Nonparametric Methods for Microarray Data Based on Exchangeability and Borrowed Power

ABSTRACT

This article proposes nonparametric inference procedures for analyzing microarray gene expression data that are reliable, robust, and simple to implement. They are conceptually transparent and require no special-purpose software. The analysis begins by normalizing gene expression data in a unique way. The resulting adjusted observations consist of gene-treatment interaction terms (representing differential expression) and error terms. The error terms are considered to be exchangeable, which is the only substantial assumption. Thus, under a family null hypothesis of no differential expression, the adjusted observations are exchangeable and all permutations of the observations are equally probable. The investigator may use the adjusted observations directly in a distribution-free test method or use their ranks in a rank-based method, where the ranking is taken over the whole data set. For the latter, the essential steps are as follows: 1. Calculate a Wilcoxon rank-sum difference or a corresponding Kruskal-Wallis rank statistic for each gene. 2. Randomly permute the observations and repeat the previous step. 3. Independently repeat the random permutation a suitable number of times. Under the exchangeability assumption, the permutation statistics are independent random draws from a null cumulative distribution function (c.d.f.) approximated by the empirical c.d.f. Reference to the empirical c.d.f. tells if the test statistic for a gene is outlying and, hence, shows differential expression. This feature is judged by using an appropriate rejection region or computing a p-value for each test statistic, taking into account multiple testing. The distribution-free analog of the rank-based approach is also available and has parallel steps which are described in the article. The proposed nonparametric analysis tends to give good results with no additional refinement, although a few refinements are presented that may interest some investigators. The implementation is illustrated with a case application involving differential gene expression in wild-type and knockout mice of an E.coli lipopoly-saccharide (LPS) endotoxin treatment, relative to a baseline untreated condition.     

Colonoscopy without Sedation

The results of ambulatory 24-h esophageal pH monitoring in 67 patients with gastroesophageal reflux symptoms (endoscopic esophagitis, n = 44; normal endoscopy, n = 23) were compared with those of 27 normal subjects without reflux symptoms. Patients with reflux symptoms had significantly increased gastroesophageal reflux compared with normal subjects. Acid reflux time was significantly (p < 0.001) correlated with the severity of endoscopic esophagitis. Linear discriminant analysis was used to differentiate, for each reflux variable, between patients and controls. When the percentage of overall time at pH below 4 was used as a single determinant of gastroesophageal reflux, the sensitivity and specificity were 81% and 85%, respectively, with 4% as upper limit of normal. Pathologic reflux was found in 61% of the patients with negative endoscopy. Long-term ambulatory pH-metry is of clinical value in detecting pathologic reflux in symptomatic patients with negative endoscopy.     

Instrumental variables estimation and inference in the presence of many exogenous regressors

Summary We consider a standard instrumental variables model contaminated by the presence of a large number of exogenous regressors. In an asymptotic framework where this number is proportional to the sample size, we study the impact of their ratio on the validity of existing estimators and tests. When the instruments are few, the inference using the conventional 2SLS estimator and associated t and J statistics, as well as the Anderson–Rubin and Kleibergen tests, is still valid. When the instruments are many, the LIML estimator remains consistent, but the presence of many exogenous regressors changes its asymptotic variance. Moreover, the conventional bias correction of the 2SLS estimator is no longer appropriate. We provide asymptotically correct versions of bias correction for the 2SLS estimator, derive its asymptotically correct variance estimator, extend the Hansen–Hausman–Newey LIML variance estimator to the case of many exogenous regressors, and propose asymptotically valid modifications of the J overidentification tests based on the LIML and bias‐corrected 2SLS estimators.     

Physiological variables involved in heart valve substitute calcification

Biochemical, histological and genetic studies using in vitro/in vivo models have demonstrated that pathological calcification of bioprosthetic heart valves (BHV) is regulated by various mechanisms associated with physiological variables. The major objective of this review is to characterize physiological variables involved in BHV calcification. This review examines our understanding of the systemic cellular behavior and physiological regulation processes behind BHV calcification and its clinical applications.     

Influence of pitch size and age category on the physical and physiological responses of young football players during small-sided games using GPS devices

ABSTRACT

This study aimed to compare the physical and physiological responses of young football players of different categories during small-sided games (SSGs) played on different pitch sizes. Forty-eight (24 U-13 and 24 U-14) athletes played a 3 vs. 3 + 1 SSG in two experimental conditions: regular (36 × 27 m) and large pitch sizes (40 × 29 m). The total distance covered, the distances covered at different speed zones (0 to 6.9 km/h, 6.9 to 14.3, and 14.3 to 21.4), maximum heart rate, and mean heart rate were recorded. The results showed that older athletes covered larger distances during SSGs (p = 0.001; d = 0.937; large effect) and lower distances at the lowest (0–6.9 km/h) speed zone (p = 0.001; d = 0.657; moderate-to-large effect). Neither the physical nor physiological variables (except for distance covered between 14.3 and 21.4 km/h) differed between pitch sizes. This result indicates that pitch size may not impact the physical or physiological responses of U-13 and U-14 players during SSGs, but differences between categories were found. In conclusion, the development of tactical skills may be desirable to better explore the available space in the same age categories.     

Measuring socioeconomic health inequalities in presence of multiple categorical information

While many of the measurement approaches in health inequality measurement assume the existence of a ratio-scale variable, most of the health information available in population surveys is given in the form of categorical variables. Therefore, the well-known inequality indices may not always be readily applicable to measure health inequality as it may result in the arbitrariness of the health concentration index's value. In this paper, we address this problem by changing the dimension in which the categorical information is used. We therefore exploit the multi-dimensionality of this information, define a new ratio-scale health status variable and develop positional stochastic dominance conditions that can be implemented in a context of categorical variables. We also propose a parametric class of population health and socioeconomic health inequality indices. Finally we provide a twofold empirical illustration using the Joint Canada/United States Surveys of Health 2004 and the National Health Interview Survey 2010.     

Not just data: A method for improving prediction with knowledge

Many medical conditions are only indirectly observed through symptoms and tests. Developing predictive models for such conditions is challenging since they can be thought of as ‘latent’ variables. They are not present in the data and often get confused with measurements. As a result, building a model that fits data well is not the same as making a prediction that is useful for decision makers. In this paper, we present a methodology for developing Bayesian network (BN) models that predict and reason with latent variables, using a combination of expert knowledge and available data. The method is illustrated by a case study into the prediction of acute traumatic coagulopathy (ATC), a disorder of blood clotting that significantly increases the risk of death following traumatic injuries. There are several measurements for ATC and previous models have predicted one of these measurements instead of the state of ATC itself. Our case study illustrates the advantages of models that distinguish between an underlying latent condition and its measurements, and of a continuing dialogue between the modeller and the domain experts as the model is developed using knowledge as well as data.     

Environmental variables associated with an increased risk of invasive aspergillosis

Information on the environmental variables that may affect the incidence of invasive aspergillosis (IA) is scarce. We sought to determine the relationship between airborne spore counts, climatic conditions and IA. We also examined whether circulating respiratory viruses predispose patients to IA in a multicentre cohort study of hospitalized adults with IA. Data on environmental mould spores, climatic conditions and circulating respiratory viruses were obtained from the Environmental Department of the Autonomous University of Barcelona, the Meteorological Service of Catalonia and the Acute Respiratory Infection Surveillance Project in Catalonia, respectively. Between 2008 and 2011, 165 patients with IA were identified. Diagnosis was based on one or more of the following: culture (125 cases), galactomannan antigen (98) and histology (34). One hundred and twenty-seven cases (77%) had criteria for probable IA and the remainder for proven IA. Environmental mould spore counts from the period 28–42 days preceding infection presented significant associations with admissions due to IA. None of the climatic conditions were associated with an increased risk of IA, but the presence of circulating respiratory viruses was associated with a higher risk of infection: the most strongly associated viruses were respiratory syncytial virus, influenza A(H1N1)pdm09 and adenovirus. In conclusion, the presence of high numbers of spores in the air increases the risk of admission due to IA. Circulating respiratory viruses appear to be associated with a higher risk of developing IA. Physicians should be aware of this association in order to optimize prevention and diagnosis strategies for IA during viral epidemic periods.     

Interrater Reliability of Criteria Used in Assessing Blunt Head Injury Patients for Intracranial Injuries

OBJECTIVE: To determine the interrater reliability of potential predictor variables that may be used to construct a clinical decision rule for emergency computed tomography of the head in blunt head injury victims. METHODS: As a substudy of the NEXUS II study of blunt head trauma, physicians from 21 different emergency departments performed paired evaluations of patients undergoing computed tomography of the head after blunt head injury. Each physician independently determined, for each subject, the presence or absence of each of 19 separate clinical characteristics. The physicians were either residents or attending physicians in the participating emergency departments. Paired responses on a sample of 3,951 patients were compared for raw level of agreement and for interrater concordance using the kappa statistic. If the lower margin of the 95% confidence interval for raw agreement was at least 85% and was equal to or greater than 0.50 for kappa, this was predetermined to represent substantial interrater agreement. RESULTS: There was substantial interobserver agreement by both measures for 17 of the 19 candidate variables in patients with blunt head trauma. Interobserver agreement was substantial for all candidate variables except presence of seizure (kappa = 0.57 [95% CI = 0.47 to 0.67]; raw agreement = 96.5%) and abnormal cerebellar function (kappa = 0.54 [95% CI = 0.41 to 0.67]; raw agreement = 96.5%). CONCLUSIONS: The clinicians in our study had a substantial level of agreement regarding most clinical criteria assessed in this large sample of patients with blunt head injury.