R/ge_winners.R

In metan: Multi Environment Trials Analysis

#' Genotype-environment winners
#' @description
#' `r badge('stable')`
#'
#' Computes the ranking for genotypes within environments and return the winners.
#'
#' @param .data The dataset containing the columns related to Environments,
#'   Genotypes, and the response variable(s).
#' @param env The name of the column that contains the levels of the
#'   environments.
#' @param gen The name of the column that contains the levels of the genotypes.
#' @param resp The response variable(s). To analyze multiple variables in a
#'   single procedure a vector of variables may be used. For example `resp
#'   = c(var1, var2, var3)`. Select helpers are also allowed.
#' @param type The type of results. Defaults to `"winners"` (default),
#'   i.e., a two-way table with the winner genotype in each environment. If
#'   `type = "ranks"` return the genotype ranking within each environment.
#' @param better A vector of the same length of the number of variables to rank
#'   the genotypes according to the response variable. Each element of the
#'   vector must be one of the `'h'` or `'l'`. If `'h'` is used
#'   (default), the genotypes are ranked from maximum to minimum. If `'l'`
#'   is used then the are ranked from minimum to maximum. Use a comma-separated
#'   vector of names. For example, `better = c("h, h, h, h, l")`, for
#'   ranking the fifth variable from minimum to maximum.
#' @return A tibble with two-way table with the winner genotype in each
#'   environment (default) or the genotype ranking for each environment (if
#'   `type = "ranks"`).
#' @author Tiago Olivoto \email{tiagoolivoto@@gmail.com}
#' @export
#' @examples
#'\donttest{
#' library(metan)
#' ge_winners(data_ge, ENV, GEN, resp = everything())
#'
#' # Assuming that for 'GY' lower values are better.
#' ge_winners(data_ge, ENV, GEN,
#'            resp = everything(),
#'            better = c("l, h"))
#'
#' # Show the genotype ranking for each environment
#' ge_winners(data_ge, ENV, GEN,
#'            resp = everything(),
#'            type = "ranks")
#' }
#'
ge_winners <- function(.data, env, gen, resp, type = "winners", better = NULL) {
  if(!type  %in% c("ranks", "winners")){
    stop("The argument 'type' must be one of the 'ranks' or 'winners'")
  }
  factors  <-
    .data %>%
    select({{env}}, {{gen}}) %>%
    mutate(across(everything(), as.factor))
  vars <- .data %>% select({{resp}}, -names(factors))
  vars %<>% select_numeric_cols()
  factors %<>% set_names("ENV", "GEN")
  listres <- list()
  nvar <- ncol(vars)
  if(!missing(better)){
    better <- unlist(strsplit(better, split = ", ")) %>% all_lower_case()
  } else {
    better <- rep("h", nvar)
  }
  if (length(better) != ncol(vars)){
    stop("The vector 'better' should have length " , nvar, " (the number of variables in 'resp')", call. = FALSE)
  }
  if (any(!better %in% c("h", "l"))){
    stop("Invalid values in argument 'better'. It must have 'h' or 'l' only.", call. = FALSE)
  }
  for (var in 1:nvar) {
    temp <- factors %>%
      mutate(Y = vars[[var]])
    if(has_na(temp)){
      temp <- remove_rows_na(temp)
      has_text_in_num(temp)
    }
    if (length(better) == 1) {
      if (better == "h") {
        temp <-
          temp %>%
          mean_by(ENV, GEN) %>%
          group_by(ENV) %>%
          arrange(desc(Y), .by_group = TRUE)
      }
      if (better == "l") {
        temp <-
          temp %>%
          mean_by(ENV, GEN) %>%
          group_by(ENV) %>%
          arrange(Y, .by_group = TRUE)
      }
    } else {
      if (better[[var]] == "h") {
        temp <-
          temp %>%
          mean_by(ENV, GEN) %>%
          group_by(ENV) %>%
          arrange(desc(Y), .by_group = TRUE)
      }
      if (better[[var]] == "l") {
        temp <-
          temp %>%
          mean_by(ENV, GEN) %>%
          group_by(ENV) %>%
          arrange(Y, .by_group = TRUE)
      }
    }
    listres[[paste(names(vars[var]))]] <- temp
  }
  if (type == "ranks"){
    bind <- do.call(
      cbind,
      lapply(listres, function(x) {
        as.character(x[["GEN"]])
      })) %>%
      as_tibble() %>%
      mutate(ENV = listres[[1]][["ENV"]]) %>%
      select(ENV, everything()) %>%
      ungroup()
  } else{
    bind <- do.call(
      cbind,
      lapply(listres, function(x) {
        as.character(x[["GEN"]])
      })) %>%
      as_tibble() %>%
      mutate(ENV = listres[[1]][["ENV"]]) %>%
      select(ENV, everything()) %>%
      group_by(ENV) %>%
      select_rows(1) %>%
      ungroup()
  }
  return(bind)
}