R/allele-count-mism.R

In hlaR: Tools for HLA Data

#' @name CountAlleleMism
#' @title Count HLA mismatch at the allele level
#' @description Donor and recipient HLA(Human Leukocyte Antigen) typing data is compared to determine allele level mismatch. The output of EvalAlleleMism is used as input for this function. Allele level mismatch can be calculated for both high and low resolution data. The generated count will return NA if the input alleles are NA.
#' @param dat_in
#' A data frame with donor and recipient mismatched alleles. It's a output from EvalAlleleMism function.
#' @param names_in
#' A vector of HLA loci name to count mismatch for.
#' @return
#' A tibble of input data (subject id and hla loci) followed by mismatch hla count of each subject.
#' @export
#'
#' @import
#' tidyverse
#'
#' @examples
#' hla <- read.csv(system.file("extdata/example", "HLA_MisMatch_count_test.csv", package = "hlaR"))
#' classI <- CountAlleleMism(hla, c("mism.a1", "mism.a2", "mism.b1", "mism.b2"))
#' classII <- CountAlleleMism(hla, c("mism.dqa12", "mism.dqb11", "mism.dqb12"))

CountAlleleMism <- function(dat_in, names_in){
  names <- syms(names_in)
  len <- length(names) + 1
  dat_out <- dat_in %>%
              select(1, !!!names) %>%
              mutate(num_nas = apply(is.na(.), 1, sum)) %>%
              mutate(mism_total = ifelse(num_nas == length(names), NA, rowSums(.[2:len], na.rm = T))) %>%
              select(-num_nas)

  return(dat_out)
}