R/krsa_group_diff.R

In CogDisResLab/KRSA: KRSA: Kinome Random Sampling Analyzer

#' Calculates LFC based on modeled pw data and grouping
#'
#' This function takes in the modeled pw data, groups, peptides, and an option to perform the calculation per chip (byChip)
#'
#' @param data modeled pw data
#' @param groups a vector. format: (case, control)
#' @param peps peptide list
#' @param samples sample names
#' @param byChip T or F, to calculate per chip
#' @param Barcodes (optional) Barcodes vector
#'
#' @return LFC krsa table
#'
#' @family core functions
#'
#' @export
#'
#' @examples
#' TRUE
krsa_group_diff <- function(data, groups, peps, samples = NULL, byChip = T, Barcodes = NULL) {
    data %>%
        {
            if (!is.null(samples)) dplyr::filter(., SampleName %in% samples) else .
        } %>%
        dplyr::filter(Group %in% groups, Peptide %in% peps) %>%
        {
            if (byChip == T) {
                dplyr::group_by(., Barcode, Peptide)
            } else {
                dplyr::group_by(., Peptide, Group)
            }
        } %>%
        {
            if (byChip == T) {
                dplyr::summarise(., LFC = slope[Group == groups[1]] - slope[Group == groups[2]])
            } else {
                dplyr::summarise(., slope = mean(slope)) %>%
                    ungroup(.) %>%
                    dplyr::group_by(., Peptide) %>%
                    dplyr::summarise(., LFC = slope[Group == groups[1]] - slope[Group == groups[2]])
            }
        } %>%
        dplyr::ungroup(.) %>%
        {
            if (byChip == T) {
                dplyr::group_by(., Peptide) %>%
                    dplyr::mutate(., totalMeanLFC = mean(LFC), LFC_SD = sd(LFC)) %>%
                    dplyr::ungroup(.)
            } else {
                .
            }
        }
}