Nothing
negatives_counts <- function() {
counts <- matrix(
c(
10L,
20L,
12L,
20L,
14L,
26L,
100L,
5L,
50L,
60L
),
ncol = 2,
byrow = TRUE,
dimnames = list(
c("NEG_A", "NEG_B", "NEG_C", "GENE1", "GENE2"),
c("S1", "S2")
)
)
list(
counts = counts,
code_class = c(
"Negative",
"Negative",
"Negative",
"Endogenous",
"Endogenous"
)
)
}
test_that("background_levels() computes each statistic per sample", {
d <- negatives_counts()
level <- function(statistic, excluded = character(0)) {
NACHO:::background_levels(d$counts, d$code_class, excluded, statistic)
}
expect_null(level("none"))
expect_equal(level("mean"), c(12, 22))
expect_equal(level("mean_2sd"), c(12 + 2 * 2, 22 + 2 * sqrt(12)))
expect_equal(level("median"), c(12, 20))
expect_equal(level("max"), c(14, 26))
expect_equal(
level("geo"),
c(exp(mean(log(c(10, 12, 14)))), exp(mean(log(c(20, 20, 26)))))
)
expect_equal(level("max", excluded = "NEG_C"), c(12, 20))
})
test_that("background needs negatives", {
d <- negatives_counts()
expect_error(
NACHO:::background_levels(
d$counts,
rep("Endogenous", 5),
character(0),
"mean"
),
class = "nacho_error_bad_argument"
)
expect_error(
NACHO:::background_levels(
d$counts,
d$code_class,
c("NEG_B", "NEG_C"),
"mean_2sd"
),
regexp = "two negative",
class = "nacho_error_bad_argument"
)
})
test_that("apply_background() thresholds or subtracts, and floors only at 0", {
counts <- matrix(
c(5L, 15L, 30L, 8L),
ncol = 2,
dimnames = list(c("a", "b"), c("S1", "S2"))
)
level <- c(10, 9)
expect_identical(
NACHO:::apply_background(counts, level, "threshold"),
matrix(c(10, 15, 30, 9), ncol = 2, dimnames = dimnames(counts))
)
expect_identical(
NACHO:::apply_background(counts, level, "subtract"),
matrix(c(0, 5, 21, 0), ncol = 2, dimnames = dimnames(counts))
)
expect_identical(
NACHO:::apply_background(counts, NULL, "threshold"),
counts * 1
)
})
test_that("missing counts stay missing", {
counts <- matrix(
c(NA, 15L, 30L, 8L),
ncol = 2,
dimnames = list(c("a", "b"), c("S1", "S2"))
)
expect_identical(
NACHO:::apply_background(counts, c(10, 9), "threshold"),
matrix(c(NA, 15, 30, 9), ncol = 2, dimnames = dimnames(counts))
)
expect_identical(
NACHO:::apply_background(counts, c(10, 9), "subtract"),
matrix(c(NA, 5, 21, 0), ncol = 2, dimnames = dimnames(counts))
)
})
test_that("a build without negative probes gives an NA Negative_factor", {
keep <- GSE74821@probes$CodeClass != "Negative"
built <- suppressWarnings(
NACHO:::build_nacho(
counts = GSE74821@counts[keep, 1:3],
probes = GSE74821@probes[keep, ],
samples = GSE74821@samples[1:3, , drop = FALSE],
settings = GSE74821@settings,
thresholds = GSE74821@thresholds,
rcc_type = GSE74821@rcc_type,
provenance = GSE74821@provenance
),
classes = c(
"nacho_warning_n_comp_reduced",
"nacho_warning_metric_unavailable"
)
)
expect_identical(nacho_samples(built)$Negative_factor, rep(NA_real_, 3))
})
test_that("a sample without any kept negative count gets an NA background and a warning", {
d <- negatives_counts()
d$counts[1:3, "S2"] <- NA
for (statistic in c("geo", "mean", "mean_2sd", "median", "max")) {
expect_warning(
level <- NACHO:::background_levels(
d$counts,
d$code_class,
character(0),
statistic
),
class = "nacho_warning_metric_unavailable",
regexp = "S2"
)
expect_false(is.na(level[[1]]), label = statistic)
expect_identical(level[[2]], NA_real_, label = statistic)
}
expect_no_warning(
NACHO:::background_levels(
d$counts,
d$code_class,
character(0),
"geo",
warn = FALSE
)
)
})
test_that("a sample with fewer than two observed negative counts gets an NA mean_2sd background", {
d <- negatives_counts()
d$counts[c("NEG_B", "NEG_C"), "S2"] <- NA
level <- suppressWarnings(
NACHO:::background_levels(d$counts, d$code_class, character(0), "mean_2sd")
)
expect_equal(level, c(12 + 2 * 2, NA))
expect_false(is.nan(level[2]))
d$counts[, "S2"] <- NA
level <- suppressWarnings(
NACHO:::background_levels(d$counts, d$code_class, character(0), "mean_2sd")
)
expect_identical(level[2], NA_real_)
})
test_that("background_levels() forwards the call it is given to its errors", {
d <- negatives_counts()
error <- tryCatch(
NACHO:::background_levels(
d$counts,
d$code_class,
c("NEG_B", "NEG_C"),
"mean_2sd",
call = rlang::call2("load_rcc")
),
error = identity
)
expect_identical(rlang::call_name(error$call), "load_rcc")
})
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.