Perform t-tests (or Wilcoxon tests) for EPSC summary plots
Source:R/Process-data.R
perform_t_tests_for_summary_plot.RdThis function enables you to perform a series of paired t-tests (or Wilcoxon tests) comparing the
mean current amplitude within each interval relative to the mean current
amplitude during the baseline. This uses the pairwise_t_test or pairwise_wilcox_test functions from
rstatix, with paired = TRUE and Holm's adjustment for multiple
comparisons (p_adjust_method = "holm") by default. The resulting output table can also
be used to apply significance stars to the plot in
plot_summary_current_data().
Usage
perform_t_tests_for_summary_plot(
data,
test_category,
include_all_treatments = "yes",
list_of_treatments = NULL,
current_type = "eEPSC",
parameter = "amplitude",
baseline_interval = "t0to5",
interval_length = 5,
treatment_colour_theme,
test_type = "pairwise.t.test",
p_adjust_method = "holm",
save_output_as_RDS = "no"
)Arguments
- data
A dataframe containing the summary data generated from
make_summary_EPSC_data(). Ifcurrent_typeis "eEPSC", this must be the$summary_dataelement of the list produced bymake_summary_EPSC_data().- test_category
A numeric value describing the experimental category. In the sample dataset for this package, 2 represents experiments where insulin was applied continuously after a 5-minute baseline period.
- include_all_treatments
A character (
"yes"or"no") specifying if the plot will include data from all treatments. If"no", you must specify a list of treatments inlist_of_treatments.- list_of_treatments
A list of character values describing the treatments that will be in the plot. Defaults to
NULL, since include_all_treatments is"yes"by default.- current_type
A character describing the current type. Allowed values are
"eEPSC"or"sEPSC".- parameter
A character value specifying the parameter to be plotted on the y-axis. For evoked currents (
current_type = "eEPSC"), the available parameter is "amplitude", which contains amplitudes normalized relative to the baseline. For spontaneous currents (current_type = "sEPSC"), the available parameters are "amplitude" (normalized currents), "raw_amplitude", "frequency" (normalized frequency) or "raw_frequency".- 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 inmake_normalized_EPSC_data().- interval_length
Length of each interval (numeric; in minutes). Used to divide the dataset into broad ranges for statistical analysis. Important!
max_recording_lengthmust be evenly divisible byinterval_length. Defaults to5.- 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.
- test_type
A character (must be
"pairwise.wilcox.test"or"pairwise.t.test") describing the statistical model used in this function.- p_adjust_method
This argument is directly related to
p.adjust.methodinrstatix::t_test. This is the method used to adjust the p-value in multiple pairwise comparisons. Allowed values include"holm","hochberg","hommel","bonferroni","BH","BY","fdr","none"(although"none"is not recommended).- save_output_as_RDS
A character (
"yes"or"no") describing if the resulting object should be saved as an RDS file in the raw data folder.
References
Nutter B (2018). lazyWeave: LaTeX Wrappers for R #' Users. R package version 3.0.2, https://CRAN.R-project.org/package=lazyWeave.
See also
make_normalized_EPSC_data() for an example of how the normalized
current amplitudes were created.
Examples
perform_t_tests_for_summary_plot(
data = sample_summary_eEPSC_df$summary_data,
test_category = 2,
include_all_treatments = "yes",
list_of_treatments = NULL,
current_type = "eEPSC",
parameter = "amplitude",
baseline_interval = "t0to5",
test_type = "pairwise.t.test",
interval_length = 5,
treatment_colour_theme = sample_treatment_names_and_colours,
save_output_as_RDS = "no"
)
#> group2 asterisk_time treatment .y. group1 n1 n2
#> 1 t10to15 12.5 Control mean_P1_transformed t0to5 4 4
#> 2 t15to20 17.5 Control mean_P1_transformed t0to5 4 4
#> 3 t20to25 22.5 Control mean_P1_transformed t0to5 4 4
#> 4 t5to10 7.5 Control mean_P1_transformed t0to5 4 4
#> 5 t10to15 12.5 HNMPA mean_P1_transformed t0to5 5 5
#> 6 t15to20 17.5 HNMPA mean_P1_transformed t0to5 5 5
#> 7 t20to25 22.5 HNMPA mean_P1_transformed t0to5 5 5
#> 8 t5to10 7.5 HNMPA mean_P1_transformed t0to5 5 5
#> 9 t10to15 12.5 PPP mean_P1_transformed t0to5 5 5
#> 10 t15to20 17.5 PPP mean_P1_transformed t0to5 5 5
#> 11 t20to25 22.5 PPP mean_P1_transformed t0to5 5 5
#> 12 t5to10 7.5 PPP mean_P1_transformed t0to5 5 5
#> 13 t10to15 12.5 PPP_and_HNMPA mean_P1_transformed t0to5 5 5
#> 14 t15to20 17.5 PPP_and_HNMPA mean_P1_transformed t0to5 5 5
#> 15 t20to25 22.5 PPP_and_HNMPA mean_P1_transformed t0to5 5 5
#> 16 t5to10 7.5 PPP_and_HNMPA mean_P1_transformed t0to5 5 5
#> statistic df p p.adj p.adj.signif p_string significance_stars
#> 1 4.89 3 0.016 0.033 * 0.033 *
#> 2 6.96 3 0.006 0.018 * 0.018 *
#> 3 8.01 3 0.004 0.016 * 0.016 *
#> 4 2.28 3 0.107 0.107 ns 0.11
#> 5 3.15 4 0.035 0.133 ns 0.13
#> 6 2.99 4 0.040 0.133 ns 0.13
#> 7 3.19 4 0.033 0.133 ns 0.13
#> 8 3.10 4 0.036 0.133 ns 0.13
#> 9 0.89 4 0.422 0.844 ns 0.84
#> 10 1.50 4 0.209 0.627 ns 0.63
#> 11 1.96 4 0.121 0.484 ns 0.48
#> 12 0.78 4 0.481 0.844 ns 0.84
#> 13 0.94 4 0.402 1.000 ns > 0.99
#> 14 0.84 4 0.447 1.000 ns > 0.99
#> 15 1.39 4 0.238 0.952 ns 0.95
#> 16 -0.59 4 0.586 1.000 ns > 0.99