Sigmastat

For all experiments, differences among treatments were assessed via a one- way analysis of variance (ANOVA) using SigmaStat within SigmaPlot 11.0; when data sets failed normality tests, Kruskal-Wallis ANOVAs by ranks were performed. The Student-Newman-Keuls method was used as a post hoc, pairwise, multiple comparison procedures. A one-way ANOVA was used for interannual comparisons of nutrient concentrations, nutrient ratios, and meteorological data between years or seasons. Additionally, Spearman’s rank order correlation matrices were used to assess the extent to which Dinophysis densities co-varied with other members of the plankton community during experiments. Lastly, multiple regression analyses were performed to determine the dependence of Dinophysis acuminata abundances and toxicity on all measured environmental variables (i.e. inorganic nutrients, organic nutrients, plankton groups, temperature, salinity, chlorophyll a, etc) using forward stepwise selection (F to enter p < 0.049; F to remove p > 0.050) in SigmaStat within SigmaPlot 11.0. Cell densities were log transformed prior to analysis. The small bloom (Table 1) that occurred in 2009 during which okadaic acid or pectenotoxins were not detected [32 (link)] was not considered in statistical analyses.

All figures were prepared with Microsoft Excel, SigmaPlot, and Adobe Photoshop software. Power analyses were conducted with preliminary data to determine final sample sizes. Data were expressed as mean ± SEM (standard error of mean). Holm–Sidak test and one-way ANOVA were performed data in figures with SigmaStat when necessary. Otherwise, student paired/unpaired t-tests were used between wt and het mice or between pre- and post-groups with Excel or SigmaStat.

All the data obtained from different experiments were evaluated using statistical analysis. An unpaired t-test and a one-way analysis of variance (ANOVA) (the Fischer LSD, Waltham, MA, USA) test were conducted to compare the mean differences using Sigma Stat version 3.5. Comparisons with p < 0.05 were considered significantly different.

Normality assumptions of continuous variables were examined with Kolmogorov–Smirnov test. Normally distributed variables are expressed as mean and standard deviation values, whereas non-normally distributed variables are presented as median and inter-quartile range values (IQR; 25th–75th percentile). Categorical variables were compared among the CAE, CAD, and control groups using the chi-square test for normally distributed variables and Fischer’s exact test for non-normally distributed variables. One-way analysis of variance (ANOVA) was used to test whether these three groups differed with regards to various continuous parameters of interest. A post hoc analysis with Tukey test was performed for identifying individual levels of statistical significance. The non-parametric Kruskal–Wallis test and post hoc analysis with Mann–Whitney U test were used if the homogeneity of variance assumption was violated. SigmaStat (Systat, San Jose, CA, USA) and GraphPad Prism (Dotmatics, Boston, MA, USA) were used for statistical analysis and generation of graphics, respectively.