tests/testthat/test_replace_zeros.R

In pmartR/pmartR: Panomics Marketplace - Quality Control and Statistical Analysis for Panomics Data

context('replace missing values')

test_that('as.xxx correctly replaces missing values (0s) with NAs', {
  # Load: isobaricpepData ------------------------------------------------------

  load(system.file('testdata',
    'little_isodata.RData',
    package = 'pmartR'
  ))

  # Mutate all NAs in edata.
  edata[is.na(edata)] <- 0

  # Test: isobaricpepData ------------------------------------------------------

  # Fabricate an isobaricpepData object.
  expect_message(
    isodata <- as.isobaricpepData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'Peptide',
      fdata_cname = 'Sample',
      emeta_cname = 'Protein'
    ),
    '400 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(isodata)$data_info$num_edata,
    150
  )
  expect_equal(
    attributes(isodata)$data_info$num_miss_obs,
    400
  )
  expect_equal(
    round(attributes(isodata)$data_info$prop_missing, 4),
    0.2222
  )
  expect_equal(
    attributes(isodata)$data_info$num_samps,
    12
  )

  # Ensure the number of replaced elements is correct.
  expect_equal(
    sum(is.na(isodata$e_data)),
    400
  )

  # Examine the class of the new isodata object.
  expect_s3_class(
    isodata,
    c('isobaricpepData', 'pepData')
  )

  # Load: lipidData ------------------------------------------------------------

  load(system.file('testdata',
    'lipidData.RData',
    package = 'pmartR'
  ))

  # Transmogrify all NAs in edata.
  edata[is.na(edata)] <- 0

  # Test: lipidData ------------------------------------------------------------

  # Produce a lipidData object.
  expect_message(
    ldata <- as.lipidData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'LipidCommonName',
      fdata_cname = 'Sample_Name',
      emeta_cname = 'LipidClass'
    ),
    '884 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(ldata)$data_info$num_edata,
    146
  )
  expect_equal(
    attributes(ldata)$data_info$num_miss_obs,
    884
  )
  expect_equal(
    round(attributes(ldata)$data_info$prop_missing, 4),
    0.5504
  )
  expect_equal(
    attributes(ldata)$data_info$num_samps,
    11
  )

  # Ensure the number of replaced elements is correct.
  expect_equal(
    sum(is.na(ldata$e_data)),
    884
  )

  # Examine the class of the new ldata object.
  expect_s3_class(
    ldata,
    'lipidData'
  )

  # Load: metabData ------------------------------------------------------------

  load(system.file('testdata',
    'metaboliteData.RData',
    package = 'pmartR'
  ))

  # Transmute all NAs in edata.
  edata[is.na(edata)] <- 0

  # Test: metabData ------------------------------------------------------------

  # Forge a metabData object.
  expect_message(
    mdata <- as.metabData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'Metabolite',
      fdata_cname = 'SampleID',
      emeta_cname = 'MClass'
    ),
    '148 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(mdata)$data_info$num_edata,
    80
  )
  expect_equal(
    attributes(mdata)$data_info$num_miss_obs,
    148
  )
  expect_equal(
    round(attributes(mdata)$data_info$prop_missing, 4),
    0.1542
  )
  expect_equal(
    attributes(mdata)$data_info$num_samps,
    12
  )

  # Ensure the number of replaced elements is accurate.
  expect_equal(
    sum(is.na(mdata$e_data)),
    148
  )

  # Examine the class of the new mdata object.
  expect_s3_class(
    mdata,
    'metabData'
  )

  # Load: nmrData --------------------------------------------------------------

  load(system.file('testdata',
    'nmrData.RData',
    package = 'pmartR'
  ))

  # Alter the original nmr data frame to contain zeros
  set.seed(22)

  # Select row indices.
  rows <- sample(1:38, 338, replace = TRUE)

  # Select column indices.
  columns <- sample(2:42, 338, replace = TRUE)

  # Loop through each row column pair.
  for (e in 1:338) {
    # Replace current value in edata with zero.
    edata[rows[[e]], columns[[e]]] <- 0
  }

  # Test: nmrData --------------------------------------------------------------

  # Fabricate an nmrData object.
  expect_message(
    nmrdata <- as.nmrData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'Metabolite',
      fdata_cname = 'SampleID',
      emeta_cname = 'nmrClass'
    ),
    '302 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(nmrdata)$data_info$num_edata,
    38
  )
  expect_equal(
    attributes(nmrdata)$data_info$num_miss_obs,
    302
  )
  expect_equal(
    round(attributes(nmrdata)$data_info$prop_missing, 4),
    0.1938
  )
  expect_equal(
    attributes(nmrdata)$data_info$num_samps,
    41
  )

  # Ensure the number of replaced elements is correct.
  expect_equal(
    sum(is.na(nmrdata$e_data)),
    302
  )

  # Examine the class of the new nmrdata object.
  expect_s3_class(
    nmrdata,
    'nmrData'
  )

  # Load: pepData --------------------------------------------------------------

  load(system.file('testdata',
    'little_pdata.RData',
    package = 'pmartR'
  ))

  # Transfigure all NAs in edata.
  edata[is.na(edata)] <- 0

  # Test: pepData --------------------------------------------------------------

  # Construct a pepData object.
  expect_message(
    pdata <- as.pepData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'Mass_Tag_ID',
      fdata_cname = 'SampleID',
      emeta_cname = 'Protein'
    ),
    '341 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(pdata)$data_info$num_edata,
    150
  )
  expect_equal(
    attributes(pdata)$data_info$num_miss_obs,
    341
  )
  expect_equal(
    round(attributes(pdata)$data_info$prop_missing, 4),
    0.1894
  )
  expect_equal(
    attributes(pdata)$data_info$num_samps,
    12
  )

  # Ensure the number of replaced elements is accurate.
  expect_equal(
    sum(is.na(pdata$e_data)),
    341
  )

  # Examine the class of the new pdata object.
  expect_s3_class(
    pdata,
    'pepData'
  )

  # Load: proData --------------------------------------------------------------

  load(system.file('testdata',
    'little_prdata.RData',
    package = 'pmartR'
  ))

  # Transmute all NAs in edata.
  edata[is.na(edata)] <- 0

  # Test: proData --------------------------------------------------------------

  # Generate a proData object.
  expect_message(
    prdata <- as.proData(
      e_data = edata,
      f_data = fdata,
      e_meta = emeta,
      edata_cname = 'Reference',
      fdata_cname = 'SampleID',
      emeta_cname = 'PClass'
    ),
    '234 instances of 0 have been replaced with NA'
  )

  # Check that the elements of the data_info attribute are all correct.
  expect_equal(
    attributes(prdata)$data_info$num_edata,
    150
  )
  expect_equal(
    attributes(prdata)$data_info$num_miss_obs,
    234
  )
  expect_equal(
    round(attributes(prdata)$data_info$prop_missing, 4),
    0.1418
  )
  expect_equal(
    attributes(prdata)$data_info$num_samps,
    11
  )

  # Ensure the number of replaced elements is accurate.
  expect_equal(
    sum(is.na(prdata$e_data)),
    234
  )

  # Examine the class of the new prdata object.
  expect_s3_class(
    prdata,
    'proData'
  )
})