Classification Using Generalized Partial Least Squares

The advances in computational biology have made simultaneous monitoring of thousands of features possible. The high throughput technologies not only bring about a much richer information context in which to study various aspects of gene functions but they also present challenge of analyzing data with large number of covariates and few samples. As an integral part of machine learning, classification of samples into two or more categories is almost always of interest to scientists. In this paper, we address the question of classification in this setting by extending partial least squares (PLS), a popular dimension reduction tool in chemometrics, in the context of generalized linear regression based on a previous approach, Iteratively ReWeighted Partial Least Squares, i.e. IRWPLS (Marx, 1996). We compare our results with two-stage PLS (Nguyen and Rocke, 2002A; Nguyen and Rocke, 2002B) and other classifiers. We show that by phrasing the problem in a generalized linear model setting and by applying bias correction to the likelihood to avoid (quasi)separation, we often get lower classification error rates.

Gauss-Seidel Estimation of Generalized Linear Mixed Models with Application to Poisson Modeling of Spatially Varying Disease Rates

Generalized linear mixed models (GLMMs) provide an elegant framework for the analysis of correlated data. Due to the non-closed form of the likelihood, GLMMs are often fit by computational procedures like penalized quasi-likelihood (PQL). Special cases of these models are generalized linear models (GLMs), which are often fit using algorithms like iterative weighted least squares (IWLS). High computational costs and memory space constraints often make it difficult to apply these iterative procedures to data sets with very large number of cases. This paper proposes a computationally efficient strategy based on the Gauss-Seidel algorithm that iteratively fits sub-models of the GLMM to subsetted versions of the data. Additional gains in efficiency are achieved for Poisson models, commonly used in disease mapping problems, because of their special collapsibility property which allows data reduction through summaries. Convergence of the proposed iterative procedure is guaranteed for canonical link functions. The strategy is applied to investigate the relationship between ischemic heart disease, socioeconomic status and age/gender category in New South Wales, Australia, based on outcome data consisting of approximately 33 million records. A simulation study demonstrates the algorithm's reliability in analyzing a data set with 12 million records for a (non-collapsible) logistic regression model.

Basic and clinical immunology of Siglecs

Siglecs are cell-surface proteins found primarily on hematopoietic cells. By definition, they are members of the immunoglobulin gene super-family and bind sialic acid. Most contain cytoplasmic tyrosine motifs implicated in cell signaling. This review will first summarize characteristics common and unique to Siglecs, followed by a discussion of each human Siglec in numerical order, mentioning in turn its closest murine ortholog or paralog. Each section will describe its pattern of cellular expression, latest known immune functions, ligands, and signaling pathways, with the focus being predominantly on CD33-related Siglecs. Potential clinical and therapeutic implications of each Siglec will also be covered.

Linear and Piecewise Functions

This morning Dr. Battle will review basic concepts of linear functions and piecewise functions and how they can be used as models for real-world applications. She will also introduce techniques for using a spreadsheet to analyze data.

Transfer Functions in Artificial Neural Networks - A Simulation-Based Tutorial

Artificial neural networks are based on computational units that resemble basic information processing properties of biological neurons in an abstract and simplified manner. Generally, these formal neurons model an input-output behaviour as it is also often used to characterize biological neurons. The neuron is treated as a black box; spatial extension and temporal dynamics present in biological neurons are most often neglected. Even though artificial neurons are simplified, they can show a variety of input-output relations, depending on the transfer functions they apply. This unit on transfer functions provides an overview of different transfer functions and offers a simulation that visualizes the input-output behaviour of an artificial neuron depending on the specific combination of transfer functions.

Design and Impact of Awareness Functions – A Study about Social Media in Virtual Teams

The usage of social media in leisure time settings has become a prominent research topic. However, less research has been done on the design of social media in collaboration settings. In this study, we investigate how social media can support asynchronous collaboration in virtual teams and specifically how they can increase activity awareness. On the basis of an open source social networking platform, we present two prototype designs: a standard platform with basic support for information processing, communication and process – as suggested by Zigurs and Buckland (1998) – and an advanced platform with additional support for activity awareness via specialfeed functions. We argue that the standard platform already conveys activity awareness to a certain extent, however, that this awareness can be increased even more by the feeds in the advanced platform. Both prototypes are tested in a field experiment and evaluated with respect to their impact on perceived activity awareness, coordination and satisfaction. We show that the advanced design increases coordination and satisfaction through increased perceived activity awareness.

Targeting GRPR in urological cancers--from basic research to clinical application

Gastrin releasing peptide (GRP) is a regulatory peptide that acts through its receptor (GRPR) to regulate physiological functions in various organs. GRPR is overexpressed in neoplastic cells of most prostate cancers and some renal cell cancers and in the tumoral vessels of urinary tract cancers. Thus, targeting these tumours with specifically designed GRP analogues has potential clinical application. Potent and specific radioactive, cytotoxic or nonradioactive GRP analogues have been designed and tested in various animal tumour models with the aim of receptor targeting for tumour diagnosis or therapy. All three categories of compound were found suitable for tumour targeting in animal models. The cytotoxic and nonradioactive GRP analogues have not yet shown convincing tumour-reducing effects in human trials; however, the first clinical studies of radioactive GRP analogues--both agonists and antagonists--suggest promising opportunities for both diagnostic tumour imaging and radiotherapy of prostate and other GRPR-expressing cancers.

Generalized fuzzy clustering for segmentation of multi-spectral magnetic resonance images.

An integrated approach for multi-spectral segmentation of MR images is presented. This method is based on the fuzzy c-means (FCM) and includes bias field correction and contextual constraints over spatial intensity distribution and accounts for the non-spherical cluster's shape in the feature space. The bias field is modeled as a linear combination of smooth polynomial basis functions for fast computation in the clustering iterations. Regularization terms for the neighborhood continuity of intensity are added into the FCM cost functions. To reduce the computational complexity, the contextual regularizations are separated from the clustering iterations. Since the feature space is not isotropic, distance measure adopted in Gustafson-Kessel (G-K) algorithm is used instead of the Euclidean distance, to account for the non-spherical shape of the clusters in the feature space. These algorithms are quantitatively evaluated on MR brain images using the similarity measures.

Improving executive functions in 5- and 6-year-olds: Evaluation of a small group intervention in prekindergarten and kindergarten children

Research suggests a central role of executive functions for children's cognitive and social development during preschool years, especially in promoting school readiness. Interventions aiming to improve executive functions are therefore being called for. The present study examined the effect of a small group intervention implemented in kindergarten settings focusing on basic components of executive functions, i.e., working memory, interference control and cognitive flexibility. A total of 135 children enrolled in Swiss prekindergarten (5-year-olds) and kindergarten (6-year-olds) were involved. Results revealed that the small group intervention promoted gains in all three included components of executive functions: prekindergarten children substantially improved their working memory and cognitive flexibility processes, whereas significant training effects were found for the kindergarten children in interference control. Implications of these findings for early intervention programs and for tailoring preschool curricula are discussed, particularly with respect to children's school readiness. Copyright © 2011 John Wiley & Sons, Ltd.

The Gut Microbiome and Cirrhosis: Basic Aspects

In this chapter the basic aspects helping to understand the microbiome in terms of quantity, diversity, complexity, function, and interaction with the host are discussed. First the nomenclature, definitions of taxa, and measures of diversity as well as methods to unravel this kingdom are outlined. A brief summary on its physiological relevance for general health and the functions exerted specifically by the microbiome is presented. Differences in the composition of the microbiome along the gastrointestinal tract and across the gut wall and its interindividual variations, enterotypes, and stability are highlighted. The reader will be familiarized with all different modulators impacting on the microbiome, namely, intrinsic and extrinsic factors. Intrinsic factors include gastrointestinal secretions (gastric acid, bile, pancreatic juice, mucus), antimicrobial peptides, motility, enteric nervous system, and host genotype. Extrinsic factors are mainly dietary choices, hygiene, stress, alcohol consumption, exercise, and medications. The second part of the chapter focuses on quantitative and qualitative changes in microbiome in liver cirrhosis. The mechanisms contributing to dysbiosis, small intestinal bacterial overgrowth, and bacterial translocation are delineated underscoring their role for the liver-gut axis.

COMMUNITY-FOCUSED FUNCTIONS IN PUBLIC HEALTH NURSING: A COMPARISON OF PERCEPTIONS

This descriptive cross-sectional survey compared the perceptions of public health nursing practitioners, educators and administrators along two dimensions: the importance of community-focused functions in public health nursing and which occupational categories in public health are responsible for those functions. More than 50 percent of the mailed questionnaires that were sent to a systematic stratified nationwide sample of public health nurses were returned. In general, respondents: were female, were in their 40s, received their basic nursing education in baccalaureate programs, had either a baccalaureate or a master's degree, worked in official agencies or schools, and had approximately 14 years of experience in public health with six in their present position.^ Significant differences between practitioners, educators and administrators were found in their perceptions of both the importance of community-focused functions in public health nursing and in which occupational category they indicated as having the major responsibility to perform those functions. Educators and administrators perceived community-focused functions as more important than did practitioners. Overall the occupational category of administrator was indicated as having the major responsibility for performing community-focused functions.^

A matrix pencil approach to the existence of compactly supported reconstruction functions in average sampling

The aim of this work is to solve a question raised for average sampling in shift-invariant spaces by using the well-known matrix pencil theory. In many common situations in sampling theory, the available data are samples of some convolution operator acting on the function itself: this leads to the problem of average sampling, also known as generalized sampling. In this paper we deal with the existence of a sampling formula involving these samples and having reconstruction functions with compact support. Thus, low computational complexity is involved and truncation errors are avoided. In practice, it is accomplished by means of a FIR filter bank. An answer is given in the light of the generalized sampling theory by using the oversampling technique: more samples than strictly necessary are used. The original problem reduces to finding a polynomial left inverse of a polynomial matrix intimately related to the sampling problem which, for a suitable choice of the sampling period, becomes a matrix pencil. This matrix pencil approach allows us to obtain a practical method for computing the compactly supported reconstruction functions for the important case where the oversampling rate is minimum. Moreover, the optimality of the obtained solution is established.

A Web Implementation of A Generalized NEP

The Networks of Evolutionary Processors (NEPs) are computing mechanisms directly inspired from the behavior of cell populations more specifically the point mutations in DNA strands. These mechanisms are been used for solving NP-complete problems by means of a parallel computation postulation. This paper describes an implementation of the basic model of NEP using Web technologies and includes the possibility of designing some of the most common variants of it by means the use of the web page design which eases the configuration of a given problem. It is a system intended to be used in a multicore processor in order to benefit from the multi thread use.

Leader Rural Development Program in Spain: Profile and Functions of Workers in Local Action Groups.

Occupations in the labor market are linked with to a minimum basic training and other capacities. Hired workers should be able to accomplish required functions related to their specific job. Regarding the rural development labor market, local action groups? workers have defined performance areas?projects, strategy, organization and training & market?but specific functions within each of these areas are not as clearly defined. Neither both, basic training and capacities needed to perform each job profile within the local action group are defined. This communication analyses training and other capacities linked to each of the job profiles within the local action group. Functions within each of the performance areas previously defined are also analyzed regarding the job profiles.