santaR_auto_summary: Summarise, report and save the results of a santaR analysis

Description

After multiple variables have been analysed using santaR_auto_fit, santaR_auto_summary helps identify significant results and summarise them in an interpretable fashion. Correction for multiple testing can be applied to generate Bonferroni [1], Benjamini-Hochberg [2] or Benjamini-Yekutieli [3] corrected p-values. P-values can be saved to disk in .csv files. For a given significance cut-off (plotCutOff), the number of variables significantly altered is reported and plots are automatically saved to disk by increasing p-value. The aspect of the plots can be altered such as the representation of confidence bands (showConfBand) or the generation of a mean curve across all samples (showTotalMeanCurve) to help assess difference between groups when group sizes are unbalanced.

Usage

santaR_auto_summary(
  SANTAObjList,
  targetFolder = NA,
  summaryCSV = TRUE,
  CSVName = "summary",
  savePlot = TRUE,
  plotCutOff = 0.05,
  showTotalMeanCurve = TRUE,
  showConfBand = TRUE,
  legend = TRUE,
  fdrBH = TRUE,
  fdrBY = FALSE,
  fdrBonf = FALSE,
  CIpval = TRUE,
  plotAll = FALSE
)

Value

A list: result$pval.all

data.frame of p-values, with all variables as rows and different p-value corrections as columns. result$pval.summary

data.frame of number of variables with a p-value inferior to a cut-off. Different metric and p-value correction as rows, different cut-off (Inf 0.05, Inf 0.01, Inf 0.001) as columns.

Arguments

SANTAObjList: A list of SANTAObj with p-values calculated, as generated by santaR_auto_fit.
targetFolder: (NA or str) NA or the path to a folder in which to save summary.xls and plots. If NA no outputs are saved to disk. If targetFolder does not exist, folders will be created. Default is NA.
summaryCSV: If TRUE save the (corrected if applicable) p-values to 'CSVName'_summary.csv, 'CSVName'_pvalue-all.csv, 'CSVName'_pvalue-dist.csv, 'CSVName'_pvalue-dist.csv (default summary_summary.csv,...). Default is TRUE.
CSVName: (string) Filename of the csv to save. Default is 'summary'.
savePlot: If TRUE save to targetFolder all variables with p < plotCutOff ordered by p-values. Default is TRUE.
plotCutOff: (float) P-value cut-off value to save summary plots to disk. Default 0.05.
showTotalMeanCurve: If TRUE add the mean curve across all groups on the plots. Default is TRUE.
showConfBand: If TRUE plot the confidence band for each group. Default is TRUE.
legend: If TRUE add a legend to the plots. Default is TRUE.
fdrBH: If TRUE add the Benjamini-Hochberg corrected p-value to the output. Default is TRUE.
fdrBY: If TRUE add the Benjamini-Yekutieli corrected p-value to the output. Default is FALSE.
fdrBonf: If TRUE add the Bonferroni corrected p-value to the output. Default is FALSE.
CIpval: If TRUE add the upper and lower confidence interval on p-value to the output. Default is TRUE.
plotAll: If TRUE override the plotCutOff parameter and plot all variables. Default is FALSE.

References

[1] Bland, J. M. & Altman, D. G. Multiple significance tests: the Bonferroni method. British Medial Journal 310, 170 (1995).

[2] Benjamini, Y. & Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society 57, 1, 289-300 (1995).

[3] Benjamini, Y. & Yekutieli, D. The control of the false discovery rate in multiple testing under depencency. The Annals of Statistics 29, 1165-1188 (2001).

Examples

Run this code

## 2 variables, 56 measurements, 8 subjects, 7 unique time-points
## Default parameter values decreased to ensure an execution < 2 seconds
inputData     <- acuteInflammation$data[,1:2]
ind           <- acuteInflammation$meta$ind
time          <- acuteInflammation$meta$time
group         <- acuteInflammation$meta$group
SANTAObjList  <- santaR_auto_fit(inputData, ind, time, group, df=5, ncores=0, CBand=TRUE,
                                pval.dist=TRUE, nBoot=100, nPerm=100)
# Input data generated: 0.02 secs
# Spline fitted: 0.03 secs
# ConfBands done: 0.53 secs
# p-val dist done: 0.79 secs
# total time: 1.37 secs
result <- santaR_auto_summary(SANTAObjList)
print(result)
# $pval.all
#              dist dist_upper  dist_lower     curveCorr    dist_BH
# var_1 0.03960396 0.09783202 0.015439223 -0.2429725352 0.03960396
# var_2 0.00990099 0.05432519 0.001737742  0.0006572238 0.01980198
#
# $pval.summary
#       Test Inf 0.05 Inf 0.01 Inf 0.001
# 1    dist        2        1         0
# 2 dist_BH        2        0         0