R/extract_node_data.R

In mselensky/bngal: Biological Network Graphing and Learning

#' Import network data
#'
#' @param network.data Output from [`bngal::export_network_data()`]
#' @param tax.level Taxonomic level of network data output from `[bngal::export_network_data]`
#'
#' @return
#' @export
#'
#' @examples
extract_node_data <- function(network.data, tax.level) {

  tax.levels <- c("phylum", "class", "order", "family", "genus", "asv")

  if (!is.null(nrow(network.data$nodes[[tax.level]]))) {
    # list nodes and edges each under "all" for downstream syntax
    network.data.list=list()
      network.data.list$nodes[[tax.level]][["all"]] = network.data$nodes[[tax.level]]
      network.data.list$edges[[tax.level]][["all"]] = network.data$edges[[tax.level]]
  } else {
    network.data.list=network.data
  }
  rm(network.data)

  # loop to add "tax_level" and "subcomm" to each node df
  for (x in names(network.data.list$nodes[[tax.level]])) {
    network.data.list$nodes[[tax.level]][[x]] <- network.data.list$nodes[[tax.level]][[x]] %>%
      dplyr::mutate(tax_level = tax.level,
                    sub_comm = x)

  }

  network.nodes = list()
  network.nodes[[tax.level]] <- Reduce(bind_rows, network.data.list$nodes[[tax.level]])

  network.nodes[[tax.level]] <- network.nodes[[tax.level]] %>%
    distinct(label, edge_btwn_cluster, sub_comm, .keep_all = T) %>%
    dplyr::rename(degree=value,
                  taxon_=label)


  Reduce(bind_rows, network.nodes)


  # tax.levels <- c("phylum", "class", "order", "family", "genus", "asv")
  #
  # if (!is.null(nrow(network.data$nodes$asv))) {
  #   # list nodes and edges each under "all" for downstream syntax
  #   network.data.list=list()
  #   for (i in tax.levels) {
  #     network.data.list$nodes[[i]][["all"]] = network.data$nodes[[i]]
  #     network.data.list$edges[[i]][["all"]] = network.data$edges[[i]]
  #   }
  # } else {
  #   network.data.list=network.data
  # }
  # rm(network.data)
  #
  # # a horrid double for loop to add "tax_level" to each node dataframe for each subnetwork
  #
  # network.nodes = list()
  # for (i in tax.levels) {
  #   for (x in names(network.data.list$nodes[[i]])) {
  #     network.data.list$nodes[[i]][[x]] <- network.data.list$nodes[[i]][[x]] %>%
  #       dplyr::mutate(tax_level = i,
  #                     sub_comm = x)
  #
  #   }
  #
  #   network.nodes[[i]] <- Reduce(bind_rows, network.data.list$nodes[[i]])
  #
  #   network.nodes[[i]] <- network.nodes[[i]] %>%
  #     distinct(label, edge_btwn_cluster, sub_comm, .keep_all = T) %>%
  #     dplyr::rename(degree=value,
  #                   taxon_=label)
  #
  # }
  #
  # Reduce(bind_rows, network.nodes)

}