R/calc_perturbation.R

In ceRNAnetsim: Regulation Simulator of Interaction between miRNA and Competing RNAs (ceRNA)

#' Calculates average expression changes of all nodes except
#' trigger and finds the perturbed node count for a given node.
#'
#' @importFrom tibble tibble
#'
#' @details calc_perturbation calculates mean expression changes
#' of elements except trigger after the change in the network
#' in terms of percentage. It also calculates the number of nodes
#' that have expression changes after the change occur in the network.
#' The function determines the perturbation efficiency and number
#' of perturbed nodes after given change with how, cycle and
#' limit parameter.
#'
#' @param input_graph the graph object that was processed with
#' priming graph in previous step.
#' @param node_name The node that is trigger for simulation.
#' @param how The change of count of the given node in terms of fold change.
#' @param cycle The iteration of simulation.
#' @param limit The minimum fold change which can be taken into
#' account for perturbation calculation on all nodes in terms of percentage.
#'
#' @return a tibble with two columns, the perturbation
#' efficiency and number of perturbed nodes.
#'
#' @examples
#'
#' data('minsamp')
#'
#' minsamp%>%
#'    priming_graph(competing_count = Competing_expression,
#'        miRNA_count = miRNA_expression)%>%
#'    calc_perturbation('Gene6', how= 3, cycle = 4)
#'
#'  minsamp%>%
#'    priming_graph(competing_count = Competing_expression, miRNA_count = miRNA_expression,
#'        aff_factor = c(energy,seed_type), deg_factor = region)%>%
#'    calc_perturbation('Gene6',3, cycle = 4)
#'
#'
#' @export


calc_perturbation <- function(input_graph, node_name, how = 1, cycle = 1,
                              limit = 0) {

  res <- input_graph %>% update_how(node_name, how) %>% simulate(cycle) %>%
    tidygraph::as_tibble() %>% dplyr::filter(name != node_name)

  perturbation_eff <- as.double((res %>% summarise(mean(abs(count_current -
                                                              initial_count) * 100/initial_count)))[[1]])

  perturbed_count <- as.double((res %>% filter((abs(count_current - initial_count) *
                                                  100/initial_count) > limit) %>% count())[[1]])

  return(tibble(perturbation_efficiency = perturbation_eff, perturbed_count = perturbed_count))
}