context("calculateLb")

################################################################################
# CHANGE LOG
# 22.03.2019: Changed deprecated 'matches' to 'expect_match'.
# 22.03.2019: Added tests for new option "peak".
# 29.08.2016: Added  kit = "SGMplus" for performance.
# 01.03.2016: Upgraded to work with version 1.6.0.9002.
# 26.12.2015: First version.
#
# require(strvalidator)
# require(testthat)
# test_dir("inst/tests/")
# test_file("tests/testthat/test-calculateLb.r")
# test_dir("tests/testthat")

test_that("calculateLb", {
  # Get test data.
  data(set1)
  data(ref1)
  data(set2)
  data(ref2)

  # Extract sample PC2 and fix datasets.
  set1Fat <- set1[set1$Sample.Name == "PC2", ]
  set1 <- slim(data = set1Fat, fix = c("Sample.Name", "Marker", "Dye"), stack = c("Allele", "Height"))
  ref1 <- slim(data = ref1, fix = c("Sample.Name", "Marker"), stack = c("Allele"))

  # TEST 01 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for data including
  # one locus dropout.

  # Analyse dataframe.
  res <- calculateLb(
    data = set2, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$TPPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(862 / 3005))
  expect_that(res$Lb[2], equals(506 / 3005))
  expect_that(res$Lb[3], equals(914 / 3005))
  expect_that(res$Lb[4], equals(723 / 3005))
  expect_that(res$Lb[5], equals(1219 / 3363))
  expect_that(res$Lb[6], equals(619 / 3363))
  expect_that(res$Lb[7], equals(766 / 3363))
  expect_that(res$Lb[8], equals(759 / 3363))
  expect_that(res$Lb[9], equals(592 / 1760))
  expect_that(res$Lb[10], equals(743 / 1760))
  expect_that(res$Lb[11], equals(425 / 1760))

  expect_that(res$Lb[12], equals(440 / 1530))
  expect_that(res$Lb[13], equals(304 / 1530))
  expect_that(res$Lb[14], equals(439 / 1530))
  expect_that(res$Lb[15], equals(347 / 1530))
  expect_that(res$Lb[16], equals(638 / 1750))
  expect_that(res$Lb[17], equals(309 / 1750))
  expect_that(res$Lb[18], equals(402 / 1750))
  expect_that(res$Lb[19], equals(401 / 1750))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(3005, 3363, 1760, 1530, 1750, NA)))


  # TEST 02 -------------------------------------------------------------------
  # Test that global proportional Lb can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "prop", by.dye = FALSE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$TPPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(862 / 8128))
  expect_that(res$Lb[2], equals(506 / 8128))
  expect_that(res$Lb[3], equals(914 / 8128))
  expect_that(res$Lb[4], equals(723 / 8128))
  expect_that(res$Lb[5], equals(1219 / 8128))
  expect_that(res$Lb[6], equals(619 / 8128))
  expect_that(res$Lb[7], equals(766 / 8128))
  expect_that(res$Lb[8], equals(759 / 8128))
  expect_that(res$Lb[9], equals(592 / 8128))
  expect_that(res$Lb[10], equals(743 / 8128))
  expect_that(res$Lb[11], equals(425 / 8128))

  expect_that(res$Lb[12], equals(as.numeric(NA)))
  expect_that(res$Lb[13], equals(as.numeric(NA)))
  expect_that(res$Lb[14], equals(as.numeric(NA)))
  expect_that(res$Lb[15], equals(as.numeric(NA)))
  expect_that(res$Lb[16], equals(as.numeric(NA)))
  expect_that(res$Lb[17], equals(as.numeric(NA)))
  expect_that(res$Lb[18], equals(as.numeric(NA)))
  expect_that(res$Lb[19], equals(as.numeric(NA)))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(8128, NA)))


  # TEST 03 -------------------------------------------------------------------
  # Test that normalized Lb by dye can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "norm", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$MTPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$MTPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(862 / 914))
  expect_that(res$Lb[2], equals(506 / 914))
  expect_that(res$Lb[3], equals(914 / 914))
  expect_that(res$Lb[4], equals(723 / 914))
  expect_that(res$Lb[5], equals(1219 / 1219))
  expect_that(res$Lb[6], equals(619 / 1219))
  expect_that(res$Lb[7], equals(766 / 1219))
  expect_that(res$Lb[8], equals(759 / 1219))
  expect_that(res$Lb[9], equals(592 / 743))
  expect_that(res$Lb[10], equals(743 / 743))
  expect_that(res$Lb[11], equals(425 / 743))

  expect_that(res$Lb[12], equals(440 / 440))
  expect_that(res$Lb[13], equals(304 / 440))
  expect_that(res$Lb[14], equals(439 / 440))
  expect_that(res$Lb[15], equals(347 / 440))
  expect_that(res$Lb[16], equals(638 / 638))
  expect_that(res$Lb[17], equals(309 / 638))
  expect_that(res$Lb[18], equals(402 / 638))
  expect_that(res$Lb[19], equals(401 / 638))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Maximum total peak height.
  expect_that(unique(res$MTPH), equals(c(914, 1219, 743, 440, 638, NA)))


  # TEST 04 -------------------------------------------------------------------
  # Test that global normalized Lb can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "norm", by.dye = FALSE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$MTPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$MTPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(862 / 1219))
  expect_that(res$Lb[2], equals(506 / 1219))
  expect_that(res$Lb[3], equals(914 / 1219))
  expect_that(res$Lb[4], equals(723 / 1219))
  expect_that(res$Lb[5], equals(1219 / 1219))
  expect_that(res$Lb[6], equals(619 / 1219))
  expect_that(res$Lb[7], equals(766 / 1219))
  expect_that(res$Lb[8], equals(759 / 1219))
  expect_that(res$Lb[9], equals(592 / 1219))
  expect_that(res$Lb[10], equals(743 / 1219))
  expect_that(res$Lb[11], equals(425 / 1219))

  expect_that(res$Lb[12], equals(as.numeric(NA)))
  expect_that(res$Lb[13], equals(as.numeric(NA)))
  expect_that(res$Lb[14], equals(as.numeric(NA)))
  expect_that(res$Lb[15], equals(as.numeric(NA)))
  expect_that(res$Lb[16], equals(as.numeric(NA)))
  expect_that(res$Lb[17], equals(as.numeric(NA)))
  expect_that(res$Lb[18], equals(as.numeric(NA)))
  expect_that(res$Lb[19], equals(as.numeric(NA)))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Maximum total peak height.
  expect_that(unique(res$MTPH), equals(c(1219, NA)))


  # TEST 05 -------------------------------------------------------------------
  # Test that centred quantities Lb by dye can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "cent", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$MPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$MPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals((862 - 751.25) / sqrt(751.25)))
  expect_that(res$Lb[2], equals((506 - 751.25) / sqrt(751.25)))
  expect_that(res$Lb[3], equals((914 - 751.25) / sqrt(751.25)))
  expect_that(res$Lb[4], equals((723 - 751.25) / sqrt(751.25)))
  expect_that(res$Lb[5], equals((1219 - 840.75) / sqrt(840.75)))
  expect_that(res$Lb[6], equals((619 - 840.75) / sqrt(840.75)))
  expect_that(res$Lb[7], equals((766 - 840.75) / sqrt(840.75)))
  expect_that(res$Lb[8], equals((759 - 840.75) / sqrt(840.75)))
  expect_that(res$Lb[9], equals((592 - (586 + 2 / 3)) / sqrt(586 + 2 / 3)))
  expect_that(res$Lb[10], equals((743 - (586 + 2 / 3)) / sqrt(586 + 2 / 3)))
  expect_that(res$Lb[11], equals((425 - (586 + 2 / 3)) / sqrt(586 + 2 / 3)))

  expect_that(res$Lb[12], equals((440 - 382.5) / sqrt(382.5)))
  expect_that(res$Lb[13], equals((304 - 382.5) / sqrt(382.5)))
  expect_that(res$Lb[14], equals((439 - 382.5) / sqrt(382.5)))
  expect_that(res$Lb[15], equals((347 - 382.5) / sqrt(382.5)))
  expect_that(res$Lb[16], equals((638 - 437.5) / sqrt(437.5)))
  expect_that(res$Lb[17], equals((309 - 437.5) / sqrt(437.5)))
  expect_that(res$Lb[18], equals((402 - 437.5) / sqrt(437.5)))
  expect_that(res$Lb[19], equals((401 - 437.5) / sqrt(437.5)))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Mean peak height.
  expect_that(unique(res$MPH), equals(c(751.25, 840.75, 586 + 2 / 3, 382.5, 437.5, NA)))


  # TEST 06 -------------------------------------------------------------------
  # Test that global centred quantity Lb can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "cent", by.dye = FALSE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$MPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_true(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$MPH)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(round(res$Lb[1], 4), equals(round((862 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[2], 4), equals(round((506 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[3], 4), equals(round((914 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[4], 4), equals(round((723 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[5], 4), equals(round((1219 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[6], 4), equals(round((619 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[7], 4), equals(round((766 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[8], 4), equals(round((759 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[9], 4), equals(round((592 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[10], 4), equals(round((743 - 738.9091) / sqrt(738.9091), 4)))
  expect_that(round(res$Lb[11], 4), equals(round((425 - 738.9091) / sqrt(738.9091), 4)))

  expect_that(res$Lb[12], equals(as.numeric(NA)))
  expect_that(res$Lb[13], equals(as.numeric(NA)))
  expect_that(res$Lb[14], equals(as.numeric(NA)))
  expect_that(res$Lb[15], equals(as.numeric(NA)))
  expect_that(res$Lb[16], equals(as.numeric(NA)))
  expect_that(res$Lb[17], equals(as.numeric(NA)))
  expect_that(res$Lb[18], equals(as.numeric(NA)))
  expect_that(res$Lb[19], equals(as.numeric(NA)))
  expect_that(res$Lb[20], equals(as.numeric(NA)))
  expect_that(res$Lb[21], equals(as.numeric(NA)))
  expect_that(res$Lb[22], equals(as.numeric(NA)))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(862))
  expect_that(res$TPH[2], equals(506))
  expect_that(res$TPH[3], equals(914))
  expect_that(res$TPH[4], equals(723))
  expect_that(res$TPH[5], equals(1219))
  expect_that(res$TPH[6], equals(619))
  expect_that(res$TPH[7], equals(766))
  expect_that(res$TPH[8], equals(759))
  expect_that(res$TPH[9], equals(592))
  expect_that(res$TPH[10], equals(743))
  expect_that(res$TPH[11], equals(425))

  expect_that(res$TPH[12], equals(440))
  expect_that(res$TPH[13], equals(304))
  expect_that(res$TPH[14], equals(439))
  expect_that(res$TPH[15], equals(347))
  expect_that(res$TPH[16], equals(638))
  expect_that(res$TPH[17], equals(309))
  expect_that(res$TPH[18], equals(402))
  expect_that(res$TPH[19], equals(401))
  expect_that(res$TPH[20], equals(284))
  expect_that(res$TPH[21], equals(362))
  expect_that(res$TPH[22], equals(as.numeric(NA)))

  # Check result: Mean peak height.
  expect_that(unique(res$MPH), equals(c(738.9091, NA)))


  # TEST 07 -------------------------------------------------------------------
  # Test that peak ratio Lb by dye can be calculated for data including
  # one locus dropout.

  # Analyse dataframe.
  res <- calculateLb(
    data = set2, ref = NULL, option = "peak", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$Min))
  expect_false(is.null(res$Max))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Dye)))
  expect_true(any(is.na(res$Min)))
  expect_true(any(is.na(res$Max)))
  expect_false(any(is.na(res$Peaks)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(402 / 723))
  expect_that(res$Lb[2], equals(361 / 766))
  expect_that(res$Lb[3], equals(359 / 592))

  expect_that(res$Lb[4], equals(198 / 347))
  expect_that(res$Lb[5], equals(195 / 402))
  expect_that(res$Lb[6], equals(as.numeric(NA)))

  # Check result: Peaks.
  expect_that(res$Peaks[1], equals(6))
  expect_that(res$Peaks[2], equals(6))
  expect_that(res$Peaks[3], equals(4))

  expect_that(res$Peaks[4], equals(6))
  expect_that(res$Peaks[5], equals(6))
  expect_that(res$Peaks[6], equals(4))

  # Check result: Min peak height.
  expect_that(unique(res$Min), equals(c(402, 361, 359, 198, 195, NA)))

  # Check result: Max peak height.
  expect_that(unique(res$Max), equals(c(723, 766, 592, 347, 402, NA)))


  # TEST 08 -------------------------------------------------------------------
  # Test that peak ratio Lb can be calculated for data including
  # one locus dropout.

  res <- calculateLb(
    data = set2, ref = NULL, option = "peak", by.dye = FALSE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Min))
  expect_false(is.null(res$Max))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_true(any(is.na(res$Min)))
  expect_true(any(is.na(res$Max)))
  expect_false(any(is.na(res$Peaks)))
  expect_true(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(359 / 766))

  expect_that(res$Lb[2], equals(as.numeric(NA)))

  # Check result: Peaks.
  expect_that(res$Peaks[1], equals(16))

  expect_that(res$Peaks[2], equals(16))

  # Check result: Min peak height.
  expect_that(unique(res$Min), equals(c(359, NA)))

  # Check result: Max peak height.
  expect_that(unique(res$Max), equals(c(766, NA)))


  # TEST 09 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for unfiltered data.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7324 / 34486))
  expect_that(res$Lb[2], equals(7648 / 34486))
  expect_that(res$Lb[3], equals(7911 / 34486))
  expect_that(res$Lb[4], equals(6774 / 34486))
  expect_that(res$Lb[5], equals(4829 / 34486))
  expect_that(res$Lb[6], equals(5482 / 26214))
  expect_that(res$Lb[7], equals(8098 / 26214))
  expect_that(res$Lb[8], equals(6629 / 26214))
  expect_that(res$Lb[9], equals(6005 / 26214))
  expect_that(res$Lb[10], equals(5548 / 24822))
  expect_that(res$Lb[11], equals(4975 / 24822))
  expect_that(res$Lb[12], equals(7320 / 24822))
  expect_that(res$Lb[13], equals(6979 / 24822))
  expect_that(res$Lb[14], equals(16865 / 50136))
  expect_that(res$Lb[15], equals(10906 / 50136))
  expect_that(res$Lb[16], equals(10649 / 50136))
  expect_that(res$Lb[17], equals(11716 / 50136))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7324))
  expect_that(res$TPH[2], equals(7648))
  expect_that(res$TPH[3], equals(7911))
  expect_that(res$TPH[4], equals(6774))
  expect_that(res$TPH[5], equals(4829))
  expect_that(res$TPH[6], equals(5482))
  expect_that(res$TPH[7], equals(8098))
  expect_that(res$TPH[8], equals(6629))
  expect_that(res$TPH[9], equals(6005))
  expect_that(res$TPH[10], equals(5548))
  expect_that(res$TPH[11], equals(4975))
  expect_that(res$TPH[12], equals(7320))
  expect_that(res$TPH[13], equals(6979))
  expect_that(res$TPH[14], equals(16865))
  expect_that(res$TPH[15], equals(10906))
  expect_that(res$TPH[16], equals(10649))
  expect_that(res$TPH[17], equals(11716))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(34486, 26214, 24822, 50136)))


  # TEST 10 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for unfiltered data
  # and off-ladder peaks removed.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7131 / 34293))
  expect_that(res$Lb[2], equals(7648 / 34293))
  expect_that(res$Lb[3], equals(7911 / 34293))
  expect_that(res$Lb[4], equals(6774 / 34293))
  expect_that(res$Lb[5], equals(4829 / 34293))
  expect_that(res$Lb[6], equals(5482 / 26214))
  expect_that(res$Lb[7], equals(8098 / 26214))
  expect_that(res$Lb[8], equals(6629 / 26214))
  expect_that(res$Lb[9], equals(6005 / 26214))
  expect_that(res$Lb[10], equals(5548 / 24822))
  expect_that(res$Lb[11], equals(4975 / 24822))
  expect_that(res$Lb[12], equals(7320 / 24822))
  expect_that(res$Lb[13], equals(6979 / 24822))
  expect_that(res$Lb[14], equals(16865 / 50136))
  expect_that(res$Lb[15], equals(10906 / 50136))
  expect_that(res$Lb[16], equals(10649 / 50136))
  expect_that(res$Lb[17], equals(11716 / 50136))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7131))
  expect_that(res$TPH[2], equals(7648))
  expect_that(res$TPH[3], equals(7911))
  expect_that(res$TPH[4], equals(6774))
  expect_that(res$TPH[5], equals(4829))
  expect_that(res$TPH[6], equals(5482))
  expect_that(res$TPH[7], equals(8098))
  expect_that(res$TPH[8], equals(6629))
  expect_that(res$TPH[9], equals(6005))
  expect_that(res$TPH[10], equals(5548))
  expect_that(res$TPH[11], equals(4975))
  expect_that(res$TPH[12], equals(7320))
  expect_that(res$TPH[13], equals(6979))
  expect_that(res$TPH[14], equals(16865))
  expect_that(res$TPH[15], equals(10906))
  expect_that(res$TPH[16], equals(10649))
  expect_that(res$TPH[17], equals(11716))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(34293, 26214, 24822, 50136)))


  # TEST 11 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for unfiltered data,
  # with off-ladder peaks, and sex markers removed.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = TRUE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7648 / 27162))
  expect_that(res$Lb[2], equals(7911 / 27162))
  expect_that(res$Lb[3], equals(6774 / 27162))
  expect_that(res$Lb[4], equals(4829 / 27162))
  expect_that(res$Lb[5], equals(5482 / 26214))
  expect_that(res$Lb[6], equals(8098 / 26214))
  expect_that(res$Lb[7], equals(6629 / 26214))
  expect_that(res$Lb[8], equals(6005 / 26214))
  expect_that(res$Lb[9], equals(5548 / 24822))
  expect_that(res$Lb[10], equals(4975 / 24822))
  expect_that(res$Lb[11], equals(7320 / 24822))
  expect_that(res$Lb[12], equals(6979 / 24822))
  expect_that(res$Lb[13], equals(16865 / 50136))
  expect_that(res$Lb[14], equals(10906 / 50136))
  expect_that(res$Lb[15], equals(10649 / 50136))
  expect_that(res$Lb[16], equals(11716 / 50136))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7648))
  expect_that(res$TPH[2], equals(7911))
  expect_that(res$TPH[3], equals(6774))
  expect_that(res$TPH[4], equals(4829))
  expect_that(res$TPH[5], equals(5482))
  expect_that(res$TPH[6], equals(8098))
  expect_that(res$TPH[7], equals(6629))
  expect_that(res$TPH[8], equals(6005))
  expect_that(res$TPH[9], equals(5548))
  expect_that(res$TPH[10], equals(4975))
  expect_that(res$TPH[11], equals(7320))
  expect_that(res$TPH[12], equals(6979))
  expect_that(res$TPH[13], equals(16865))
  expect_that(res$TPH[14], equals(10906))
  expect_that(res$TPH[15], equals(10649))
  expect_that(res$TPH[16], equals(11716))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(27162, 26214, 24822, 50136)))


  # TEST 12 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for filtered data.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = ref1, option = "prop", by.dye = TRUE,
    ol.rm = FALSE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7131 / 33079))
  expect_that(res$Lb[2], equals(7308 / 33079))
  expect_that(res$Lb[3], equals(7911 / 33079))
  expect_that(res$Lb[4], equals(6282 / 33079))
  expect_that(res$Lb[5], equals(4447 / 33079))
  expect_that(res$Lb[6], equals(5033 / 24651))
  expect_that(res$Lb[7], equals(7847 / 24651))
  expect_that(res$Lb[8], equals(6088 / 24651))
  expect_that(res$Lb[9], equals(5683 / 24651))
  expect_that(res$Lb[10], equals(4757 / 22933))
  expect_that(res$Lb[11], equals(4573 / 22933))
  expect_that(res$Lb[12], equals(7005 / 22933))
  expect_that(res$Lb[13], equals(6598 / 22933))
  expect_that(res$Lb[14], equals(16099 / 47097))
  expect_that(res$Lb[15], equals(9805 / 47097))
  expect_that(res$Lb[16], equals(10287 / 47097))
  expect_that(res$Lb[17], equals(10906 / 47097))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7131))
  expect_that(res$TPH[2], equals(7308))
  expect_that(res$TPH[3], equals(7911))
  expect_that(res$TPH[4], equals(6282))
  expect_that(res$TPH[5], equals(4447))
  expect_that(res$TPH[6], equals(5033))
  expect_that(res$TPH[7], equals(7847))
  expect_that(res$TPH[8], equals(6088))
  expect_that(res$TPH[9], equals(5683))
  expect_that(res$TPH[10], equals(4757))
  expect_that(res$TPH[11], equals(4573))
  expect_that(res$TPH[12], equals(7005))
  expect_that(res$TPH[13], equals(6598))
  expect_that(res$TPH[14], equals(16099))
  expect_that(res$TPH[15], equals(9805))
  expect_that(res$TPH[16], equals(10287))
  expect_that(res$TPH[17], equals(10906))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(33079, 24651, 22933, 47097)))


  # TEST 13 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for filtered data
  # and off-ladder peaks removed.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = ref1, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7131 / 33079))
  expect_that(res$Lb[2], equals(7308 / 33079))
  expect_that(res$Lb[3], equals(7911 / 33079))
  expect_that(res$Lb[4], equals(6282 / 33079))
  expect_that(res$Lb[5], equals(4447 / 33079))
  expect_that(res$Lb[6], equals(5033 / 24651))
  expect_that(res$Lb[7], equals(7847 / 24651))
  expect_that(res$Lb[8], equals(6088 / 24651))
  expect_that(res$Lb[9], equals(5683 / 24651))
  expect_that(res$Lb[10], equals(4757 / 22933))
  expect_that(res$Lb[11], equals(4573 / 22933))
  expect_that(res$Lb[12], equals(7005 / 22933))
  expect_that(res$Lb[13], equals(6598 / 22933))
  expect_that(res$Lb[14], equals(16099 / 47097))
  expect_that(res$Lb[15], equals(9805 / 47097))
  expect_that(res$Lb[16], equals(10287 / 47097))
  expect_that(res$Lb[17], equals(10906 / 47097))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7131))
  expect_that(res$TPH[2], equals(7308))
  expect_that(res$TPH[3], equals(7911))
  expect_that(res$TPH[4], equals(6282))
  expect_that(res$TPH[5], equals(4447))
  expect_that(res$TPH[6], equals(5033))
  expect_that(res$TPH[7], equals(7847))
  expect_that(res$TPH[8], equals(6088))
  expect_that(res$TPH[9], equals(5683))
  expect_that(res$TPH[10], equals(4757))
  expect_that(res$TPH[11], equals(4573))
  expect_that(res$TPH[12], equals(7005))
  expect_that(res$TPH[13], equals(6598))
  expect_that(res$TPH[14], equals(16099))
  expect_that(res$TPH[15], equals(9805))
  expect_that(res$TPH[16], equals(10287))
  expect_that(res$TPH[17], equals(10906))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(33079, 24651, 22933, 47097)))


  # TEST 14 -------------------------------------------------------------------
  # Test that proportional Lb by dye can be calculated for filtered data,
  # with off-ladder peaks, and sex markers removed.

  # Analyse dataframe.
  res <- calculateLb(
    data = set1, ref = ref1, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = TRUE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Check return class.
  expect_match(class(res), class(data.frame()))

  # Check that expected columns exist.
  expect_false(is.null(res$Sample.Name))
  expect_false(is.null(res$Marker))
  expect_false(is.null(res$TPH))
  expect_false(is.null(res$Peaks))
  expect_false(is.null(res$Dye))
  expect_false(is.null(res$TPPH))
  expect_false(is.null(res$Lb))

  # Check for NA's.
  expect_false(any(is.na(res$Sample.Name)))
  expect_false(any(is.na(res$Marker)))
  expect_false(any(is.na(res$TPH)))
  expect_false(any(is.na(res$Peaks)))
  expect_false(any(is.na(res$Dye)))
  expect_false(any(is.na(res$TPPH)))
  expect_false(any(is.na(res$Lb)))

  # Check result: Locus balance.
  expect_that(res$Lb[1], equals(7308 / 25948))
  expect_that(res$Lb[2], equals(7911 / 25948))
  expect_that(res$Lb[3], equals(6282 / 25948))
  expect_that(res$Lb[4], equals(4447 / 25948))
  expect_that(res$Lb[5], equals(5033 / 24651))
  expect_that(res$Lb[6], equals(7847 / 24651))
  expect_that(res$Lb[7], equals(6088 / 24651))
  expect_that(res$Lb[8], equals(5683 / 24651))
  expect_that(res$Lb[9], equals(4757 / 22933))
  expect_that(res$Lb[10], equals(4573 / 22933))
  expect_that(res$Lb[11], equals(7005 / 22933))
  expect_that(res$Lb[12], equals(6598 / 22933))
  expect_that(res$Lb[13], equals(16099 / 47097))
  expect_that(res$Lb[14], equals(9805 / 47097))
  expect_that(res$Lb[15], equals(10287 / 47097))
  expect_that(res$Lb[16], equals(10906 / 47097))

  # Check result: Total peak height.
  expect_that(res$TPH[1], equals(7308))
  expect_that(res$TPH[2], equals(7911))
  expect_that(res$TPH[3], equals(6282))
  expect_that(res$TPH[4], equals(4447))
  expect_that(res$TPH[5], equals(5033))
  expect_that(res$TPH[6], equals(7847))
  expect_that(res$TPH[7], equals(6088))
  expect_that(res$TPH[8], equals(5683))
  expect_that(res$TPH[9], equals(4757))
  expect_that(res$TPH[10], equals(4573))
  expect_that(res$TPH[11], equals(7005))
  expect_that(res$TPH[12], equals(6598))
  expect_that(res$TPH[13], equals(16099))
  expect_that(res$TPH[14], equals(9805))
  expect_that(res$TPH[15], equals(10287))
  expect_that(res$TPH[16], equals(10906))

  # Check result: Total profile peak height.
  expect_that(unique(res$TPPH), equals(c(25948, 24651, 22933, 47097)))


  # TEST 15 -------------------------------------------------------------------
  # Test when a marker is missing from a sample.

  # Remove TH01 from the first sample.
  setMissing <- set2[!(set2$Marker == "TH01" & set2$Sample.Name == "SampleA01"), ]

  # Test that proportional Lb by dye can be calculated for filtered data,
  # with off-ladder peaks removed and markers are missing.

  # This work since ref is provided (data is filtered and missing markers added).
  res <- calculateLb(
    data = setMissing, ref = ref2, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Same as above but global Lb.
  res <- calculateLb(
    data = setMissing, ref = ref2, option = "prop", by.dye = FALSE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # This work in version >=1.6.0.9002 since the missing marker is detected and added.
  res <- calculateLb(
    data = setMissing, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Same as above but global Lb (this work in version >=1.6.0.9002).
  res <- calculateLb(
    data = setMissing, ref = NULL, option = "prop", by.dye = FALSE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )


  # TEST 16 -------------------------------------------------------------------
  # Test when all markers in one dye channel is missing from a sample.

  # Remove yellow dye channel from the first sample.
  setMissing <- set2[!(set2$Dye == "Y" & set2$Sample.Name == "SampleA01"), ]

  # Test that proportional Lb by dye can be calculated for filtered data,
  # with off-ladder peaks removed and markers are missing.

  # This work since ref is provided (data is filtered and missing markers added).
  res <- calculateLb(
    data = setMissing, ref = ref2, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Same as above but global Lb.
  res <- calculateLb(
    data = setMissing, ref = ref2, option = "prop", by.dye = FALSE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # This work in version >=1.6.0.9002 since the missing marker is detected and added.
  res <- calculateLb(
    data = setMissing, ref = NULL, option = "prop", by.dye = TRUE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )

  # Same as above but global Lb (this work in version >=1.6.0.9002).
  res <- calculateLb(
    data = setMissing, ref = NULL, option = "prop", by.dye = FALSE,
    ol.rm = TRUE, sex.rm = FALSE, na = NULL, kit = "SGMplus",
    ignore.case = TRUE, word = FALSE, exact = FALSE
  )


  # TEST 17 -------------------------------------------------------------------
  # Test that 'fat' data throws error.

  # Expect error due to 'fat' data.
  expect_error(calculateLb(data = set1Fat))
})