Plot variance comparison for a treatment

plot_variance_comparison_data() creates a connected plot with time as a categorical variable (i.e. baseline/before and after) on the x-axis and the variance measure on the y-axis. There are also lines connecting the "before" data point to the "after" data point for each letter.

Usage

plot_variance_comparison_data(
  data,
  plot_category,
  plot_treatment,
  facet_by_sex = "no",
  variance_measure = "cv",
  baseline_interval = "t0to5",
  baseline_label = "Baseline",
  post_hormone_interval = "t20to25",
  post_hormone_label = "Insulin",
  included_sexes = "both",
  male_label = "Male",
  female_label = "Female",
  left_sex = "Female",
  test_type,
  y_variable_signif_brackets = NULL,
  map_signif_level_values = F,
  geom_signif_family = "",
  geom_signif_text_size = 5,
  large_axis_text = "no",
  geom_signif_size = 0.4,
  mean_line_thickness = 1.2,
  mean_point_size = 2.5,
  treatment_colour_theme,
  theme_options,
  save_plot_png = "no",
  filename_suffix = "",
  ggplot_theme = patchclampplotteR_theme()
)

Arguments

data: A dataframe generated using make_variance_data().
plot_category: A numeric value specifying the category, which can be used to differentiate different protocol types. In the sample dataset for this package, plot_category == 2 represents experiments where insulin was applied continuously after a 5-minute baseline period.
plot_treatment: A character value specifying the treatment you would like to plot (e.g. "Control"). plot_treatment represents antagonists that were present on the brain slice, or the animals were fasted, etc.
facet_by_sex: A character value ("yes" or "no") describing if the plots should be faceted by sex. This is only available if included_sexes is "both". The resulting plot will be split in two, with male data on the left and female data on the right.
variance_measure: A character value ("cv" or "VMR"). The variance measures can be either the inverse coefficient of variation squared (variance_measure == "cv") or variance-to-mean ratio (variance_measure == "VMR").
baseline_interval: A character value indicating the name of the interval used as the baseline. Defaults to "t0to5", but can be changed. Make sure that this matches the baseline interval that you set in make_normalized_EPSC_data().
baseline_label: A character value for the x-axis label applied to the pre-hormone state. Defaults to "Baseline".
post_hormone_interval: A character value specifying the interval used for the data points after a hormone or protocol was applied. This must match the post_hormone_interval used in make_variance_data().
post_hormone_label: A character value for x-axis label applied to the post-hormone or post-protocol state. Defaults to "Post-hormone" but you will likely change this to the hormone or protocol name.
included_sexes: A character value ("both", "male" or "female"). Useful if you want to have a plot with data from one sex only. Defaults to "both". If you choose a single sex, the resulting plot will have "-males-only" or "-females-only" in the file name. WARNING!! If you choose "male" or "female", you MUST ensure that the t_test_df contains data that has been filtered to only include one sex. Otherwise, the significance stars will represent both sexes and it will be inaccurate.
male_label: A character value used to describe how males are encoded in the sex column of the dataframe used in data. This MUST match the value for male data in the sex column, and it must be consistent across data sheets. Defaults to "Male".
female_label: A character value used to describe how females are encoded in the sex column of the dataframe used in data. This MUST match the value for female data in the sex column, and it must be consistent across data sheets. This must be consistent in all data sheets. Defaults to "Female".
left_sex: A character value ("Female" or "Male") describing the sex that will appear on the left side of a faceted plot. Only applies if facet_by_sex is "yes".
test_type: A character (must be "wilcox.test", "t.test" or "none") describing the statistical model used to create a significance bracket comparing the pre- and post-hormone groups.
y_variable_signif_brackets: A character value. You should only use this if your data did not pass assumptions and you had to transform it. y_variable_signif_brackets should be the name of the column of data which has the transformed data (e.g. log-transformed data). Raw data will be plotted, but the significance brackets (and t-test/wilcox test) will use the transformed data. If you did not transform the data, this will default to the raw data column.
map_signif_level_values: A TRUE/FALSE value or a list of character values for mapping p-values. If TRUE, p-values will be mapped with asterisks (e.g. \* for p < 0.05, for p < 0.01). If FALSE, raw p-values will display. You can also insert a list of custom mappings or a function. For example, use map_signif_level_values = function(p) if (p < 0.1) {round(p, 3)} else {"ns"} to only display the p-values when they are below 0.1.
geom_signif_family: A character value describing the font family used for the p-value annotations used by ggsignif::geom_signif(). Defaults to "" (empty value, will be replaced with default system font), but can be replaced with a named font. Use a package like extrafont to load system fonts into R.
geom_signif_text_size: A numeric value describing the size of the text annotations (significance stars or p-values) on the plot. Defaults to 8.
large_axis_text: A character ("yes" or "no"). If "yes", a ggplot theme layer will be applied which increases the axis text.
geom_signif_size: A numeric value describing the size of the geom_signif bracket size. Defaults to 0.4, which is a good thickness for most applications.
mean_line_thickness: A numeric value describing the thickness of the line used to indicate the mean for a group. Defaults to 1.2.
mean_point_size: A numeric value describing the size of the points used to indicate the means. Defaults to 2.5.
treatment_colour_theme: A dataframe containing treatment names and their associated colours as hex values. See sample_treatment_names_and_colours for an example of what this dataframe should look like.
theme_options: A dataframe containing theme options. See sample_theme_options for an example of what this dataframe should look like.
save_plot_png: A character ("yes" or "no"). If "yes", the plot will be saved as a .png using ggsave. The filepath depends on the current type, but they will all go in subfolders below Figures/ in your project directory.
filename_suffix: Optional character value to add a suffix to the filename of the .png file created with this plot. Could be useful if you want to specify anything about the data (for example, to distinguish between recordings produced in MiniAnalysis vs. Clampfit).
ggplot_theme: The name of a ggplot theme or your custom theme. This will be added as a layer to a ggplot object. The default is patchclampplotteR_theme(), but other valid entries include theme_bw(), theme_classic() or the name of a custom ggplot theme stored as an object.

Value

A ggplot object. If save_plot_png == "yes", it will also generate a .png file exported to Figures/Evoked-currents/Variance-plots. The plot will be named in the form of "Variance-comparison-category-plot_category-plot_treatment-variance_measure.png". An example filename is "Variance-comparison-category-2-Control-cv.png".

Details

If you specify a test_type, the function will perform a paired t-test or paired wilcox test and add brackets with significance stars through ggsignif::geom_signif().

This allows you to visually determine if a change in synaptic plasticity is due to a pre- or post-synaptic mechanism. For more information, please see Huijstee & Kessels (2020).

Examples


# Without facets

plot_variance_comparison_data(
  data = sample_eEPSC_variance_df,
  plot_category = 2,
  plot_treatment = "Control",
  variance_measure = "cv",
  baseline_interval = "t0to5",
  post_hormone_interval = "t20to25",
  included_sexes = "both",
  facet_by_sex = "no",
  post_hormone_label = "Insulin",
  test_type = "wilcox.test",
  large_axis_text = "no",
  treatment_colour_theme = sample_treatment_names_and_colours,
  theme_options = sample_theme_options
)


# With facets

plot_variance_comparison_data(
  data = sample_eEPSC_variance_df,
  plot_category = 2,
  plot_treatment = "Control",
  variance_measure = "cv",
  baseline_interval = "t0to5",
  post_hormone_interval = "t20to25",
  included_sexes = "both",
  facet_by_sex = "yes",
  post_hormone_label = "Insulin",
  test_type = "wilcox.test",
  large_axis_text = "no",
  treatment_colour_theme = sample_treatment_names_and_colours,
  theme_options = sample_theme_options
)

Usage

Arguments

Value

Details

See also

Examples