Skip to contents

Plot Event Value (Average)

Source: R/plot_event_avg.R
plot_event_avg.Rd

A highly abstracted plot that can be used to plot an average value within and across experiments. This is useful for examining features of synaptic events, e.g. amplitude, failure rate, potency, paired-pulse ratio. The plot requires pre-processing of the feature(s) of interest and a column with unique values to group by, i.e. experiment recording IDs.

Usage

plot_event_avg(
  df,
  x_axis,
  y_axis,
  plot_grouping,
  plot_mean = FALSE,
  weighted_mean = FALSE,
  sd_value = NULL,
  experiment_name = NULL,
  x_label = NULL,
  y_label = NULL,
  ymax = NA,
  ...
)

Arguments

df

Data frame or Tibble, grouped experiment data set.

x_axis

Character - unquoted, non-standard evaluation. Column name to use for x-axis.

y_axis

Character - unquoted, non-standard evaluation. Column name to use for y-axis.

plot_grouping

Character, column name to use for grouping.

plot_mean

Boolean, add group mean to plot. Default is FALSE.

weighted_mean

Boolean, plot weighted mean, requires df column labelled 'w' of weights \(w\). Default is FALSE.

sd_value

Optional character, std deviation measurement to use, accepts "sd" or "sem". Default is NULL.

experiment_name

Optional character, add a subtitle with experiment name. Default is NULL.

x_label

Optional character, custom label for x-axis. Default NULL, will use the x_axis column name.

y_label

Optional character, custom label for y-axis. Default NULL, will use the y_axis column name.

ymax

Optional integer, set a y-axis limit. Default is NULL and the y-axis will be automatically determined by ggplot.

...

Optional vargs for additional ggplot function calls, e.g. ylim(), ylab().

Value

A ggplot2 object with user defined features.

Details

The weight column, w, is assumed to frequency weights - where a weight equals the number of occurrences. Weighted mean and sd are calculated using the following formulas:

  • weighted mean: \(\bar{x}=\frac{\sum\limits_{i=1}^n w_i x_i}{\sum\limits_{i=1}^n w_i}\)

  • weighted sd: \(\sqrt{\hat{\sigma}^2_\mathrm{w}}\)

  • unbiased variance: \(\hat{\sigma}^2_\mathrm{w}=\frac{\sum\limits_{i=1}^N w_i \left(x_i - \mu^{*}\right)^2 }{\sum_{i=1}^N w_i - 1}\)

References

See also

Examples

simple_tbl <-
  data.frame(
    recording_id = rep(1:2, each = 3),
    experiment_id = rep("example", 6),
    condition = rep(c("control", "drug_1", "drug_2"), 2),
    amplitude = c(1, 0.5, 1, 2, 1, 2)
  )

plot_event_avg(
  df = simple_tbl,
  x_axis = condition,
  y_axis = amplitude,
  plot_grouping = "recording_id",
  plot_mean = TRUE,
  sd_value = "sem",
  experiment_name = "example",
  ymax = 3
) + ggplot2::coord_fixed(ratio = 3)


# remove average and add custom labels
plot_event_avg(
  df = simple_tbl,
  x_axis = condition,
  y_axis = amplitude,
  plot_grouping = "recording_id",
  plot_mean = FALSE,
  experiment_name = "example",
  x_label = "custom x-label",
  y_label = "custom y-label",
  ymax = 3
)


# plot a weighted average
weighted_tbl <-
  data.frame(
    recording_id = rep(1:2, each = 3),
    experiment_id = rep("example", 6),
    condition = rep(c("control", "drug_1", "drug_2"), 2),
    amplitude = c(1, 0.5, 1, 2, 1, 2),
    w = rep(c(1, 100), each = 3)
  )

plot_event_avg(
  df = weighted_tbl,
  x_axis = condition,
  y_axis = amplitude,
  plot_grouping = "recording_id",
  plot_mean = TRUE,
  weighted_mean = TRUE,
  sd_value = "sd",
  experiment_name = "plot weighted mean",
  x_label = "condition",
  y_label = "amplitude",
  ymax = 3,
)