ANOVA is to partition the variation The Sources of variation plot presents the relative contribution of each factor included in the ANOVA model towards explaining the variability of the data for a feature analyzed by the ANOVA. 

ANOVA partitions the variation for a feature among all the observations into different components. One component is due to the variation among the different groups of the factors in the model, which represents signal; another component is due to the variation within groups, which represents noise (it . The variance within groups is also called error, because it is the variability that is not explained by any factors ).   F-statistic is a ratio of the two components, that is signal vs noise ratio

F = Between group variance / within group variance

Large A large F value of for a factor indicates that between group variation is large relative to the greater than within group variation, which results in a small p-value . On the other hand, for that factor. Conversely, a small F value of for a factor indicates the between group variation is small relative to the smaller than within group variation, which results in a big p-value for that factor.

In Partek^® Flow^®,   source the Sources of variation plot is using uses a bar chart to show the F ratio of value for each factor in the ANOVA model relative to the error. It The plot is available on ANOVA reportfor each gene in an ANOVA result. To open the Sources of variation plot for a gene in the ANOVA results table, click on the icon    next to the gene symbol to invoke the plot.  It is useful to exam the variation examine the plot to understand the contribution of all the factors in the ANOVA model to the overall variance in the data set relative to the their within group variation (an example plot is on Figure 1).

The chart title is based on the feature ID (e.g. gene or transcript) that the plot was invoked on. The yY-axis is the F value, X-axis represent categories are the factors specified in ANOVA model as well as and the Error term which . The Error term will always have the an F value as of 1 (since error because error, i.e., within group variance, is the denominator , so in the F value would be 1)calculation.

Plot controls are in the panel on the left. The <Previous and Next> buttons enable you to switch between genomic features as they appear in the parent Feature listANOVA results table.  The bar chart colors are controlled by the Customize colors option. If you want to change a color, select customize the colors, click the  icon to choose configure the color palette. 

...