intensity.diff.plot: Intensity difference plot
In s-andrews/intensitydiff: Intensity Dependent Magntiude of Effect Test

Description Usage Arguments Details Examples

This function takes the output from the intensity_difference function and draws a number of diagnostic plots to help to evaluate the results of the analysis

1	intensity.diff.plot(intensity.diff.data)

intensity.diff.data

The output data frame from the intensity.difference function

The function will split the current plot area into 4 sections and will plot 4 separate graphs:

1. The distribution of difference values from the data. This should ideally form a single coherent distribution with a median of zero.

2. The relationship between average intensity and standard deviation. The entire point of this test is that there should be linkage between these two values, so if you don't see a relationship here then this test is not appropriate to use. Normally you would expect that the standard deviation would fall as the average intensity rises.

3. An MA plot showing the difference between datasets plotted against the average intensity. The colouring of points on the plot reflects whether they were found to be signficant by the statistical test which was run. The colouring is grey (not significant), blue (significant by raw p-value), but not after multiple testing correction, red (significant after multiple testing correction). You should always see the blue points occupying the outer 5 You may or may not see any red points depending on the behaviour of your data.

4. A Z-score MA plot. This is the same as 3 but using the normalised z-score differences instead of the raw differences. You would expect that the width of the cloud of points should be roughly equal across this plot, and there shouldn't be an intensity dependent effect visible across the plot.