tests/testthat/test_design.R

In fedeago/NewWave: Negative binomial model for scRNA-seq

context("Test newModel class and methods.")
set.seed(1234)


test_that("newFit works with genewise dispersion", {

    bio <- gl(2, 3)
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    m <- newFit(counts, X=model.matrix(~bio), commondispersion = TRUE)
    m <- newFit(counts, X=model.matrix(~bio), commondispersion = FALSE)
    m <- newFit(counts, X=model.matrix(~bio), verbose = TRUE)
})

test_that("newFit stops if one gene has only 0 counts", {
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    counts <- rbind(counts, rep(0, ncol(counts)))
    expect_error(newFit(counts), "only 0 counts")
})

test_that("newFit stops if one sample has only 0 counts", {
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    counts <- cbind(counts, rep(0, nrow(counts)))
    expect_error(newFit(counts), "only 0 counts")
})

test_that("newFit works without X and V", {
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    m1 <- newFit(counts, X = matrix(0, ncol=1, nrow=ncol(counts)))
    m2 <- newFit(counts, V = matrix(0, ncol=1, nrow=nrow(counts)))
    m3 <- newFit(counts, X = matrix(0, ncol=1, nrow=ncol(counts)),
                  V = matrix(0, ncol=1, nrow=nrow(counts)))

    expect_equal(sum(as.vector(m1@beta)), 0)
    expect_equal(sum(as.vector(m2@gamma)), 0)
    expect_equal(sum(as.vector(m3@beta)), 0)
    expect_equal(sum(as.vector(m3@gamma)), 0)

})

test_that("newFit gives the same results with matrix and SE", {
    counts <- matrix(rnbinom(60, mu=5,size=2), nrow=10, ncol=6)
    se <- SummarizedExperiment(counts)

    set.seed(1234)
    m1 <- newFit(counts)
    set.seed(1234)
    m2 <- newFit(se)
    expect_equal(m1, m2)
})

test_that("newFit gives the same results with matrix and formula", {
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    bio <- gl(2, 3)
    gcc <- rnorm(10)
    se <- SummarizedExperiment(counts, colData=data.frame(Bio=bio),
                               rowData=data.frame(GCC=gcc))

    m1 <- newFit(se, X = model.matrix(~bio))
    m2 <- newFit(se, X = "~Bio")
    expect_equivalent(m1, m2)

    m3 <- newFit(se, V = model.matrix(~gcc))
    m4 <- newFit(se, V = "~GCC")
    expect_equivalent(m3, m4)

    # misstyping
    expect_error(newFit(se, V = "~gc"), "V must be a matrix or a formula")

    # colData / rowData missmatch
    expect_error(newFit(se, V = "~BIO"), "V must be a matrix or a formula")

})

test_that("newSim works", {
    a <- newmodel(n=5, J=10)
    sim <- newSim(a)

    expect_true(all(NewWave:::.is_wholenumber(sim$counts)))

})

test_that("newMu and have the right dimensions", {
    bio <- gl(2, 3)
    counts <- matrix(rpois(60, lambda=5), nrow=10, ncol=6)
    m <- newFit(counts, X=model.matrix(~bio), commondispersion = TRUE)

    expect_equal(dim(newMu(m)), c(numberSamples(m), numberFeatures(m)))
    expect_equal(dim(newLogMu(m)), c(numberSamples(m), numberFeatures(m)))
    expect_equal(dim(newW(m)), c(numberSamples(m), numberFactors(m)))
    expect_equal(length(newTheta(m)), numberFeatures(m))
    expect_equal(length(newZeta(m)), numberFeatures(m))
})

test_that("Initialization works", {

    ## no arguments specified
    newmodel()

    ## specify W
    mat <- matrix(rnorm(10), ncol=2)
    m <- newmodel(W = mat)
    expect_equal(numberSamples(m), nrow(mat))

    ## specify X
    m <- newmodel(X = mat)
    expect_equal(numberSamples(m), nrow(mat))

    ## specify V
    m <- newmodel(V = mat)
    expect_equal(numberFeatures(m), nrow(mat))


    ## specify empty X
    m <- newmodel(X = matrix(0, ncol=0, nrow=10))
    expect_equal(numberSamples(m), 10)

    ## specify empty V
    m <- newmodel(V = matrix(0, ncol=0, nrow=10))
    expect_equal(numberFeatures(m), 10)

    ## specify empty X and V
    m <- newmodel(X = matrix(0, ncol=0, nrow=10),
                   V = matrix(0, ncol=0, nrow=10))
    expect_equal(numberSamples(m), 10)
    expect_equal(numberFeatures(m), 10)

})