types of intervention, in order to assess their impact on the dynamics of weight loss in overweight and obese people. As experimental design, the study was based on caloric restriction, which, in some cases, added a training protocol (strength, endurance, or combined in equal volume and intensity). The main research variables comprising this thesis were: body weight and body composition outcomes (fat mass and fat-free mass). Conclusions. In this thesis, for weight loss programs in overweight and obese people with 25-30% of caloric restriction, the body weight was significantly decreased in both sexes, regardless the age and type of followed treatment. According to the results of the study, the weight loss performed by an individual (male or female) during six months can be represented by any of the five functions (linear, power law, exponential, logarithmic and quadratic) in both sexes, being the quadratic one which tends to represent it better. In addition, it can be concluded that the body weight loss is affected by the initial body mass index and sex condition, being greater for the obese people than for the overweight one, showing differences between sexes only in the overweight condition. Moreover, it is possible to calculate the final body weight of any participant engaged in an intervention using the PRONAF Project methodology only knowing their initial body composition variables. Furthermore, the four types of treatments had similar results on body composition changes at the end of the intervention period, with the only exception of fat-free mass, being the training groups the ones that maintained it during the caloric restriction. On the other hand, only the combined group achieved to maintain the fat mass (%) reduced 3 years after the end of the intervention.

Models for stochastic climate prediction

There has been a recent burst of activity in the atmosphere/ocean sciences community in utilizing stable linear Langevin stochastic models for the unresolved degree of freedom in stochastic climate prediction. Here several idealized models for stochastic climate modeling are introduced and analyzed through unambiguous mathematical theory. This analysis demonstrates the potential need for more sophisticated models beyond stable linear Langevin equations. The new phenomena include the emergence of both unstable linear Langevin stochastic models for the climate mean and the need to incorporate both suitable nonlinear effects and multiplicative noise in stochastic models under appropriate circumstances. The strategy for stochastic climate modeling that emerges from this analysis is illustrated on an idealized example involving truncated barotropic flow on a beta-plane with topography and a mean flow. In this example, the effect of the original 57 degrees of freedom is well represented by a theoretically predicted stochastic model with only 3 degrees of freedom.

On Managing Stochastic Decentralized Projects

Thesis (Ph.D.)--University of Washington, 2016-06

Assessing the performance of a nucleus estate and smallholder scheme for oil palm production in West Sumatra: a stochastic frontier analysis

In this paper, we study the performance of smallholders in a nucleus estate and smallholder (NES) scheme in oil palm production schemein West Sumatra by measuring their technical efficiency using a stochastic frontier production function. Our results indicate a mean technical efficiency of 66%, which is below what we would have expected given the uniformity of the climate, soils and plantation construction among the sample farmers. The use of progressive farmers as a means of disseminating extension advice does not appear to have been successful, and more rigorous farmer selection procedures need to be put in place for similar schemes and for general agricultural extension in future. No clear relationship was established between technical efficiency and the use of female labour, suggesting there is no need to target extension services specifically at female labourers in the household. Finally, education was found to have an unexpectedly negative impact on technical efficiency, indicating that farmers with primary education may be more important than those with secondary and tertiary education as targets of development schemes and extension programs entailing non-formal education. (C) 2003 Elsevier Ltd. All rights reserved.

Stochastic modelling of the invasion process of nematodes in fly larvae

Stochastic models based on Markov birth processes are constructed to describe the process of invasion of a fly larva by entomopathogenic nematodes. Various forms for the birth (invasion) rates are proposed. These models are then fitted to data sets describing the observed numbers of nematodes that have invaded a fly larval after a fixed period of time. Non-linear birthrates are required to achieve good fits to these data, with their precise form leading to different patterns of invasion being identified for three populations of nematodes considered. One of these (Nemasys) showed the greatest propensity for invasion. This form of modelling may be useful more generally for analysing data that show variation which is different from that expected from a binomial distribution.

Approximate Bayesian techniques for inference in stochastic dynamical systems

This thesis is concerned with approximate inference in dynamical systems, from a variational Bayesian perspective. When modelling real world dynamical systems, stochastic differential equations appear as a natural choice, mainly because of their ability to model the noise of the system by adding a variant of some stochastic process to the deterministic dynamics. Hence, inference in such processes has drawn much attention. Here two new extended frameworks are derived and presented that are based on basis function expansions and local polynomial approximations of a recently proposed variational Bayesian algorithm. It is shown that the new extensions converge to the original variational algorithm and can be used for state estimation (smoothing). However, the main focus is on estimating the (hyper-) parameters of these systems (i.e. drift parameters and diffusion coefficients). The new methods are numerically validated on a range of different systems which vary in dimensionality and non-linearity. These are the Ornstein-Uhlenbeck process, for which the exact likelihood can be computed analytically, the univariate and highly non-linear, stochastic double well and the multivariate chaotic stochastic Lorenz '63 (3-dimensional model). The algorithms are also applied to the 40 dimensional stochastic Lorenz '96 system. In this investigation these new approaches are compared with a variety of other well known methods such as the ensemble Kalman filter / smoother, a hybrid Monte Carlo sampler, the dual unscented Kalman filter (for jointly estimating the systems states and model parameters) and full weak-constraint 4D-Var. Empirical analysis of their asymptotic behaviour as a function of observation density or length of time window increases is provided.

Estimating parameters in stochastic systems:a variational Bayesian approach

This work is concerned with approximate inference in dynamical systems, from a variational Bayesian perspective. When modelling real world dynamical systems, stochastic differential equations appear as a natural choice, mainly because of their ability to model the noise of the system by adding a variation of some stochastic process to the deterministic dynamics. Hence, inference in such processes has drawn much attention. Here a new extended framework is derived that is based on a local polynomial approximation of a recently proposed variational Bayesian algorithm. The paper begins by showing that the new extension of this variational algorithm can be used for state estimation (smoothing) and converges to the original algorithm. However, the main focus is on estimating the (hyper-) parameters of these systems (i.e. drift parameters and diffusion coefficients). The new approach is validated on a range of different systems which vary in dimensionality and non-linearity. These are the Ornstein–Uhlenbeck process, the exact likelihood of which can be computed analytically, the univariate and highly non-linear, stochastic double well and the multivariate chaotic stochastic Lorenz ’63 (3D model). As a special case the algorithm is also applied to the 40 dimensional stochastic Lorenz ’96 system. In our investigation we compare this new approach with a variety of other well known methods, such as the hybrid Monte Carlo, dual unscented Kalman filter, full weak-constraint 4D-Var algorithm and analyse empirically their asymptotic behaviour as a function of observation density or length of time window increases. In particular we show that we are able to estimate parameters in both the drift (deterministic) and the diffusion (stochastic) part of the model evolution equations using our new methods.

Neural network enhanced self tuning adaptive control application for non-linear control of dynamic systems

The main theme of research of this project concerns the study of neutral networks to control uncertain and non-linear control systems. This involves the control of continuous time, discrete time, hybrid and stochastic systems with input, state or output constraints by ensuring good performances. A great part of this project is devoted to the opening of frontiers between several mathematical and engineering approaches in order to tackle complex but very common non-linear control problems. The objectives are: 1. Design and develop procedures for neutral network enhanced self-tuning adaptive non-linear control systems; 2. To design, as a general procedure, neural network generalised minimum variance self-tuning controller for non-linear dynamic plants (Integration of neural network mapping with generalised minimum variance self-tuning controller strategies); 3. To develop a software package to evaluate control system performances using Matlab, Simulink and Neural Network toolbox. An adaptive control algorithm utilising a recurrent network as a model of a partial unknown non-linear plant with unmeasurable state is proposed. Appropriately, it appears that structured recurrent neural networks can provide conveniently parameterised dynamic models for many non-linear systems for use in adaptive control. Properties of static neural networks, which enabled successful design of stable adaptive control in the state feedback case, are also identified. A survey of the existing results is presented which puts them in a systematic framework showing their relation to classical self-tuning adaptive control application of neural control to a SISO/MIMO control. Simulation results demonstrate that the self-tuning design methods may be practically applicable to a reasonably large class of unknown linear and non-linear dynamic control systems.

Stochastic beamforming for cochlear implant coding

Cochlear implants are prosthetic devices used to provide hearing to people who would otherwise be profoundly deaf. The deliberate addition of noise to the electrode signals could increase the amount of information transmitted, but standard cochlear implants do not replicate the noise characteristic of normal hearing because if noise is added in an uncontrolled manner with a limited number of electrodes then it will almost certainly lead to worse performance. Only if partially independent stochastic activity can be achieved in each nerve fibre can mechanisms like suprathreshold stochastic resonance be effective. We are investigating the use of stochastic beamforming to achieve greater independence. The strategy involves presenting each electrode with a linear combination of independent Gaussian noise sources. Because the cochlea is filled with conductive salt solutions, the noise currents from the electrodes interact and the effective stimulus for each nerve fibre will therefore be a different weighted sum of the noise sources. To some extent therefore, the effective stimulus for a nerve fibre will be independent of the effective stimulus of neighbouring fibres. For a particular patient, the electrode position and the amount of current spread are fixed. The objective is therefore to find the linear combination of noise sources that leads to the greatest independence between nerve discharges. In this theoretical study we show that it is possible to get one independent point of excitation (one null) for each electrode and that stochastic beamforming can greatly decrease the correlation between the noise exciting different regions of the cochlea. © 2007 Copyright SPIE - The International Society for Optical Engineering.