geposan/R/ranking.R

91 lines
2.8 KiB
R
Raw Normal View History

2021-10-19 13:39:55 +02:00
#' Rank the results by computing a score.
#'
#' This function takes the result of [analyze()] and creates a score by
2021-10-19 13:39:55 +02:00
#' computing a weighted mean across the different methods' results.
#'
#' @param analysis Analysis object resulting from [analyze()].
#' @param weights Named list pairing method names with weighting factors. Only
#' methods that are contained within this list will be included.
2021-10-19 13:39:55 +02:00
#'
#' @returns A ranking object. The object extends the analysis with additional
#' columns containing the `score` and the `rank` of each gene. It will be
#' ordered by rank.
2021-10-19 13:39:55 +02:00
#'
#' @export
ranking <- function(analysis, weights) {
if (!"geposan_analysis" %chin% class(analysis)) {
stop("Invalid analyis. Use geposan::analyze().")
}
ranking <- copy(analysis)
ranking[, score := 0.0]
2021-10-19 13:39:55 +02:00
for (method in names(weights)) {
weighted <- weights[[method]] * ranking[, ..method]
ranking[, score := score + weighted]
2021-10-19 13:39:55 +02:00
}
# Normalize scores to be between 0.0 and 1.0.
ranking[, score := score / sum(unlist(weights))]
setorder(ranking, -score)
ranking[, rank := .I]
2021-10-19 13:39:55 +02:00
structure(
ranking,
class = c("geposan_ranking", "geposan_analysis", class(ranking))
)
2021-10-19 13:39:55 +02:00
}
#' Find the best weights to rank the results.
#'
#' This function finds the optimal parameters to [ranking()] that result in the
#' reference genes ranking particulary high.
#'
#' @param analysis Results from [analyze()] or [ranking()].
2021-10-19 13:39:55 +02:00
#' @param methods Methods to include in the score.
#' @param reference_gene_ids IDs of the reference genes.
2021-10-21 11:42:44 +02:00
#' @param target The optimization target. It may be one of "mean", "min" or
#' "max" and results in the respective rank being optimized.
2021-10-19 13:39:55 +02:00
#'
#' @returns Named list pairing method names with their optimal weights. This
#' can be used as an argument to [ranking()].
2021-10-19 13:39:55 +02:00
#'
#' @export
optimal_weights <- function(analysis, methods, reference_gene_ids,
target = "mean") {
if (!"geposan_analysis" %chin% class(analysis)) {
stop("Invalid analyis. Use geposan::analyze().")
}
2021-10-19 13:39:55 +02:00
# Create the named list from the factors vector.
weights <- function(factors) {
result <- NULL
mapply(function(method, factor) {
result[[method]] <<- factor
}, methods, factors)
result
}
2021-10-21 11:42:44 +02:00
# Compute the target rank of the reference genes when applying the weights.
target_rank <- function(factors) {
data <- ranking(analysis, weights(factors))
2021-10-21 11:42:44 +02:00
data[gene %chin% reference_gene_ids, if (target == "min") {
min(rank)
} else if (target == "max") {
max(rank)
} else {
mean(rank)
}]
2021-10-19 13:39:55 +02:00
}
2021-10-21 11:42:44 +02:00
factors <- stats::optim(rep(1.0, length(methods)), target_rank)$par
2021-10-21 16:21:55 +02:00
factors[factors < 0.0] <- 0.0
2021-10-19 13:39:55 +02:00
total_weight <- sum(factors)
weights(factors / total_weight)
}