R/load_data.R

In dragosmg/rocheshiny: A Packaged Shiny App for the Roche Genentech Coding Test

#' Load Laboratory Measurements Data
#'
#' This function performs the following operations:
#'     * loads the data
#'     * casts the variables to their correct types
#'     * transforms the variable names to snake case
#'     * performs a light touch cleaning by transforming character variables to
#'     title case
#'
#' @return a data frame (tibble) containing laboratory measurements data
#' @export
#'
#' @examples
#' \dontrun{
#' lab_data <- load_lab_data()
#' }
load_lab_data <- function() {

    folder <- system.file(package = "rocheshiny", "extdata")

    lab_file <- file.path(folder, "Random_LabValuesInfo_2020.tsv")

    cols_to_title_case_lab <- c("bmrkr2", "lbcat", "avisit")

    lab_data <- readr::read_tsv(lab_file,
                                col_types = readr::cols(
                                    STUDYID = readr::col_character(),
                                    USUBJID = readr::col_character(),
                                    BMRKR1 = readr::col_double(),
                                    BMRKR2 = readr::col_character(),
                                    LBTESTCD = readr::col_character(),
                                    LBTEST = readr::col_character(),
                                    LBCAT = readr::col_character(),
                                    AVAL = readr::col_double(),
                                    AVALU = readr::col_character(),
                                    AVISIT = readr::col_character()
                                )) %>%
        janitor::clean_names() %>%
        dplyr::mutate_at(.vars = cols_to_title_case_lab,
                         .funs = stringr::str_to_title) %>%
        dplyr::group_by(.data$usubjid, .data$lbtestcd) %>%
        dplyr::mutate(order = dplyr::row_number()) %>%
        dplyr::ungroup() %>%
        dplyr::mutate(visit = stringr::str_remove_all(.data$avisit,
                                                      pattern = "Week [1-5] "))

    lab_data
}


#' Load Patient Data
#'
#' Similar to [load_lab_data()] this function performs the following operations:
#'     * loads the data
#'     * casts the variables to their correct types
#'     * transforms the variable names to snake case
#'     * performs a light touch cleaning by transforming character variables to
#'     title case
#'
#'
#' @return a data frame (tibble) containing patient data
#' @export
#'
#' @examples
#'\dontrun{
#'patient_data <- load_patient_data()
#'}
load_patient_data <- function() {
    folder <- system.file(package = "rocheshiny", "extdata")

    patient_file <- file.path(folder, "Random_PatientLevelInfo_2020.tsv")

    cols_to_title_case_pat <- c("sex", "race")

    patient_data <- readr::read_tsv(patient_file,
                                    col_types = readr::cols(
                                        STUDYID = readr::col_character(),
                                        USUBJID = readr::col_character(),
                                        AGE = readr::col_double(),
                                        SEX = readr::col_character(),
                                        RACE = readr::col_character(),
                                        ACTARM = readr::col_character(),
                                        ACTARMCD = readr::col_character()
                                    )) %>%
        janitor::clean_names() %>%
        dplyr::mutate_at(.vars = cols_to_title_case_pat,
                         .funs = stringr::str_to_title) %>%
        dplyr::mutate(race = stringr::str_replace_all(.data$race, "Or", "or"))

    patient_data
}