tests/testthat/test_functions.R

In scITD: Single-Cell Interpretable Tensor Decomposition

library(scITD)
library(testthat)
library(Matrix)


test_that("colMeanVars() functionality", {
  donor_by_gene <- rbind(c(9,2,1,5),c(3,3,1,2))
  donor_by_gene <- Matrix(donor_by_gene, sparse = TRUE)
  result <- colMeanVars(donor_by_gene,rowSel = NULL)
  expected_result <- cbind(c(6,2.5,1,3.5),c(9,.25,0,2.25),c(2,2,2,2))
  expected_result <- as.data.frame(expected_result)
  colnames(expected_result) <- c('m','v','nobs')
  expect_equal(result, expected_result)
})

test_that("form_tensor() functionality", {
  expected_result <- test_container$scMinimal_ctype[['CD4+ T']]$pseudobulk
  test_container <- form_tensor(test_container, donor_min_cells=0,
                                norm_method='trim', scale_factor=10000,
                                vargenes_method='norm_var', vargenes_thresh=500,
                                scale_var = TRUE, var_scale_power = .5)
  result <- test_container$scMinimal_ctype[['CD4+ T']]$pseudobulk
  expect_equal(result, expected_result)
})

test_that("get_lm_pvals() functionality", {
  expected_result <- test_container[["gene_score_associations"]]
  test_container <- get_lm_pvals(test_container, n.cores = 1)
  result <- test_container[["gene_score_associations"]]
  expect_equal(result, expected_result)
})

test_that("get_meta_associations() functionality", {
  expected_result <- test_container[["meta_associations"]]
  test_container <- get_meta_associations(test_container, vars_test=c('lanes'), stat_use='pval')
  result <- test_container[["meta_associations"]]
  expect_equal(result, expected_result)
})