Jmp version 10

The primary outcomes in this study were lung cancer-specific mortality and risk of death. Survival time was calculated as the time from primary surgery to death due to lung cancer, censoring at the date of last contact or non-cancer death. Chi-square or Fisher’s exact test was used to examine distributions of variables. Differences in survival were examined using log-rank test. Cox proportional hazard model was used to estimate the relative risk for death and 95% confidence intervals (CI). Statistical analyses were performed using a statistical software JMP version 10.0.2 (SAS Institute Inc., Cary, NC, USA), otherwise specified. Extended Fisher's exact test was conducted using “R”: a language and environment for statistical computing (R Core Team, 2013. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/) with a package “coin”. Hardy-Weinberg equilibrium was examined to compare the observed and expected genotype frequencies using a Chi-square test. Statistical significance of reporter experiments and gene expression analyses were analyzed using statistical softwares JMP version 10.0.2 and StatView version 5.0 software (SAS Institute Inc.).

All data were analyzed using JMP Version 10 (SAS Institute) using its embedded standard general linear model algorithms [23 ]. All data were analyzed using JMP Version 10 (SAS Institute) using its embedded standard general linear model algorithms [23 ]. T tests (T test) were used for comparisons between two groups with respect to continuous variables. One-way analysis of variance (ANOVA) was used for comparisons of groups with continuous variables and estimate the amount of variance explained by the predictor. Bivariate regression (bivariate) was used for analyses of the relationship between two continuous variables. Cotinine and carbon monoxide (CO) levels, depending on context, could be treated as either a categorical or a continuous variable. Cotinine levels of 2 ng/μl or greater were classified as categorical positives. Carbon monoxide levels of 10 parts per million (ppm) were treated as categorical positives. Self-report of substance use by subject or parent was treated as a categorical variable. Methylation and PHQ-9 scores were used as continuous variables.

For microbiota data, the statistical difference between treatments was examined using the Monte-Carlo test in package ade4 [27 ,28 ] of R 2.14.0 [29 ] as described by de Carcer et al [30 (link)]. The Monte Carlo test is a non-parametric test based on random permutations. The statistical differences of the between treatments was evaluated with the function of ade4::randtest.between. The values of zero were replaced with the detection limit, which is determined by the ratio of one to the lowest read number in the data set. The Benjamini-Hochberg procedure was applied to control the false discovery rate. The dominant genera that increased or decreased in abundance were identified by correspondence analysis in package ade4 of R 2.14.0 as described by de Carcer et al [30 (link)]. Statistical difference for relative gene expression was assessed with the Wilcoxon rank sum test (Mann–Whitney test) using JMP version 10 (SAS Institute Inc., Cary, NC) and was presented as mean ± SEM. Statistical difference was determined at a P value of 0.05 or less. Bacterial genera detected in cecum content of mice administered saline (control) or L. casei strains and fold change in the expression of the targeted genes in the ileum of mice administered L. casei were used to generate a dendrograms by the Ward method of hierarchical clustering (JMP version 10, SAS Institute Inc., Cary, NC).