R/find_common_hit.R

In EmanuelSoda/ScreenR: Package to Perform High Throughput Biological Screening

#' @title Find common hit
#' @description This method find the hit in common between the three methods
#' @param hit_zscore The matrix obtained by the \code{\link{find_zscore_hit}}
#'                   method
#' @param hit_camera The matrix obtained by the \code{\link{find_camera_hit}}
#'                   method
#' @param hit_roast The matrix obtained by the \code{\link{find_roast_hit}}
#'                  method
#' @param common_in Number of methods in which the hit has to be in common
#'                  in order to be considered a candidate hit.
#'                  The default value is 3, which means that has to be present
#'                  in the result of all the three methods.
#' @return A vector containing the common hit
#' @importFrom tibble tibble
#' @export
#' @concept find
#' @examples
#' hit_zscore <- data.frame(Gene = c("A", "B", "C", "D", "E"))
#' hit_camera <- data.frame(Gene = c("A", "B", "C", "F", "H", "G"))
#' hit_roast <- data.frame(Gene = c("A", "L", "N"))
#'
#' # common among all the three methods
#' find_common_hit(hit_zscore, hit_camera, hit_roast)
#'
#' # common among at least two of the three methods
#' find_common_hit(hit_zscore, hit_camera, hit_roast, common_in = 2)
find_common_hit <- function(hit_zscore, hit_camera, hit_roast, common_in = 3) {
    # First we gets all the Gene name of the different table
    hit_zscore_Gene <- as.vector(hit_zscore$Gene)
    hit_camera_Gene <- as.vector(hit_camera$Gene)
    hit_roast_Gene <- as.vector(hit_roast$Gene)

    name <- unique(c(hit_zscore_Gene, hit_camera_Gene, hit_roast_Gene))

    hit_common <-
        tibble::tibble(Gene = name) %>%
        dplyr::mutate(hit_zscore = ifelse(test = .data$Gene %in%
            hit_zscore_Gene, yes = 1, no = 0)) %>%
        dplyr::mutate(hit_camera = ifelse(test = .data$Gene %in%
            hit_camera_Gene, yes = 1, no = 0)) %>%
        dplyr::mutate(hit_roast = ifelse(test = .data$Gene %in%
            hit_roast_Gene, yes = 1, no = 0)) %>%
        dplyr::rowwise() %>%
        dplyr::mutate(common = sum(
            .data$hit_zscore, .data$hit_camera,
            .data$hit_roast
        )) %>%
        dplyr::filter(.data$common >= common_in) %>%
        dplyr::pull(.data$Gene)

    return(hit_common)
}

#' @title Plot common hit
#' @description This method plot the hits in common among the three methods is
#'              a wrapper for the \code{\link[ggvenn]{ggvenn}} function.
#' @param hit_zscore The list of hits of the \code{\link{find_zscore_hit}}
#' @param hit_camera The list of hits of the \code{\link{find_camera_hit}}
#' @param hit_roast The list of hits of the \code{\link{find_roast_hit}}
#' @param alpha A value for the opacity of the plot.
#'              Allowed values are in the range 0 to 1
#' @param stroke_size Stroke size for drawing circles
#' @param set_name_size Text size for set names
#' @param text_size Text size for intersect contents
#' @param text_color Text color for intersect contents
#' @param show_percentage Show percentage for each set
#' @param show_elements Show set elements instead of count/percentage.
#' @param title The title to display above the plot
#' @return A vector containing the common hit
#' @param color The three vector color for the venn
#' @importFrom ggplot2 ggtitle
#' @export
#' @examples
#' hit_zscore <- data.frame(Gene = c("A", "B", "C", "D", "E"))
#' hit_camera <- data.frame(Gene = c("A", "B", "C", "F", "H", "G"))
#' hit_roast <- data.frame(Gene = c("A", "L", "N"))
#' plot_common_hit(hit_zscore, hit_camera, hit_roast)
#'
plot_common_hit <- function(hit_zscore, hit_camera,
    hit_roast, alpha = 0.5, stroke_size = 0.5, set_name_size = 4,
    text_color = "black", text_size = 4, show_percentage = TRUE,
    title = "", color = c("#1B9E77", "#D95F02", "#7570B3"),
    show_elements = TRUE) {
    hit_list <- list(
        `Z-score Hits` = unique(hit_zscore$Gene),
        `Camera Hits` = unique(hit_camera$Gene),
        `ROAST Hits` = unique(hit_roast$Gene)
    )

    ggvenn::ggvenn(
        data = hit_list, fill_alpha = alpha,
        columns = c("Z-score Hits", "Camera Hits", "ROAST Hits"),
        fill_color = color, stroke_size = stroke_size,
        text_size = text_size, text_color = text_color,
        show_elements = show_elements, set_name_size = set_name_size,
        show_percentage = show_percentage
    ) +
        ggplot2::ggtitle(title)
}