Learning from ordered sets and applications in collaborative ranking

Ranking over sets arise when users choose between groups of items. For example, a group may be of those movies deemed 5 stars to them, or a customized tour package. It turns out, to model this data type properly, we need to investigate the general combinatorics problem of partitioning a set and ordering the subsets. Here we construct a probabilistic log-linear model over a set of ordered subsets. Inference in this combinatorial space is highly challenging: The space size approaches (N!/2)6.93145N+1 as N approaches infinity. We propose a split-and-merge Metropolis-Hastings procedure that can explore the state-space efficiently. For discovering hidden aspects in the data, we enrich the model with latent binary variables so that the posteriors can be efficiently evaluated. Finally, we evaluate the proposed model on large-scale collaborative filtering tasks and demonstrate that it is competitive against state-of-the-art methods.

Unique IL-13Rα2-based HIV-1 vaccine strategy to enhance mucosal immunity, CD8(+) T-cell avidity and protective immunity.

We have established that mucosal immunization can generate high-avidity human immunodeficiency virus (HIV)- specific CD8þ T cells compared with systemic immunization, and interleukin (IL)-13 is detrimental to the functional avidity of these T cells. We have now constructed two unique recombinant HIV-1 vaccines that co-express soluble or membrane-bound forms of the IL-13 receptor a2 (IL-13Ra2), which can ‘‘transiently’’ block IL-13 activity at the vaccination site causing wild-type animals to behave similar to an IL-13 KO animal. Following intranasal/intramuscular prime-boost immunization, these IL-13Ra2-adjuvanted vaccines have shown to induce (i) enhanced HIV-specific CD8þ Tcells with higher functional avidity, with broader cytokine/chemokine profiles and greater protective immunity using a surrogate mucosal HIV-1 challenge, and also (ii) excellent multifunctional mucosal CD8þ T-cell responses, in the lung, genito-rectal nodes (GN), and Peyer’s patch (PP). Data revealed that intranasal delivery of these IL-13Ra2-adjuvanted HIV vaccines recruited large numbers of unique antigen-presenting cell subsets to the lung mucosae, ultimately promoting the induction of high-avidity CD8þ Tcells. We believe our novel IL-13R cytokine trap vaccine strategy offers great promise for not only HIV-1, but also as a platform technology against range of chronic infections that require strong sustained high-avidity mucosal/systemic immunity for protection.

Exploiting side information in distance dependent Chinese restaurant processes for data clustering

Multimedia contents often possess weakly annotated data such as tags, links and interactions. The weakly annotated data is called side information. It is the auxiliary information of data and provides hints for exploring the link structure of data. Most clustering algorithms utilize pure data for clustering. A model that combines pure data and side information, such as images and tags, documents and keywords, can perform better at understanding the underlying structure of data. We demonstrate how to incorporate different types of side information into a recently proposed Bayesian nonparametric model, the distance dependent Chinese restaurant process (DD-CRP). Our algorithm embeds the affinity of this information into the decay function of the DD-CRP when side information is in the form of subsets of discrete labels. It is flexible to measure distance based on arbitrary side information instead of only the spatial layout or time stamp of observations. At the same time, for noisy and incomplete side information, we set the decay function so that the DD-CRP reduces to the traditional Chinese restaurant process, thus not inducing side effects of noisy and incomplete side information. Experimental evaluations on two real-world datasets NUS WIDE and 20 Newsgroups show exploiting side information in DD-CRP significantly improves the clustering performance.

A preliminary study of the 20-m multistage shuttle run as a predictor of peak VO2 in Hong Kong Chinese students

The purpose of the present study was to determine the validity of the 20-meter multistage shuttle run (MSR) for predicting peak VO2 in Hong Kong Chinese students, ages 12-15 years. Fifty-five subjects, 27 boys and 28 girls, performed the MSR in the school environment and had peak VO2 determined in the laboratory. A correlation of 0.72 (p is less than 0.001) was found between peak VO2 and predicted peak VO2 using an equation previously developed with Canadian children (6). However, maximal shuttle run speed alone was a better predictor in this group (r=0.74, SEE=4.6 ml.kg-1.min-1, p is less than 0.001). Multiple-regression analysis (best-subsets) was performed and the best predictor variables were maximal speed and sex with either triceps skinfold or weight. For practical application in the school setting, the equation peak VO2=24.2-5.0(sex) minus 0.8(age) plus 3.4(maximal speed) (r=0.82, SEE=4.0), where for sex, male = 0 and female = 1, is suggested.

HIV-1 Infection of T cells and macrophages are differentially modulated by virion-associated Hck: a nef-dependent phenomenon

The proline repeat motif (PxxP) of Nef is required for interaction with the SH3 domains of macrophage-specific Src kinase Hck. However, the implication of this interaction for viral replication and infectivity in macrophages and T lymphocytes remains unclear. Experiments in HIV-1 infected macrophages confirmed the presence of a Nef:Hck complex which was dependent on the Nef proline repeat motif. The proline repeat motif of Nef also enhanced both HIV-1 infection and replication in macrophages, and was required for incorporation of Hck into viral particles. Unexpectedly, wild-type Hck inhibited infection of macrophages, but Hck was shown to enhance infection of primary T lymphocytes. These results indicate that the interaction between Nef and Hck is important for Nef-dependent modulation of viral infectivity. Hck-dependent enhancement of HIV-1 infection of T cells suggests that Nef-Hck interaction may contribute to the spread of HIV-1 infection from macrophages to T cells by modulating events in the producer cell, virion and target cell.

Analysis of serious non-AIDS events among HIV-infected adults at Latin American sites

Acquired immune deficiency appears to be associated with serious non-AIDS (SNA)-defining conditions such as cardiovascular disease, liver and renal insufficiency and non-AIDS-related malignancies. We analysed the incidence of, and factors associated with, several SNA events in the LATINA retrospective cohort.

Foxp3 expression in macrophages associated with RENCA tumors in mice.

The transcription factor Foxp3 represents the most specific functional marker of CD4+ regulatory T cells (TRegs). However, previous reports have described Foxp3 expression in other cell types including some subsets of macrophages, although there are conflicting reports and Foxp3 expression in cells other than Treg is not well characterized. We performed detailed investigations into Foxp3 expression in macrophages in the normal tissue and tumor settings. We detected Foxp3 protein in macrophages infiltrating mouse renal cancer tumors injected subcutaneously or in the kidney. Expression was demonstrated using flow cytometry and Western blot with two individual monoclonal antibodies. Further analyses confirmed Foxp3 expression in macrophages by RT PCR, and studies using ribonucleic acid-sequencing (RNAseq) demonstrated a previously unknown Foxp3 messenger (m)RNA transcript in tumor-associated macrophages. In addition, depletion of Foxp3+ cells using diphtheria toxin in Foxp3DTR mice reduced the frequency of type-2 macrophages (M2) in kidney tumors. Collectively, these results indicate that tumor-associated macrophages could express Foxp3.

A novel aggregate gene selection method for microarray data classification

This paper introduces a novel method for gene selection based on a modification of analytic hierarchy process (AHP). The modified AHP (MAHP) is able to deal with quantitative factors that are statistics of five individual gene ranking methods: two-sample t-test, entropy test, receiver operating characteristic curve, Wilcoxon test, and signal to noise ratio. The most prominent discriminant genes serve as inputs to a range of classifiers including linear discriminant analysis, k-nearest neighbors, probabilistic neural network, support vector machine, and multilayer perceptron. Gene subsets selected by MAHP are compared with those of four competing approaches: information gain, symmetrical uncertainty, Bhattacharyya distance and ReliefF. Four benchmark microarray datasets: diffuse large B-cell lymphoma, leukemia cancer, prostate and colon are utilized for experiments. As the number of samples in microarray data datasets are limited, the leave one out cross validation strategy is applied rather than the traditional cross validation. Experimental results demonstrate the significant dominance of the proposed MAHP against the competing methods in terms of both accuracy and stability. With a benefit of inexpensive computational cost, MAHP is useful for cancer diagnosis using DNA gene expression profiles in the real clinical practice.

Sensitivity to MK-801 in phospholipase C-β1 knockout mice reveals a specific NMDA receptor deficit

Phospholipase C-β1 (PLC-β1) is a critical component of multiple signalling pathways downstream of neurotransmitter receptors. Mice lacking this enzyme display a striking behavioural phenotype with relevance to human psychiatric disease. Glutamatergic dysfunction is strongly associated with several abnormal behavioural states and may underlie part of the phenotype of the phospholipase C-β1 knockout (KO) mouse. A heightened response to glutamatergic psychotomimetic drugs is a critical psychosis-related endophenotype, and in this study it was employed as a correlate of glutamatergic dysfunction. Control (n=8) and PLC-β1 KO mice (n=6) were treated with MK-801, a NMDA receptor (NMDAR) antagonist, following either standard housing or environmental enrichment, and the motor function and locomotor activity thus evoked was assessed. In addition, MK-801 binding to the NMDAR was evaluated through radioligand autoradiography in post-mortem tissue (on a drug-naive cohort). We have demonstrated a significantly increased sensitivity to the effects of the NMDA antagonist MK-801 in the PLC-β1 KO mouse. In addition, we found that this mouse line displays reduced hippocampal NMDAR expression, as measured by radioligand binding. We previously documented a reversal of specific phenotypes in this mouse line following housing in an enriched environment. Enrichment did not alter this heightened MK-801 response, nor NMDAR expression, indicating that this therapeutic intervention works on specific pathways only. These findings demonstrate the critical role of the glutamatergic system in the phenotype of the PLC-β1 KO mouse and highlight the role of these interconnected signalling pathways in schizophrenia-like behavioural disruption. These results also shed further light on the capacity of environmental factors to modulate subsets of these phenotypes.

Hidden Markov models for cancer classification using gene expression profiles

This paper introduces an approach to cancer classification through gene expression profiles by designing supervised learning hidden Markov models (HMMs). Gene expression of each tumor type is modelled by an HMM, which maximizes the likelihood of the data. Prominent discriminant genes are selected by a novel method based on a modification of the analytic hierarchy process (AHP). Unlike conventional AHP, the modified AHP allows to process quantitative factors that are ranking outcomes of individual gene selection methods including t-test, entropy, receiver operating characteristic curve, Wilcoxon test and signal to noise ratio. The modified AHP aggregates ranking results of individual gene selection methods to form stable and robust gene subsets. Experimental results demonstrate the performance dominance of the HMM approach against six comparable classifiers. Results also show that gene subsets generated by modified AHP lead to greater accuracy and stability compared to competing gene selection methods, i.e. information gain, symmetrical uncertainty, Bhattacharyya distance, and ReliefF. The modified AHP improves the classification performance not only of the HMM but also of all other classifiers. Accordingly, the proposed combination between the modified AHP and HMM is a powerful tool for cancer classification and useful as a real clinical decision support system for medical practitioners.

Cover, not caging, influences chronic physiological stress in a ground-nesting bird

Predator exclosures ('nest cages') around nests are increasingly used to enhance hatching success of declining ground-nesting birds. However, such exclosures are contentious and have been suggested to have detrimental effects on the species which they aim to protect. This study examines whether exclosures increase physiological stress of incubating birds, a hitherto unrecognised and untested potential drawback of exclosures. Red-capped plover Charadrius ruficapillus hatching success was radically altered and significantly higher for nests with exclosures (96.2%) compared with those without (6.8%). Chronic physiological stress in parents (as measured by the heterophil/lymphocyte [H/L] ratio in blood) did not vary between nests with and without exclosures, or between the sexes. However the absence of vegetative cover at the nest site was associated with a 62.7% elevation in H/L ratio, indicating that incubating birds which place their nests in the open are subject to increased levels of chronic stress. The results from this study demonstrate the fundamental importance of predation for the nesting success of this species and confirm that chronic stress levels are not a detrimental side effect of exclosure use.

The construction of a subset of ICNP® for patients with dementia: a Delphi consensus and a group interview study

BACKGROUND: The International Classification for Nursing Practice (ICNP®) 2013 includes over 4000 concepts for global nursing diagnoses, outcomes and interventions and is a large and complex set of standardised nursing concepts and expressions. Nurses may use subsets from the ICNP as concepts and expressions for research, education and clinical practice. The objective of this study was to identify and validate concepts for an ICNP subset to guide observations and documentation of nursing care for patients with dementia.

METHOD: The process model for developing ICNP subsets was followed, according to the guidelines adopted by the International Council of Nursing (ICN). To identify relevant and useful concepts for the subset, a modified form of the Delphi method was used. Six nurses working in healthcare services in three municipalities in Norway with postgraduate education in geriatric psychiatry and dementia care participated in two Delphi sessions. The participants reviewed and scored the concepts included in the suggested subset and had an opportunity to rewrite them and offer alternatives. To validate the subset after the Delphi study, a group interview was conducted with six other nurses with postgraduate education in geriatric psychiatry and dementia care. The group interview was recorded and transcribed, and summative content analysis was used.

RESULTS: Suitable concepts for an ICNP subset to guide observations and documentation of nursing care for patients with dementia were identified. In total, 301 concepts were identified, including 77 nursing diagnoses, 78 outcomes and 146 nursing interventions. An increased focus on concepts to describe basic psychosocial needs such as identity, comfort, connection, inclusion and engagement was recommended by nurses in the validation process.

CONCLUSIONS: Relevant and pre-formulated nursing diagnoses, goals and interventions were identified, which can be used to develop care plans and facilitate accuracy in the documentation of individuals with dementia. The participants believed that it may be difficult to find formulations for all steps of the nursing process. In particular, nursing diagnoses and psychosocial needs are often inadequately documented. The participants highlighted the need for the subset to contain essential information about psychosocial needs and communication.

Modelling human preferences for ranking and collaborative filtering: a probabilistic ordered partition approach

Learning preference models from human generated data is an important task in modern information processing systems. Its popular setting consists of simple input ratings, assigned with numerical values to indicate their relevancy with respect to a specific query. Since ratings are often specified within a small range, several objects may have the same ratings, thus creating ties among objects for a given query. Dealing with this phenomena presents a general problem of modelling preferences in the presence of ties and being query-specific. To this end, we present in this paper a novel approach by constructing probabilistic models directly on the collection of objects exploiting the combinatorial structure induced by the ties among them. The proposed probabilistic setting allows exploration of a super-exponential combinatorial state-space with unknown numbers of partitions and unknown order among them. Learning and inference in such a large state-space are challenging, and yet we present in this paper efficient algorithms to perform these tasks. Our approach exploits discrete choice theory, imposing generative process such that the finite set of objects is partitioned into subsets in a stagewise procedure, and thus reducing the state-space at each stage significantly. Efficient Markov chain Monte Carlo algorithms are then presented for the proposed models. We demonstrate that the model can potentially be trained in a large-scale setting of hundreds of thousands objects using an ordinary computer. In fact, in some special cases with appropriate model specification, our models can be learned in linear time. We evaluate the models on two application areas: (i) document ranking with the data from the Yahoo! challenge and (ii) collaborative filtering with movie data. We demonstrate that the models are competitive against state-of-the-arts.