2018 session 1 ================ Activation of Sonic hedgehog signaling in ventricular cardiomyocytes exerts cardioprotection -------------------------------------------------------------------------------------------- against ischemia reperfusion injuries ------------------------------------- Statistics. Presented numerical data are Mean +/- standard error of mean (SEM). Results were considered significant if p , 0.05, with one-way, two-tailed Analysis of Variance (ANOVA) with Bonferroni post-test. Further details on methods are included in Supplementary material online. Interactions Between beta-Catenin and Transforming Growth Factor-beta Signaling Pathways ----------------------------------------------------------------------------------------- Mediate Epithelial-Mesenchymal Transition and[...] --------------------------------------------------- Statistical Analysis : data are shown as means ± S.E.; n is the number of observations. We used z-tests to determine whether the ratiometric data (i.e. normalized against control) are different from control. p < 0.05 was considered significant. The severity of NAFLD is associated with gut dysbiosis and shift in the metabolic function of the gut microbiota ---------------------------------------------------------------------------------------------------------------- Statistics : Quantitative variables were expressed as median with 1st and 3rd quartiles into brackets. Raw observation counts in taxa summary plots were normalized by calculating relative abundance. Qualitative variables were compared using the Fisher's exact test and quantitative variable using the Mann Whitney test. A p value less than 0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 18.0 software (IBM, Armonk, NY, USA).! Melanoma antigen A12 regulates cell cycle via tumor suppressor p21 expression ----------------------------------------------------------------------------- [...]with data expressed as the ratio of values for cells transfected with MAGE-A12 to values for the control siRNAs (mean ± SD; n = 3). *** p < 0.001 by t-test. Mild palmitate treatment increases mitochondrial mass but does not affect EA.hy926 endothelial cells viability -------------------------------------------------------------------------------------------------------------- Data are shown as means of ratios of treatment to control values ± SD, for the number of replicates indicated in figure legends. Confidence intervals for geometric means of ratios between treatment and appropriate controls were found using the t method for paired design (TTEST procedure) [27] with 95% confidence level. Associated p-values for the null hypothesis that the ratio equals one are also reported. Functional diversity and properties of multiple xylanases from Penicillium oxalicum GZ-2 ---------------------------------------------------------------------------------------- The transcript level of xyn11A was significantly higher (P < 0.05) than that of the other xylanase genes. pH-responsive lipid nanocapsules for enhanced cellular uptake ------------------------------------------------------------- Results (n=4) are expressed as mean measure ± standard deviation. Kruskal-Wallis, post-hoc Dunn's, correction Hochberg. For statistical analysis of size, BLK LNC was used as reference, ***p<0.001 , ** p< 0.01, *p<0.05. Characterisation of adipocyte-derived extracellular vesicle subtypes[...] ------------------------------------------------------------------------- All statistical procedures were carried out under the R v3.1.1 environment [31] using base packages developed by the PAPPSO platform (http://pappso.inra.fr/index.php?). Only peptides present in less than 2% of the sample and belonging to multiple proteins were removed. Changes in protein abundance were detected by analysis of variance. For both methods (XIC and SC), proteins were considered significant when the adjusted p value was < 0.05. The protein content of each EV pellet normalised to the protein content of the original productive 3T3-L1 adipocytes is presented. n=3-9 independent EVpreparations, **p<0.01, (Mann-Whitney rank test). 3T3-L1 adipocytes secrete substantiallymore sEVs than lEVs in serum-deprived conditions. n = 4 biological samples for lEVs and for sEVs, *p < 0.05 (Mann-Whitney rank test). 2017 session 2 ============== Canavate et al : interspecific variability in phosphorus-induces lipid (eukaryotic phytoplankton) ------------------------------------------------------------------------------------------------- The significance of differences in BL, PL, BL : PL, SQDG : PG and DGDG : PC as a result of phosphate treatment was determined at different culture stages by performing a one-way ANOVA of the log + 1 transformed data, followed by a Tukey post hoc test. Permutational analysis of variance (PERMANOVA; Anderson, 2001) was conducted using the PERMANOVA+ add-on package of PRIMER v.6 (Clarke & Gorley, 2006) for comparison between lipid class profiles. The data were converted into similarity triangular matrices using a BrayCurtis resemblance measure (Clarke & Warwick, 2001). Each term in the analysis was tested using 999 unique permutations, with significant terms investigated using a posteriori pair-wise comparisons with the PERMANOVA t statistic. Patterns of similarity between lipid class profiles were determined after similarity percentage (SIMPER) analysis to identify those lipids that contributed more to between-group similarities, according to Clarke (1993). The criteria for assigning relevance to percentages was based on dissimilarity to SD ratios (Diss : SD) > 2.0 and percentage contributions > 10%. SD is a useful measure of how consistently each lipid class contributes to dissimilarity. If average dissimilarity is large and SD small (corresponding to an increased Diss : SD ratio), then that lipid class not only contributes much to the dissimilarity between groups but also does so consistently and it is thus a good discriminating lipid (Clarke, 1993). Principal coordinate analysis (PCO), a flexible ordination technique based on Bray-Curtis similarities that preserves actual dissimilarities in Euclidean space, was used for assessing sources of variation affecting lipid class profiles and the arranging of microalgal species according to their lipid response to phosphate availability. YUAN Peng et al : Oral etoposide monotherapy is effective for metastatic breast cancer with heavy prior therapy --------------------------------------------------------------------------------------------------------------- Statistics: Intention-to-treat (ITT) analyses were carried out for both efficacy and safety measures. Descriptive statistics were used to summarize the safety and laboratory observations. Median PFS and OS, with 95% confidence interval (CI), were estimated by Kaplan-Meier method. Response rate in subgroups was analyzed by cross-tabulation and testing. Morille, Van Thanh et al : pharmacologically active microcarriers ----------------------------------------------------------------- Statistical analysis: XLSTAT 2008 (Addinsoft Paris, France) was used for that purpose. Statistical significance for each experiment was determined by a Dunnett's test or a Student's t-test. The tests were considered as significant with p values of less than 0.05 Roman et al : Cytokinins Are Initial Targets of Light in the Control of Bud Outgrowth -------------------------------------------------------------------------------------- Statistical Analysis: Three replicates of each experiment were conducted for the morphological analysis, with at least 10 plants per repetition. Standard statistical analyses (means and SE) and graphs were made using Microsoft Excel. Statistical differences were calculated by Student's t test when comparing two conditions or by multiple analyses ANOVA using R 2.13.2 software when comparing more than two conditions (0.05 significance level). 2017 session 1 ============== Chan_et_al_2011 : Colorectal_cancer ----------------------------------- Relative risks, confidence intervals Relative risk estimates were pooled using fixed-effects and random-effects models. Means or medians of the intake categories were used when reported in the articles; if not reported, midpoints were assigned to the relative risk of the corresponding category. To assess heterogeneity, we computed the Cochran Q test and I 2 statistic. Sources of heterogeneity were explored in stratified analysis and by linear meta-regression. We further examined the potential non-linear dose-response relationship between red and processed meats and colorectal cancer using fractional polynomial models. All analyses were conducted using Stata version 9.2 Chao_de_la_Barca_ABC_2015 : Métabolomique et spectrométrie de masse ------------------------------------------------------------------- On obtient une matrice d'ordre n×k 2 des intensités ou des surfaces sous la courbe de chaque pic. C'est cette matrice qui est utilisée pour les analyses statistiques. Le but est, à ce stade, d'identifier les variables importantes dans la séparation des groupes. On dispose toujours d'une infor- mation supplémentaire : l'appartenance de chacun des n échantillons à l'un des groupes (malades/non malades ou traités/contrôles, etc.). Cette information est codée comme une variable réponse qualitative à plusieurs modalités (ex : 0-malades et 1-non malades) que l'on va chercher à « expli- quer » par les variables de la matrice n×k 2 appelées alors variables explicatives. Les méthodes statistiques utilisées par la suite sont supervisées ou non selon qu'elles tiennent compte de l'appartenance des échantillons à chaque groupe. L'analyse en composantes principales (ACP) est une des méthodes non supervisées les plus utilisées. L'ACP va permettre la détection des groupes homogènes d'individus (clustering) et la détection d'individus atypiques (outliers) voire aber- rants qui pourrons être écartés des analyses statistiques après consultation des métadonnées avec le biologiste. Les analyses supervisées comprennent, entre autres, la régres- sion PLS (partial least square ou projection to latent structures) couplée à l'analyse discriminante (PLS-DA) KIm, Sueng-Won et al., Medi-96-e5759.pdf : Colorectal adenoma in korean adults ------------------------------------------------------------------------------ Baseline characteristics of the study population were presented as mean with standard deviation for continuous variables and number with proportion for categorical variables. The x 2 test for categorical variables and student t test for continuous variables were performed for statistical comparisons in the study population. All nutrient intakes were adjusted by total energy intake using the residual regression method. [19] Each type of fat and fatty acid intake was stratified into quintiles to examine trends in risk by level of exposure. Odds ratios (ORs) for colorectal adenomatous polyps in the higher intake groups (Q 2 -Q 5 ) were calculated using multivariable logistic regression analysis by using the lowest (Q 1 ) as the reference group. All statistical analyses were performed using STATA version 14.1 (Stata Co., College Station, TX). A P value <0.05 was considered to represent a statistically significant outcome. Ono_et_al.2016 : Widespread Genetic Incompatibilities between First-Step Mutations ---------------------------------------------------------------------------------- In a large-scale screen for genetic interactions in which mutations in most of the *6,000 genes in the yeast S. cerevisiae were tested pairwise in 23 million double mutants (including mutations in both nonessential and essential genes)... Outlier Detection and Removal Outliers were detected by performing a two-sided Grubbs test, allowing us to detect a maximum of one outlier per strain and medium, using the R package outliers and the method grubbs.test Epistasis for maximum growth rate was assessed with mixed-effects models run on either all haploid or all diploid strains together, including the genotype at each gene, their pairwise interactions, and mating type (for the haploids) as fixed effects and plate within day as a ran- dom effect, fit using restricted maximum likelihood with the lmer function from the lme4 package in R. For diploids, the models were first run using only strains that were homozygous (either mutant or ancestral) for comparison to the haploid data. Significance of interaction terms (and mating type) was determined by performing an ANOVA between the full model and a model dropping that term using the anova function in R and fitting models using maximum likelihood. To determine the type of epistasis present for each pair of genes, the package lsmeans [60] was used to both determine the least-squares mean for each strain in the model and to make comparisons between strains using the contrast function. The type of epistasis was determined by comparing the double mutant to each single mutant and each single mutant to the ancestor, and only these planned comparisons were performed. The p-value was adjusted for the num- ber of tests performed using the multivariate t distribution (mvt method) in lsmeans. To be conservative, we based our categorization of epistasis solely on statistically significant differ- ences. For example, if the double mutant had a lower growth rate than both single mutants but this difference was only significant in one of the two cases, it was considered an example of sign epistasis (significantly lower than one single mutant but not the other) rather than recip- rocal sign epistasis. A similar procedure was then undertaken including heterozygous diploid strains. A model was run using the lmer function including all diploid strains together, with plate within day as a random effect. Least-squares means were determined for all diploid genotype For the tolerance assay assessed across a range of concentrations of nystatin, we performed Welch's t-tests of OD after 24 hours... Data and analyses deposited in the Dryad repository: http://dx.doi.org/10.5061/dryad.vs370. Brunel-Muguet, Mary, Durr : Seed nitrogen reserves and external nitrogen uptake ------------------------------------------------------------------------------- Data fitting and statistical analyses Two-way ANOVA was performed to test the effects of genotype (G), N supply (S) and G x S interactions of the measured variables (STATGRAPHICS Plus 3.1. software). The N supply in each genotype was compared using Bonferroni's multiple comparison procedure. For a given level of N supply, the genotype effect was tested using a one-way ANOVA. A model was built to calculate N fluxes to estimate the relative proportions of nitrogen originating from mobilisation of seed reserves and N absorption using 15N labelling (Fig. 1). To estimate N fluxes, the non-parametric Kruskal-Wallis procedure was performed to test the genotype and the N supply effects because sample sizes were reduced (two replicates per modality i.e. n=8 and n=4 for testing the genotype and the N supply effects respectively) and the variance equality was verified (Levene's test). 2016 session 2 ============== Konradsen et al. : interleukin-26 as a biomarker in pediatric asthma -------------------------------------------------------------------- To address our hypothesis, we utilized data from a previously well-characterized cohort of school-age children (n = 76) with severe uncontrolled (uncontrolled) or persistent controlled (controlled) asthma. Boxplots in Fig. 1 Median concentrations (logarithmically transformed) of IL-26 (ng/mL) in induced sputum from children with uncontrolled versus controlled asthma and a low concentrations of eosinophils in blood (<=0.3 × 10^9/L) or b low levels of exhaled nitric oxide (<= 18.3 p.p.b). Footnotes. The IL-26 concentrations are presented as medians with inter-quartile ranges The median values in the entire cohort of the respective biomarker were used to define the cut-off between high and low levels of the various biomarkers. B-EOS blood eosinophils, FENO the fraction of nitric oxide in exhaled air, in parts per billion. *Independent samples, analyzed by Mann-Whitney U-test (SPSS® version 20) Baysal et al. : induction of defence-related enzymes ----------------------------------------------------- Experimental design and statistical analyses All experiments were arranged in a completely randomized split-plot design with three replicates of 20 plants per treatment and repeated at least twice. Data obtained at different intervals after inoculation were analysed using Minitab Software, version 13·32 (Minitab Inc., State College, Pennsylvania, USA). Pearson's correlation coefficient was used to establish any relationships between disease index, treatments (time intervals between ASM treatments and inoculation) and days after inoculation. For disease severity studies, significance of differences among the treatments was determined by a nonparametric Mann-Whitney rank sum test. A simple multiple linear regression (MLR) was performed to explain variations in the response variable of bacterial growth as a function of the explanatory variables (treatments and days after inoculation). Log-transformed data were subjected to analyses of variance, and differences between treatments assessed by Student's two-sample t-test at P < 0·05. For enzyme activities, analysis of variance (anova) was carried out, and the significance of differences among the treatments was determined according to Duncan's Multiple Range Test at P < 0·05. Youm et al : ketone metabolite beta-hydroxybutyrate --------------------------------------------------- Data are expressed as mean ± sem (*P < 0.05) from cells derived from n = 12 (a-d); All bar graphs in a-e represent[...] The differences between means and the effects of treatments were determined by one-way analysis of variance (ANOVA) using Tukey's test. Khan et al : genetic analysis of metabolites in apple fruits ------------------------------------------------------------ MetaNetwork implements a two-part parametric model per trait, combining a non-parametric approach (Wilcoxon-Mann- Whitney test; Brem et al., 2002; Yvert et al., 2003) with a parametric test [analysis of variance (ANOVA)]. The non-parametric test uses a user-defined spike value to distinguish qualitative segregation from quantitative differences. The value chosen as the spike was 37, because this value was the noise level in the LC-MS analysis. MetaNetwork also allows setting a threshold for the significance of mQTL by performing permutation tests on samples. A bootstrap procedure was performed with a type I error of 5% (default value of MetaNetwork) for finding an mQTL considering all genetic markers. This procedure rendered a -10log(P) threshold of 3.8 for individual marker-trait combinations. This threshold was used for all analyses. Schuerger et al. : comparison of two hyperspectral imaging ---------------------------------------------------------- 2.8. Statistical analysis Statistical analyses were conducted with the PC-based Statistical Analysis System (SAS Institute, Cary, NC, USA). A completely randomized experimental design was used in growing bahia grass plants at the different concentrations of Zn on greenhouse benches. Plant biometric, hyperspectral imaging, and laser-induced fluorescence data were analyzed with orthogonal polynomial contrast analyses and protected least-squares mean separation tests (n = 16). Orthogonal coefficients for unevenly spaced treatments were generated with PROC IML in SAS. The a levels for various tests are listed as one of the following: (a) not significant (NS) ( P>0.05), (b) significant (*) ( PV0.05), or (c) highly significant (**) (PV0.01). A series of linear regression analyses (PROC REG in SAS) were used for describing the relationships of NDVI, RVI, BL/GR fluorescence ratios, and R/FR fluorescence ratios to plant stress induced by different Zn concentrations in nutrient solutions used to irrigate bahia grass plants. Hersant et al. : étude d'un phénomène de dégradation par usure --------------------------------------------------------------- IV. ANALYSE STATISTIQUE Entre deux essais, il est possible que certains paramètres fixes au cours du temps aient différentes valeurs. Nous devons donc réaliser une étude statistique dans le but de vérifier cela. Dans notre cas, l'étude statistique consiste à vérifier les points suivants : - Tester l'indépendance des trajectoires de dégradation : Test de Kolmogorov-Smirnov. - Dans le cas où les trajectoires sont indépendantes, tester l'indépendance des variances : LR Test ou test du rapport de vraisemblance. - Dans le cas où les variances sont égales, réaliser un test de comparaison des moyennes avec variances égales. - Dans le cas où les variances sont différentes, réaliser un de comparaison des moyennes avec des variances différentes : Test d'Aspin-Welch. Khalyfa et al. : Circulating Plasma Extracellular Microvesicle miRNA Cargo -------------------------------------------------------------------------- Data Analysis Results are presented as means ± SD, unless stated otherwise. All numerical data were subjected to statistical analysis using independent Student t tests or analysis of variance followed by post-hoc tests (Tukey) as appropriate. Chi square analysis was performed on categorical data concerning demographic characteristics of the various groups. Pearson correlation testing was conducted to establish association between several study parameters, including ECIS-derived normalized resistance changes and Tmax in endothelial function tests. Finally, canonical correlation analyses were performed to explore the relationships between sets of variables. Statistical analyses were performed using SPSS software (version 21.0; SPPS Inc., Chicago, IL). For all comparisons, a 2-tailed p < 0.05 was considered to define statistical significance.x Schulte and Tack : Weakly Monotonic Propagators ----------------------------------------------- pas de calculs statistiques. Lecomte et al. : Partial Resistance of Carrot to Alternaria dauci ----------------------------------------------------------------- Log(AUDPC) was calculated from the visual assessments, log(I+1) from the qPCR experiments. Both were subjected to variance analysis followed by a Waller-Duncan multiple comparison. As could be expected, the two parameters were closely correlated (r2 = 0.793). Interestingly, log(AUDPC) seemed to show a higher resolution, as the homogeneity groups appeared to be more numerous (4 vs 2). Homogeneity goups were calculated using the Waller-Duncan multiple comparison following an ANOVA analysis. Mina et al. : Purification and Characterization of a Mycelial Catalaset ----------------------------------------------------------------------- Statistical analysis was performed using the Wilcoxon-Mann-Whitney test, and results were considered significantly different at a P value of 0.01. Liu : Sentiment Analysis and Subjectivity ----------------------------------------- pas de calculs statistiques, un peu de probabilités. 2016 session 1 ============== Kent : ambulatory Blood Pressure Monitoring in Individuals with HIV -------------------------------------------------------------------- Statistical Analyses Meta-analyses were conducted among all studies with both HIV+ and HIV- individuals. We examined the pooled associations of HIV status with clinic SBP and DBP and with ABPM-mea- sured mean 24-hour SBP and DBP, mean daytime SBP and DBP, mean nighttime SBP and DBP, percentage nocturnal SBP and DBP declines, and non-dipping SBP pattern. For each study, the absolute differences in continuous BP phenotypes and 95% confidence intervals (CIs) were calcu- lated between HIV+ and HIV- individuals. We also calculated odds ratios (ORs) and 95% CIs for the association between HIV+ status and a non-dipping SBP pattern. To obtain overall effect esti- mates, pooled weighted mean differences for continuous BP measures, ORs for non-dipping SBP pattern, and 95% CIs for all measures were calculated from inverse-variance weighted random- effects models. Heterogeneity was examined using heterogeneity I2 and chi-squared statistics. In sensitivity analyses, all meta-analysis models were re-calculated in a leave-one-out method by individually removing each included study, one at a time, to examine whether pooled estimates and heterogeneity statistics changed. Data management and analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC) and Stata 13.1 (Stata Inc., College Station, TX). Souchu : nutrient limitation ---------------------------- Median values for the above water quality parameters were calculated for the pooled data from each lagoon or from each individual sector of lagoons within which the sector tor median water quality was found to differ significantly (Kruskal-Wallis, p < 0.05). Individual sectors were then kept for Bages-Sigean and Méjean lagoons, which both had significant between-station differences for both TN and TP. Andrieux : impact of oyster farming ----------------------------------- The normality of data sets was first assessed with the Shapiro and Wilk's test (www.anastats.fr). Spearman or Bravais-Pearson tests were used to evaluate the relationships between variables (Xlstat). Mann-Whitney tests were performed to determine the signif- icant differences in biogeochemical parameters between the Reference and Oyster stations at each study period (www.anastats.fr). For all tests, values were considered significant at p < 0.01. Kun-Darbois : botulinum toxin ----------------------------- Statistical analysis Statistical analysis was performed using the Systat statistical soft- ware release 13.0 (Systat Software Inc., San Jose, CA). All data were expressed as mean ± standard error of the mean (SEM). Differences among groups were analyzed by a non-parametric ANOVA (Kruskall- Wallis) and between groups by the Mann and Whitney's U test. Data from right and left hemimandibles were compared using a paired t-test. Differences were considered significant when p < 0.05. Paquete : pareo local optimum ------------------------------ When comparing the performance of two algorithms A and B, the null hypothesis is stated as follows H0: Given a specific instance, there is no difference in the performance between A and B in terms of the EAF; and the alternative as H1: Given a specific instance, there is a difference in the performance between A and B in terms of the EAF. Given independence between the runs of both algorithms, a suitable test statistic is the maximum absolute difference between the EAFs, which is analogous to the Kolmogorov- Smirnov test statistic. The decision about acceptance or rejection of H0 can be taken by comparing the test statistic obtained in the original samples with the critical value Cv, that is, the 1-alpha quantile of the distribution of the test statistic: H0is rejected if the former is greater or equal then Cv . Usually, alpha is between 0.01 and 0.05. Since the distribution of the test statistic is unknown, a possible way of computing this bound is to use permutation tests [12], which can be described as follows: 1. label each outcome obtained by each algorithm; 2. compute the test statistic given the original labels; 3. permute the labels and recalculate the test statistic until all permutations considered; 4. take decision on acceptance or rejection of H0 based on the distribution of the test statistic. If the total number of possible permutations is too large, one can carry out a random- ization test, by using only a limited number of random permutations in the above pro- cedure. Venkatasubramanian : process fault detection and diagnosis ---------------------------------------------------------- Methods that extract quantitative information can be broadly classified as non-statistical or statistical methods. Neural networks are an impor- tant class of non-statistical classifiers. Principal compo- nent analysis (PCA)/partial least squares (PLS) and statistical pattern classifiers form a major component of statistical feature extraction methods. The different ways in which knowledge can be extracted from process history are schematically presented in Fig. 1. We review these approaches in this part of the review paper. Reglink : linkages between aggregate formation, porosity... ------------------------------------------------------------- For each CZO, mean, minimum and maximum values of the soil chemical and physical parameters are given in Table A1 and the R2 values from linear regression analysis are given in Tables A2-4. First, the results of the soil characterization of the three CZO's will be given followed by the relations between the physical and chemical properties. Baron-Menguy : involvement of angiotensin II -------------------------------------------- Statistical analysis Results are expressed as mean +/- .e.m. Significance of the difference between arteries was determined by analysis of variance (one-factor analysis of variance or analysis of variance for consecutive measurements, when appropriate). Means were compared by paired t-test or by the Bonferroni test for multigroup comparisons. Values of P<0.05 were considered to be significant. van den Akker : redox state of transglutaminase 2 ------------------------------------------------- Statistics Data are shown as mean +/- SEM. For all measurements of P,d- curves, differences in diameter between groups were tested at each pressure level using a paired T-test with Bonferroni correction when appropriate. Distensibility was calculated as the lumen diameter of the artery at 120 mmHg divided by the diameter at 5 mmHg. For quantification of fluorescence, 3 images were averaged per vessel. In all figures, P-values smaller than 0.05 resp. 0.01 are indicated by single or double symbols (e.g. * and **). Dumont : Time-related alteration in Flow- (Shear Stress-) mediated remodeling ----------------------------------------------------------------------------- Statistical Analysis Results were expressed as means +/- standard error (SEM). Sig- nificance of the differences between groups was determined by analysis of variance (two-way ANOVA for consecutive measurements followed by the Bonferroni ð‘¡-test) to compare pressure-diameter curves in the different groups. In the other set of experiments, means were compared by unpaired Student's t-test. P values less than 0.05 were considered to be significant. Recoskie : [...] case study of the 2009 H1N1 pandemic ----------------------------------------------------- Estimating a density and assessing goodness-of-fit... Maximum likelihood estimation (MLE) when data is not grouped. Least squares estimation (LSE) when data is grouped. Measuring the centre of a distribution. Sample size and hypothesis testing Vasenev : How to map soil organic carbon stocks... ---------------------------------------------------- 2.5. Statistical analysis and modeling Normality of the distribution of SOC values was checked by Shapiro- Wilk's W test and homogeneous of variances was checked by Levene's test. One-way ANOVA was used to check significance of difference in SOC between different groups (soil types, land-use types, cities of different age, size etc.). A statistical, multiple regression model correlat- ed SOC contents to the explanatory variables to predict the topsoil and subsoil SOC contents in the area. Auxiliary information included both continuous and categorical data. Relationships between SOC contents and secondary data were analyzed using general linear models (GLMs). Categorical variables were introduced as dummy variables (equal 1 if a location was within a particular unit and 0 otherwise). The correlation between explanatory variables was analyzed in advance in order to prevent multicollinearity. All the above models were developed separately for the topsoil and the subsoil. The general linear models were obtained by a backward stepwise linear regression keeping the statistically significant variables (p < 0.05). The predictive power of each statistical model was character- ized by the coefficients of determination R2 and R2adj. Statistical analysis was performed in STATISTICA 6.0 (Borovikov, 2003). Smith : trial of the benzoquinone idebenone ------------------------------------------- Statistics Values are expressed as mean ± SEM. Statistical significance between groups was estimated by one sample, single-tailed Student's t-tests, with Bonferroni correction where necessary, and values of p <= 0.05 were considered significant. Power calculations were performed in R, and for the treatment arm a sample number of n = 14 was sufficient to give us a power of 0.8 for changes as small as 15% for most metrics. Two- step cluster analysis was performed in SPSS Statistics 22. Keravis : Delphinidin Inhibits Tumor Growth ------------------------------------------- Statistical analysis Results are expressed as mean ± SEM of n experiments. Studentâs t test for unpaired data was used for statistical analysis, with P < 0.05 being considered significant. Jourdain : L-Lactate protects neurons against excitotoxicity -------------------------------------------------------------- Statistical analysis. All data are presented as means ± SEM. Unpaired Student's t-test or One-way ANOVA followed by Dunnett's post hoc test (vs respective controls) have been used to determine statistical significance (p < 0.05). 2015 session 2 ============== Rougeron : environmental reservoirs ----------------------------------- 119 soil samples were collected from 11 area types representing various types of human activities. A statistical analysis was conducted to characterize the area types by the microorganisms biota divided into the eight groups presented in Table 1. The hierarchical tree obtained by agglomerative hierarchical classification revealed... Statistical analysis Two-factorial correspondence analysis followed by an agglomerative hierarchical classification (Greenacre, 1984; Benzécri, 1992) was conducted to evaluate the association between the 11 area types and microorganisms. [...] A dendrogram was established to determine the number of classes, and specific characteristics of each class were defined. A P value <= 0.05 was considered statistically significant. Leung et al. : Fulminant Community-Acquired [...] Pneumonia ----------------------------------------------------------- This is a retrospective case-control study comparing CAP-AB and HAP-AB patients, which was performed at United Christian Hospital between July 2000 and December 2003. Results: There were 19 cases of CAP-AB and 74 cases of HAP-AB. [...] Statistical Analysis Data were expressed as the mean (SD) or the median (range) as appropriate unless otherwise specified. The 2 test or Fisher exact test was used to compare categoric variables between CAP-AB and HAP-AB groups where appropriate. The Student t test or the Mann-Whitney U test was used to compare continuous variables between the two groups, depending on the nature of the data. Life table analysis was performed to analyze the survival of CAP-AB patients. Potential risk factors for worse prognosis in the CAP-AB group were tested by log-rank test (categoric variables) or Cox regression (continuous variables). A two-tailed p value of 0.05 was considered to be statistically significant. All statistical analyses were performed using statistical software (SPSS, version 11.0; SPSS Inc; Chicago, IL). Berges et al. : tubulin binding peptide targets glioma cells ------------------------------------------------------------ Statistical analysis. Data are presented as mean ± SEM (bars). Cell counting, cellular viability data, and tumor volumes were analyzed by Student's t test using Prism version 3.00 (GraphPad Software, San Diego, CA). Asterisks indicate significant level versus control *P < 0.05; **P < 0.005; ***P < 0.001. Vallet et al. : urbanization filters plant species of small woodlands --------------------------------------------------------------------- [...]Thirty-six woodlands of about 1.5 ha 11 in Rennes, 12 in Nantes and 13 in Angers were surveyed according to these criteria. Data collection We collected three datasets: a floristic matrix named L, an environmental matrix named R and a biological trait matrix Q. Data analysis We computed principal component analysis (PCA) on the environmental matrix R to ordinate woodlands according to adjacent land cover (e.g. Legendre & Legendre 1998). The floristic matrix L that contains the frequency of each species occurring in each of the 36 woodlands was analysed by correspondence analysis (CA). The trait matrix Q was analysed by principal coordinate analysis (PCoA) applied on a distance matrix computed with the mixed-variables coefficient of distance (Pavoine et al. 2009) based on a generalization of Gower's distance. To investigate the relationships between traits and environmental variables, we used a RLQ analysis based on the CA of the species frequency matrix L (Dole´dec et al. 1996; Ribera et al. 2001). [...] To determine which traits were linked with the first axis of the RLQ analysis, we looked for links between the score of a species (standardized by the corresponding eigenvalue) on the first axis of the RLQ analysis and the value of the traits taken by the species. For each quantitative variable, we used Pearson correlation as a statistic to test for the correlation between species scores on axis 1 and species trait states. The correlation was tested by a permutation scheme: the values of the variable were permuted while the order of the species coordinates was kept constant. For nominal variables, we carried out one-way ANOVA with permutation test (the species coordinates were permuted between the attributes of the nominal variables keeping constant the number of species per attribute). For ordinal and circular variables, a Spearman correlation was carried out and its significance was determined by permutation test (the values of the variable were permuted while the order of the species coordinates was kept constant). For a given fuzzy and multichoice variable, we considered... [...] All analyses were processed using the R software (version 2.8.0.; R Foundation for Statistical Computing, Vienna, Austria) and the ade4 package (Chessel et al. 2006) in R. Basseur et al. : non-neutral landscapes [...] local search ---------------------------------------------------------- instances, mea, range. no software indicated. 2015 session 1 ============== Maitre et al : food pickiness ----------------------------- A survey with 559 French people over 65 years old was conducted to collect data on food selectivity, dependency and nutritional status. As some independent variables were categorical (e.g., dependency category, gender) and other continuous (e.g., age, selectivity score), we performed analyses of covariance (ANCOVA) with the General Linear Model (GLM) procedure of STATGRAPHIC (type III sum of squares). Least-squares means (LS-means) were computed for each factor and submitted to multiple comparisons analysis thanks to the Newman-Keuls method. All the results reported here were significant at a level of 0.05 unless otherwise stated. Means (M) are given with associated standard deviation (SD). Min, Kang et al : microbubble polymeric --------------------------------------- Statistical methods: differences between experimental and control groups were determined using one-way ANOVA and deemed statistically significant (indicated by an asterisk (*) in figure) if p < 0.05. Ghattas, Ben Ishak : sélection de variables ------------------------------------------- SVM, GLM, forêts aléatoires. Méthodes plutôt qu'analyses. Nombreux jeux de données "classiques". Matlab et R utilisés. Altay, Emmert-Streib : gene regulatory networks ----------------------------------------------- Significant maximum mutual information network, Mutual information of two random variables X and Y. The data set consists of 524 published E. coli Affymetrix microarrays collected under various conditions... Boxplots of F-scores. Subnetworks of genes graphics. For our numerical simulations we used R, SynTRen and MINET and for the visualization of the networks the igraph package. Nagy, Reduil : Quantification of Anthocyanins[...] ------------------------------------------------- Plot of peak area ratio, plot of standard error, residual plot, calibration. No software described. No statistical test. Sagin, Sahu : Effects of dietary lycopene[...] ---------------------------------------------- The data were initially analyzed by analysis of variance (ANOVA) using General Linear Models procedure of SAS, othogonal constrasts. LSD option was employed to determine contrast. The effects of the experimental diets on response variables were considered to be significant at p<0.05. Chaparro, Badri : Rhizosphere microbiome ---------------------------------------- All statistical analyses were done using SAS (ver. 9.3; SAS Institute). The PROC MIXED function was used to implement a two-way ANOVA analysis with a Tukey post-hoc adjustment to determine pairwise differences between the microbial communities at each plant developmental time point. To ensure that the data followed the assumptions of normality, sequences were log2 transformed. To identify if developmentally dependent root exudation influenced the soil microbial community, Pearson correlation analysis was performed... Comparisons [...] were done using a two-sample t-test. PCoA, PCA, 95 % confidence ellipses... Wiebke, Knofel : Allogenic CD4+CD25high T Cells ----------------------------------------------- Statistical Analysis: Obtained data were analyzed using Student t test or oneway analysis of variance (ANOVA) with Bonferroni post hoc correction, as appropriate. P values less than 0.05 were considered statistically significant. We used GraphPad Prism version 5.0 (GraphPad Software, San Diego, CA). Grau, Wolf : Biology of TAL Effector Target Sites -------------------------------------------------- Statistical model defined by its likelihood, repeat-variable diresidues (RVD), assumptions as a local mixture model, Dirichlet priors on each subset of parameters defined on a common simplex, estimation by Bayesian maximum a posteriori principle, second-ordre quasi Newton method[...], We consider as performance measure the number of predicted targets that... Performance is measured for different rank cutoffs (Top 10, 20, 50, and 100 predictions) on the predictions for each TAL effector. Software TALgetter. No statistical software described. No statistical test. Demotes-Mainard, Huché-Thélier : water restriction -------------------------------------------------- For each type of restriction (water or light), we compared the various treatment groups. Chi-squared tests were used to compare bud burst distributions as a function of bud location on the stem between groups. For all other variables, groups were compared by one-way analysis of variance (ANOVA), or by nonparametric Kruskal-Wallis tests if the data were not normally distributed or the condition of homoscedasticity was not satisfied. All statistical analyses were performed with the SAS software (V. 9.2, SAS Institute), using the glm, npar1way and freq procedures. Verdier, Lalanne : regulatory network-based approach ---------------------------------------------------- To identify probe sets differentially regulated during seed development, coefficient of variation (CV) was calculated as follows: CV = (SD/mean), where SD is the SD of relative expression and mean is the average of the relative expression value for each probe set across all the studied developmental stages. Distance analysis between different developmental stages was calculated using pairwise Pearson correlation using R (version 2.15) and visualized using MultiExperiment Viewer (MeV version 4, part of the microarray software suite; Saeed et al., 2003). Principal component analysis was performed using median centering mode in MeV software, and axes 1 and 2 were selected to visualize results. PageMan software was used to perform overrepresentation analysis of functional classes using a bin-wise Wilcoxon test, and resulting P values were adjusted according to Bonferroni. A color scale was used to show overrepresented and underrepresented functional classes in red and blue, respectively (Usadel et al., 2009). To identify seed-related or seed-specific probe sets, a Z score was calculated using the following formula: Z = (X 2 mean)/SD, where X is the relative expression value, mean is the average of relative expression of tissues available in the MtGEA (i.e. leaf, root, petiole, stem, flower, nodule, and vegetative bud [http://mtgea.noble.org/]), and SD is the SD of relative expression in all tissues available in the MtGEA. Probe sets with a Z score above 2.66 in seeds were considered as seed specific or preferentially expressed in seeds. Palmer, Harris : carpal tunnel syndrome --------------------------------------- From each paper that was finally reviewed, we abstracted a standard set of information, including details of the study populations, exposure contrasts, estimates of effect and confounders considered. Where prevalence estimates were provided but not relative risks, we calculated odds ratios (ORs) with exact 95% confidence intervals (CIs) using STATA 7.0 statistical software. Tran, Benoit : biodegradable, monodispersed microparticles ---------------------------------------------------------- No statistical analysis, no software. 2014 : Session 2/ 2 =================== Wesling / Rapid quantification of plant-powdery mildew interactions by qPCR and conidiospore counts --------------------------------------------------------------------------------------------------- For each sample, spores were counted in eight 1 mm2 fields of a Neubauer-improved haemocytometer (Marienfeld, Lauda-Königshofen, Germany) and results were averaged. Finally, spore counts were normalized to the initial weight of seedlings. Spore counts of indicated genotypes at 6 dpi normalized to seedling fresh weight. Bars represent the mean ± standard deviation of three samples (500 mg of seedlings each) from one experiment counting eight fields/sample. Asterisks indicate statistically significant differences to Col-0 in two-tailed Student's t-test (p <0,05). (Scale bars are drawn with error bars.). Jing / A novel polyethylene glycol mediated lipid nanoemulsion... ----------------------------------------------------------------- Statistical analysis. Means ± standard deviation (SD) were calculated and Student's t test was used for evaluation of statistical significance. Agouni / Endothelial Dysfunction... ----------------------------------- Data Analysis. Data are represented as mean SEM, n represents the number of experiences. Statistical analyses were performed by a one-way analysis of variance and Mann- Whitney U or analysis of variance for repeated measures and subsequent Bonferroni post hoc test. P<0.05 was considered to be statistically significant. Jick / Isotretoin use and risk of depression... ------------------------------------------------- Relative risks comparing the various exposures to the nonexposed period were calculated using multiple logistic regression models in SAS statistical software, version 6.12 (SAS Institute Inc, Cary, NC). Results are presented as point estimates with 95% confidence intervals (CIs). Qixuan / HBV promotes the proliferation of hepatic stellate cells ----------------------------------------------------------------- Statistical analysis of the results was performed by one-way ANOVA, the Newman-Keuls test, the Mann-Whitney test, and the unpaired Student's t-test when appropriate. Differences were considered to be significant at P<0.05. MESR / Usages du numérique --------------------------- Statistiques descriptives, histogrammes d'effectifs, tracés en radar. Clere 2012 / Estrogen receptor alpha[...] to promote angiogenesis ---------------------------------------------------------------- Data are presented as mean ± SEM, n represents the number of experiments repeated at least in triplicate. Statistical analyses were performed by ANOVA followed by a Bonferroni test. p < 0.05 was considered to be tatistically significant. Kano Nagao / Comparison[...] Eastman challenge problem ------------------------------------------------------ MSPC. MPCA. DISSIM. MS-PCA. Multivariate statistical process control. Moving principal component analysis. Multi-scale PCA. Grossman 2013 / CEBPA double-mutated acute myeloid leukaemia ------------------------------------------------------------ Kaplan-Meier plots. Survival analysis was performed in 91 cases with clinical follow-up data available and was restricted to patients who received intensive therapy. Median follow-up time was 238 months. 80/91 (879%) patients were treated with AMLspecific intensive treatment protocols. Segui-Simarro / [...] in meristematic cells of Arabidopsis thaliana ------------------------------------------------------------------- No statistical analysis. 2014 : Session 1/ 2 =================== Garnier / Comparative proteomics Tisochrysis lutea -------------------------------------------------- PCA, ANOVA. Significant over-abundant spots were detected at a 5% significance level (p-value <0.05). Gernigon / Proximal Ischemia ---------------------------- Statistical evaluation was performed using SPSS V 12.0F software (Cary, NC). Values were expressed as the means ± SD. We considered P values <0.05 to be statistically significant. NS is a non-significant statistical difference. Proportions are calculated with 95% confidence intervals. Comparisons between ABI-b and ABI-n participants were performed using the unpaired Student's t tests for numerical variables and Chi² test for binary variables. Concordance between symptoms by history and on the treadmill was studied with the Kappa coefficient. Huche-Thelier / Nitrogen deficiency ----------------------------------- Treatments were compared by one-way or two-way analysis of variance (ANOVA). If the necessary conditions for ANOVA were not fulfilled, we used non-parametric Kruskal-Wallis or Friedman tests (Conover, 1999). We used 2 tests to compare the percentages of buds bursting in each zone. Correlations between visual attributes and morphological characters were assessed by calculating Spearman's correlation coefficient. All statistical analyses were carried out with STATBOX software (v.6.6) or Agricolae package in the R environment (v2.11.0).9. De Quelen / n-3 polyunsaturated fatty acids in the maternal diet ---------------------------------------------------------------- Data are expressed as mean values with the standard error of the mean. The significance of differences was determined using the Student's t test. For time course experiments in Ussing chambers, a two-way ANOVA was employed. Differences were considered significant for P<= 0.05. Santagostini / visual quality of ornamental plants --------------------------------------------------- All statistical analyses were carried out in the R environment (R Development Core Team, 2011), with the stats, graphics andagricolae packages. Analysis of variance was used for variety comparisons. When the conditions for the application of this method were not fulfilled, nonparametric tests (Kruskal-Wallis or Friedman test) were used (Conover, 1999). Baumann / Adaptation of the Daphnia sp . acute toxicity test ------------------------------------------------------------- EC50 values were calculated separately for each test using the statistics software R (Version 2.13.2, http://www.r-project. org) with a logistic regression. Mean EC50 value calculation and graphical plotting were made with GraphPad Prism 5.00 (GraphPad Software, San Diego, CA, USA). Significance tests were calculated with a two-way ANOVA (repeated measurements) using GraphPad Prism. (série 2) Feringa / Ankle-Brachial Index ------------------------------ Continuous data are expressed as meanand SD or median (in- terquartile range), and compared using the t test or Mann- Whitney test when appropriate. Categorical data are pre- sented as percentage frequencies, and differences between proportions were compared using the chi2test with Yates cor- rection.Theprimaryendpointswereoverallmortalityandcar- diac death. We used univariate and multivariate Cox propor- tional hazards models to analyze the effect of clinical characteristicsandABIvaluesonsurvival.Hazardratiosaregiven with 95% confidence intervals (CIs). [...] For all tests,P<.05 (2-sided) was considered significant. All analyses were per- formedusingacommerciallyavailablesoftwareprogram(SPSS- 11.0 statistical software; SPSS Inc, Chicago, Ill). Le Chen / Mutation and Late-Onset Cardiomyopathy ------------------------------------------------ Results are expressed as mean±SEM. Results from multiple groups were compared by analysis of variance (ANOVA) followed by a Student-Neuman-Keuls test for multiple comparisons. The Student t test was used for comparisons involving only 2 groups. The Wilcoxon rank-sum test and Krusakl-Wallis ANOVA were performed when data were not normally distributed. A P<0.05 was considered significant. In PCR array analysis, q value was calculated using Q-Value software (http://genomics.princeton.edu/storeylab/qvalue/). A limitation of the study was the relatively small sample size for some experiments. Goldstein / Multicenter retrospective analysis of metastatic colorectal cancer -----------------------------------------------------------------------------è Agouni / Red Wine Polyphenols ----------------------------- Cybulska / The effect of Ca2+ and cellular structure on apple firmness ---------------------------------------------------------------------- Statistical analysis was performed using Statistica 9.0 (StatSoft, Inc., Tulsa, OK, USA). The mean values and standard deviation (SD) were determined from 12 replicates at each treatment. The calcium treatment effect was investigated with one-way ANOVA followed by the post hoc Tukey's HSD test. To reveal the correlations in detail, Pearson's correlation coefficients were calculated with Statistica and significance was estimated at p\0.05. 2013 ===== Cannavo ------- Modelling of the van Genuchten curves was analysed using root mean squared error (RMSE) and analysis of variance (ANOVA). Mean differences were tested at the 90% and 95% intervals either for respective p-values of 5% and 10% or for the first species error (R software). For other results, standard errors were calculated (between brackets in the tables). Campiglia -------- The data was analyzed for the 2-year period and the year was considered a random effect. A split plot experimental design was adopted for the cover crop characteristics where the years were treated as the main factor and the cover crop as the split factor; a split-split plot experimental design was adopted for the tomato weed characteristics where the years were treated as the main factor, cover crops the split factor and the mulches (inside and outside the mulch) as the split-split factor; a split-split plot experimental design was adopted for the tomato yield characteristics where the years were treated as the main factor, cover crops the split factor and the levels of nitrogen fertilization as the splitsplit factor. The data was processed with the analysis of variance using the ANOVA methods (SAS, 1996). The variances of some data sets were heterogeneous (Bartlett's test, Gomez and Gomez, 1984) and dictated ln or arcsine square root transformation. A value of 1 was added to all observations to allow the use of the ln transformation when the number zero was in the data set. Treatment means were compared using Fisher's protected LSD test at the 0.05 probability level. Standard errors of difference (SED) were calculated for the weed density inside and outside each mulch at 15 and 30 days after transplanting (DAT).Weed community composition at 15 and 30 days after transplanting (DAT) was analyzed by means of a canonical discriminant analysis (SAS, 1996) by taking each weed Diz --- The density of benthic foraminifera as given in Supplementary Table 1 represent the number of live individuals (>63 mucm) per volume of studied sample (50 cm3). In order to give an estimation of the density of individuals per station the averaged abundance in the centire assemblage was calculated. In this work "centire" refers to all live foraminifera (>63 mucm) picked from 0-1 cm to the deepest studied interval. The relative abundances of particular species were calculated from the entire assemblage and, for comparative purposes with other studies elsewhere, the percentage of large benthic foraminifera (>125 mubcm size fraction) was also considered. Percentages were calculated only if the number of foraminifera was higher than 100 individuals. The differences between foraminifera living in the sediment (Sediment Assemblage, SA) and on polychaete tubes (Polychaete Tube Assemblage, PTA) were evaluated by Chi-square analyses (e.g. Jorissen and Wittling, 1999) in each of the five stations where polychaete tubes were examined (stations 4, 13, 18, 22 and 41). For the sediment assemblage, the Entire Assemblage (EA) and the Surface Assemblage (SA, 0-1 cm) were tested independently. The diversity of the live assemblages (entire and >125 mucm) was calculated with the common diversity indices Fisher's alpha, ShannonWiener and the expected number of species in a rarefied sample of n individuals (ES(n)). In addition we present diversity in graphical form using rarefaction curves (PAST software, version 1.31, Hammer et al., 2001). Elachouri --------- Statistics All experiments were repeated at least three times. Statistics were inferred using the Student's t-test or ANOVA test from the GraphPad Prism software (GraphPad) as indicated in the legends of the figures. Wesley ------ For each day, the NIR-red model values had a close relationship (the average coefficient of determination, r2, was above 0.90). Ramirez ------- Statistical treatment of the results. The linear regression of increase in radius against time (in days) was used to obtain the growth rates (mm/day) under each set of treatment conditions. The percentage germination and germ-tube length after 4 and 6 h of incubation at different water potential values were logit (log x[x/(100 2 x)]) and natural log (ln) transformed respectively to homogenize variance previous to analysis of variance (ANOVA). The growth rates, percentage of germination, germ-tube length (mm), mycelial water potential and sugars and sugar alcohol concentrations were evaluated by ANOVA for each experiment to determine effect of water potential, solute, isolate and twoand three-way interactions. When the analysis was statistically significant, the Tukey's multiple-comparison procedure test was used for separation of the means. Statistical significance was determined at the level P , 0.05. Pearson correlation coefficients between endogenous reserves and mycelial water potential also were calculated. All the studies (ANOVA and correlation) were made by using SigmaStat for Windows version 2.03 (SPSS Inc.). Huang ----- Statistical Analysis - The statistical significance of the differences between experimental variables was determined using the Student's t test. The values shown represent the mean S.D. for triplicate experiments. p 0.05 was considered as statistically significant. Tauseef ------- Statistical analysis. Comparisons between experimental groups were made by one-way ANOVA and post-hoc test. Differences in mean values were considered significant at P < 0.05. Schwacke -------- To overcome these difficulties we employ a statistical model that allows information sharing across levels of a given effect and employ Markov Chain Monte Carlo methods to infer parameters from the data and parameter prior distributions. Model Parameter Inference Parameter inference in iQuantitator employs computational methods developed using the Bayesian statistical framework. Applying Bayes theorem, the probability density of the model parameters conditioned on the observed data can be written as the product of the likelihood, the prior densities of the parameters, and integration constant. The joint posterior density of the parameters is obtained by integrating the likelihood over the prior densities. When a closed form solution of this integral is unavailable, computational methods such as Markov Chain Monte Carlo (MCMC) can be used [17]. The Markov chain constructed using these methods yields samples distributed according to the posterior distribution of the parameter vector. Statistical measures of center and spread estimated from these samples can be used to characterize the parameters of interest. Applying these methods to iTRAQ data requires the construction of a statistical model describing the prior distributions for each model parameter and, potentially, hyperprior distributions for distributional parameters of those priors. The statistical model structure used here follows the recommendation of Gelman [18] for the analysis of ANOVA using hierarchical regression. In this approach, all parameters of a given effect are considered as a batch and share a common prior distribution. Myin-Germeys ------------ Statistics To estimate the effect of cognitive impairments on emotional reaction to daily life stress, a multilevel linear regression model (32) was used. Multilevel or hierarchical linear modeling techniques are a variant of the more often used unilevel linear analyses and are ideally suited for the analysis of data from experience sampling, which consist of multiple observations in one person (i.e., at two levels, beep level and subject level) (33). The beta values are the fixed regression coefficients of the predictors in the multilevel model and can be interpreted in a manner identical to that for the estimate in the unilevel linear regression analyses. The data were analyzed with the SAS PROC MIXED module (SAS Technical Report P-229, 1992). Multilevel linear regression analyses were conducted with negative affect and positive affect as the dependent variables. Hinchcliff ---------- Data analysis - To determine whether EIPH was associated with race performance, distance finished behind the winner, race earnings, and finishing position were used as indicators of performance. Examination of summary and descriptive statistics and of graphs of the data was used to determine whether continuous data were normally distributed, and continuous data that were not normally distributed were transformed to yield a normal distribution or were categorized. The modal value of the EIPH severity grades assigned by the 3 observers was used in all analyses. Presence of EIPH was defined as a dichotomous (no vs yes) variable in 2 ways: severity grade of 0 (no) versus severity grade 1 (yes) and severity grade 1 (no) versus severity grade 2 (yes). To control potential confounding, all variables that may have affected or predicted a horse's performance were included as covariates in the analyses. However, a previous study7 has shown that there is considerable colinearity among these variables. Therefore, principal component analysis was used to create orthogonal (uncorrelated) scores for these independent covariates. With distance finished behind the winner and race earnings as dependent variables, potential associations with the occurrence of EIPH (EIPH severity grade; severity grade = 0 vs severity grade 1; and severity grade 1 vs severity grade 2) were examined by means of multivariable ANOVA.d Because race earnings were highly skewed, they were logarithmically transformed, with values of 0 assigned a nominal value of $1. Multivariate logistic regression was used to determine whether occurrence of EIPH was associated with various categorical assessments of finishing position and race earnings (ie, winning [yes vs no], finishing in the first 3 positions [yes vs no], earning any money in the race [yes vs no], and being in the 90th percentile or higher for earnings in the race [yes vs no]). The Bonferroni method for multiple comparisons was used to adjust comparisons of least square means derived from ANOVA models. Odds ratios (ORs) and 95% confidence intervals (CIs) derived from likelihood ratio statistics were calculated from the logistic regression models. Data are given as mean ± SE. For all analyses, values of P < 0.05 were considered significant. Wasko ----- Statistical analysis Inter-reader correlation analysis of the BAL slides was performed on all cell types of every BAL; kappa analysis was used to establish the inter-reader agreement for classification into BAL categories. Comparisons between horse group for breed, gender and clinical signs were made using Chi-squared tests. Fisher's exact test was used when the number of horses in a cell of the! contingency table was <5. Sensitivity, specificity, and positive (PPV) and negative (NPV) predictive value were calculated for each question of the RSQ for each BAL category. Comparison of the scores between the 3 LAI categories was done using a 2 sample mean comparison test. Nonparametric receiver-operating curves (ROC) were plotted and area under the curve (AUC) was calculated comparing BAL Abnormal to BAL MLAI and SLAI to determine the effect of changing cut- off values of the RSQ. The optimal cut-off scores for BAL abnormal, BALMLAI and BAL SLAI were determined by selecting the uppermost left point of the ROC curve; any score above this point was considered positive.AP value <0.05 was considered statistically significant for all results. In addition, the results of the RSQ were analysed using a logistic regression model as follows. All parameters described by the questions were included into one multivariable logistic regression model and a backwards stepwise elimination procedure of variables was performed to determine significant clinical signs or management practice associated with a difference between SLAI and normal horses. Niebroj-Dobosz -------------- 2.3. Statistical analyses The data were presented as means±SD and range. Differences in variable values were assessed using Mann-Whitney U and Wilcoxon matched pair tests. Spearman's correlation coefficient (rho) was used to analyze the relationships between variables. The p<0.05 value being considered statistically significant. Commercially available statistical programs (Statistica 5, StatSoft, Poland) were used in data analysis. ROC analysis was accomplished using the MedCal program ver. 9.4.2.0. 2012 ==== Miyake ------ confirmatory factor analysis structural equation modeling correlation-regression analysis exploratory factor analyses latent variable analysis fit of model (chi2, aic, srmr, cfi, ifi...) SAS software Deville ------- home-developed clustering algorithm amino acid propensity, Z-score hydrophobicity profiles, chi2 homogeneity tests prediction of the linker conformation, Q-scores 10-fold cross validation procedure Benameur -------- data are represented as mean +/- SEM, statistical analyses were performed by Mann-Whitney U-tests (non-parametric) using Prism software. P<0.005 was considered to be statistically significant. Ouedraogo --------- patients were classified into groups according to their best fit to nine pre-defined mathematic models. cross-correlation analysis, intra-test test-retest reproductibility a best-rmax (Pearson's coefficient of correlation) was used to define a good fitting pearson's chi-square test was used to compare the distribution of profiles sample size estimation we estimated the mean with a type I error of 5 % and a power of 90 % (two-sided hypothesis) data management and cross correlation were performed with an Excel spreadsheet. Statistical analysis for t-tests and chi-square tests were carried out with SPSS. Results are expressed in mean +/- SD. For all statistical tests, a two-tailed probability level of P<0.05 was used to indicate statistical significance. Non-significant results are reported N.S. Tran ---- the amount of active protein was determined thanks to a standard curve. the microspheres preserved their shape (62 +/- 9) with a coefficient of variation of 15 %. Forster ------- correct RT and error rates were subjected to within-subjects two-way analyses of variance Sloat ----- Statistical analyses were completed by performing ANOVA followed by Fisher's protected least significant difference (LSD) procedure. A p value of <= 0.05 (two-tail) was considered significant. 2011 ===== 2.7. Statistical analyses The data were expressed as mean±S.D. for four animals per group. Statistical analyses were performed by one-way analysis of variance (ANOVA) followed by Dunnett's post-test using SPSS (student version 7.5, SPSS Inc., Surrey, UK). We used the significant level alpha=0.05. 2.6. Statistical analysis Statistical analyses were performed with Stata/SE 10.1 software (StataCorp LP, College Station, TX). The results were expressed as the mean±standard deviation (SD). Skewness/kurtosis tests for normality showed that the data obtained for CI-ATP per milliunit of CS and CIATP per milligram of protein were normally distributed and for mitochondrial density, CII-ATP per milliunit of CS, and CII-ATP per milligram of protein after natural logarithm transformation. Differences between groups were then evaluated on the basis of Student's ttest or one-way analysis of variance where appropriate. When significant F values were obtained, individual group means were tested for differences according to Bonferroni's a posteriori correction for multiple comparisons. The criterion for significance was P<0.05 for all comparisons. Linear regression analyses of CI-ATP and CII-ATP per milliunit of CS on age indicated a marginal (R2=0.20) but significant (Pb0.005) age effect that was lost after patient 9 of the LHON-- group (86 years of age) was omitted from the database. Therefore, that patient was not taken into further account. To examine if early- and late LHON+ differ in their ability to compensate per cell for the impaired mitochondrial-dependent ATP synthesis, a Kruskall-Wallis equality-of-populations rank test and a two-sample Kolmogorov-Smirnov test for equality-of-distribution functions were performed after assigning the distinction "positive" to values of CI-ATP per milligram of protein that were equal or higher than the lowest control value and else "negative". Statistical analysis The prevalence rates are proportions of affirmative responses of all responses. The subjects were divided into three equal-sized age groups. Differences in pain prevalence between age groups were assessed by the Mann-Whitney U-test and the differences between occupational groups by the chi-square test. To test whether pain is likely to affect multiple body sites in some individuals, the number of subjects expected to have 0, 1, 2, 3, 4, 5, 6, or 7 numbers of sites with pain was calculated by Poisson's distribution. The presence of the pain in different anatomical sites in a subject was assumed independent from the presence of pain in other sites. The distribution parameter used to generate the expected number of subjects was the average number of sites with pain per individual. The observed frequencies were compared with the expected frequencies using the chi-square test. Prevalence ratios (PR) for pain in one anatomical site relative to another were calculated using Cox's proportional hazards regression (Lee and Chia 1993; Lee 1994; Thompson et al. 1998). To reduce the chance of false positive findings, the Bonferroni correction was applied. The adjustment for multiple tests was applied when the significance of deviation of PR from 1.0 was assessed. The significance level was set at 0.001, adjusted for 42 multiple tests. To obtain all combinations of the seven symptoms, or the three groups of symptoms, each of the dichotomous (0, 1) symptoms (or the symptom categories) was multiplied with a unique power of ten, after which the variables were summed. All analyses were performed using the SPSS Version 12.0.1. Data Analysis Data are represented as mean±SEM, n represents the number of experiences. Statistical analyses were performed by a one-way analysis of variance and Mann- Whitney U or analysis of variance for repeated measures and subsequent Bonferroni post hoc test. P<0.05 was considered to be statistically significant. Statistical Analyses Incidences and spot volumes of each protein in control and CF groups, determined by image analysis, were compared by Fisher's exact test and Student t test, respectively. The Student t test was based on twoway analysis of variance, which included terms for disease status, age (adult vs. child), and their interaction. Strong control of the familywise type I error rate for each term (age, disease status, and interaction) was maintained by Holm's adjustment (18), applied to all the protein spot volumes. This procedure is a stochastically dominant modification of the Bonferroni procedure for testing each hypothesis at a p value of 0.05/K, where K is the number of protein spots analyzed. Statistical analyses were performed with S-Plus 6 software (Insightful Corporation, Seattle, WA). Differences between biomarker concentration and disease groups. Data were analyzed with a linear model, applied separately for children and for adults. The linear model for children was a fixed effects linear model, with a term representing disease status. The mixed model was fitted with the residual maximum likelihood algorithm (19) as implemented in the R package nlme (20) (see the online supplement for further detail). Statistics. Data are expressed as means±SE. One-way ANOVA for repeated measures was used to test for differences. The posttest was computed only if the overall P was < 0.05. We relied on the Bonferroni post hoc test. Differences between preflight and in-flight values after 5 mo in space were compared by the nonparametric Wilcoxon matched-pairs test. Linear regression analysis was used if indicated. A value for P<0.05 was considered significant. Statistical analysis. Microparticle levels were expressed as median and range and analyzed using nonparametric Mann-Whitney U tests. Thrombin-generating activity (TGA) data were expressed as mean±SEM, according to the distribution normality, and analyzed by analysis of variance followed by Bonferonni post-test (StatView4.51, SAS Institute, Cary, North Carolina). Differences were significant at p<0.05. Statistical Analysis Data are expressed as the mean±SEM. Differences in clinical and biochemical parameters between lean and obese women were determined using the Wilcoxon unpaired nonparametric test. Spearman coefficients were computed to examine correlations. The cellular experiments were performed at least 3 times. Statistical analysis was performed using Student t test. Comparisons between >2 groups were performed using 1-way ANOVA analysis followed by post hoc test, in which P<0.05 was considered statistically significant.