R/fct_prep_project_data_for_app.R

In teofiln/gene.editing.dash: VOR Gene Editing Dashboard

#' prep_project_data
#'
#' @description prepare an RDS file for loading into the data.
#'
#' Requirements:
#'   - RDS
#'   - A list of 6 tables corresponding to:
#'      1. Quality control table (QCTable)
#'      2. Editing frequency table (EditingFreq)
#'      3. Indel frequency table (IndelFreq)
#'      4. CRISPR summary table (CRISPRSummary)
#'      5. Sample metadata table (SampleMeta)
#'      6. Allele frequency and Sequence alignment table (AlleleAlignFreq)
#'   - The order matters
#'   - Currently, no checks are made for table composition and column types,
#'     so be sure to the table conform to original examples
#'
#' Procedure:
#'   1. Load the RDS list of tables
#'   2. Rename the tables
#'   3. Format the column names into BigCamel
#'   4. Save to a serialized, fast, QC binary file for faster reading in-app
#'
#' @param project_rds_path character, path to the project data RDS file
#' @param output_qs_path character, path for the output serialized qs file
#'
#' @importFrom qs qsave
#' @importFrom janitor clean_names
#'
#' @export
prep_project_data <- function(project_rds_path, output_qs_path) {

  table_names <- c(
    "QCTable",
    "EditingFreq",
    "IndelFreq",
    "CRISPRSummary",
    "SampleMeta",
    "AlleleAlignFreq"
  )

  # 1. load the rds
  project_data <- readRDS(project_rds_path)

  # 2. rename tables
  names(project_data) <- table_names

  # 3. rename columns
  project_data <- lapply(project_data, janitor::clean_names, case = "big_camel")

  # 4. write to file
  qs::qsave(x = project_data, file = output_qs_path, preset = "balanced")
}