R/cv_by_sample_estimation.R

In Keniajin/protGear: Protein Micro Array Data Management and Interactive Visualization

#' Summarise CV by samples
#' @title cv by sample
#'
#' @param cv_variable A character string containing the identifier 
#' of the variable with CV values.
#' @param lab_replicates A numeric value indicating the number of lab 
#' replicates.
#' @param  sampleID_var A character string containing the name of the 
#' sample identifier variable. Default set to 'sampleID'
#' @param dataCV A dataframe
#' @import dplyr tidyr
#' @importFrom tidyr gather
#' @importFrom dplyr lag filter
#' @description A function to give the summary of 
#' the CV's by the sampleID
#' @return A data frame of CV calculated by sample
#' @export
#'
#' @examples
#' dataC <- readr::read_csv(system.file("extdata", 
#' "dataC.csv", package="protGear"))
#' ## this file has 3 lab replicates and the default names
#' dataCV <- cv_estimation(dataC  ,lab_replicates=3)
#' cv_by_sample_estimation(dataCV, cv_variable = "cvCat_all",
#'  lab_replicates = 3)
cv_by_sample_estimation <-
  function(dataCV,
           cv_variable,
           lab_replicates,
           sampleID_var = 'sampleID') {
    ## creating a summary of the CV's by sampleID for each file
    ## helps in identifying samples with a high CV value
    if (lab_replicates > 1) {
      iden <- unique(dataCV$iden)
      dataC_cvSample <-  dataCV %>%
        group_by_at(c(sampleID_var, cv_variable)) %>%
        summarise(n = n()) %>%
        mutate(perc = round((n / sum(n)) * 100, 2)) %>%
        #rename(cvCat='get(cv_variable)') %>%
        ungroup() %>%
        gather(variable, value,-c(sampleID_var, cv_variable)) %>%
        unite(temp,!!(cv_variable), variable) %>%
        tidyr::spread(temp, value, fill = 0)
    } else{
      dataC_cvSample <- NULL
      warning("The experiment is specified not to have lab replicates")
    }

    return(dataC_cvSample)
  }