tests/testthat/test_plot_arch_for_clusters.R

In snikumbh/archR: Identify Different Architectures of Sequence Elements

context("Test plotting architectures/helper functions")


test_that("xtick vals are properly set", {

    expect_equal(set_xtick_labels(pos_lab = -10:5, xt_freq = 5),
                 list(breaks = c(1,6,11,16), labels = c(-10, -5, 0, 5)))

})

test_that("plot_arch_for_clusters works", {

    fname <- system.file("extdata", "example_data.fa",
                         package = "archR",
                         mustWork = TRUE)
    seq_raw <- suppressMessages(archR::prepare_data_from_FASTA(fname,
                                                        raw_seq = TRUE))

    ## Collect all sequences of clust1
    clust1_idx <- which(unlist(lapply(names(seq_raw), function(x){
        temp <- unlist(strsplit(x, split = "_"))[3] == "clust1"
        temp
    })))
    ##
    p1 <- suppressWarnings(archR::plot_arch_for_clusters(seqs = seq_raw,
                clust_list = list(clust1_idx), pos_lab = 1:100, show = FALSE))
    vdiffr::expect_doppelganger("plot_arch_for_clusters example", p1)
})

# test_that("Given samples matrix object is matrix", {
#   testSamplesMat <- rnorm(10000) # err
#   testFeaturesMat <- matrix(rnorm(10000), nrow = 200)
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "not of type matrix"
#   )
# })
#
# test_that("Given features matrix object is matrix", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 200)
#   testFeaturesMat <- rnorm(10000) # err
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "not of type matrix"
#   )
# })
#
# test_that("Handling empty samples matrix", {
#   testSamplesMat <- matrix()
#   testFeaturesMat <- matrix(rnorm(10000), nrow = 200)
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   # get_clusters(t(samplesMatrix), clustMethod = "kmeans", final_nClust)
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "Empty"
#   )
# })
#
# test_that("Handling empty features matrix", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 200)
#   testFeaturesMat <- matrix()
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "Empty"
#   )
# })
#
#
# test_that("Handling negative nCluster value", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 200)
#   testFeaturesMat <- matrix()
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "Empty"
#   )
# })
#
# test_that("Handling list as nCluster value", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 200)
#   testFeaturesMat <- matrix(rnorm(10000), nrow = 200)
#   testPositionLabels <- seq(50)
#   nClusterTest <- seq(5)
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "Expecting only one value"
#   )
# })
#
# test_that("Handling negative nCluster value", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 200)
#   testFeaturesMat <- matrix(rnorm(10000), nrow = 200)
#   testPositionLabels <- seq(50)
#   nClusterTest <- -5
#   clusterSolTest <- list(clust_sol = c(), reordering_idx = c())
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "non-negative"
#   )
# })


# test_that("Handling different nCluster value than in clustering_sol", {
#   testSamplesMat <- matrix(rnorm(10000), nrow = 50)
#   testFeaturesMat <- matrix(rnorm(10000), nrow = 200, ncol = 50)
#   testPositionLabels <- seq(50)
#   nClusterTest <- 5
#   clusterSolTest <- get_clusters(testSamplesMat,
#     clustMethod = "kmeans",
#     nCluster = nClusterTest + 2
#   )
#   seqsTest <- NULL
#   expect_error(
#     plot_arch_for_clusters(testSamplesMat, testFeaturesMat,
#       nCluster = nClusterTest,
#       clustering_sol = clusterSolTest,
#       seqs = seqsTest,
#       position_labels = testPositionLabels
#     ),
#     "mismatch"
#   )
# })