Descriptive statistics
Open the Statistics tab among the analyses tabs in the top of the main window, and press the Descriptive statistics button to load the analysis into the Control panel. There must be defined groups in the data set to be able to load the analysis. If the data has been transposed, make sure that there are groups of genes. When the data have no grouping, you can create a group collecting all samples, or genes if the data is transposed. The groups are defined in the Data manager under the Groups tab.
In the Control panel you select the genes and groups of samples you are interested in by ticking the check boxes. If the data is transposed, you select samples and groups of genes. Use the radio buttons to decide whether the error bars in the plot should reflect the standard deviation (SD), standard error of the mean (SEM), or a confidence interval (CI) of specified confidence level. Standard deviation (STDEV) quantifies scatter - how much the values vary. The estimated standard deviation does not change predictably as you acquire more data. It might go up and it might go down. On average, the standard deviation will stay the same as sample size gets larger. Standard error of mean (SEM) quantifies how accurately the true mean of the population is estimated. The standard error of mean decreases as the sample gets larger. If the scatter is caused by biological variability, you probably want to show the variation. In this case, select the STDEV for the error bars. If you have technical repeats you may instead want to show how well you have assessed the mean. Then use the SEM or a 95% CI for the error bars. There is also a Plot only check box that is ticked by default. When it is ticked the result is only presented as a bar plot with selected error bars. If it is unticked, results are also represented as one table for each selected group containing more information than the plot (see below).
The bar plot shows the mean values for each gene and group as a bar, with an error bar of the selected type. The color of the bars correspond to the colors defined for the groups in the Data manager under the Colors & Symbols tab.
If the Plot only check box was unticked, one table for each selected group is displayed. The top part of the table includes the data values sorted by gene and group, or by sample and group if the data is transposed. KS is the Kolomogorov-Smirnov test statistic that is used to test whether the data in that column is normally distributed. A high KS p-value indicates that the samples are indeed normally distributed, while a low p-value indicate that they are not. This is summarized in the row Norm. dist. (Normally distributed) which is green and states TRUE if Kolomogorov-Smirnov's test indicates that the data is normally distributed, or is red saying FALSE if not. Some statistical tests assume that the data is normally distributed, so it is important to have an indication of whether this is true for your data or not. Kolomogorov-Smirnov's test gives such an indication. If the test comes out as TRUE, parametric tests can be used such as t-test and ANOVA, but if it comes out FALSE, non-parametric tests should be used.
There are also rows in the result table that contain the Count, Average, standard deviation (STDEV), and standard error from the mean (SEM) for each column. If CI was selected as error bar, the critical t-statistic (t*), the confidence interval CI, as well as the Higher and Lower part on the interval are given.