test: add Visualization Unit Tests by bianjh-cloud · Pull Request #49 · CCBR/SCOT

bianjh-cloud · 2026-04-17T17:26:16Z

Changes

make_volcano_plot.R
Refactored volcano logic to use snake_case (avg_log_2_fc), aligned significance labels, and ensured consistent internal data-frame column naming.

Regenerated docs to match the updated argument/field naming in the R function.

Added comprehensive volcano tests covering object creation, transformations, significance logic, edge cases, and missing-column validation.

split_featurePlot.R
Added/updated split feature plot tests for layout behavior, return types, parameter options, and invalid-input handling.

make_bubble_plot.R
Added unit tests for bubble plot creation, structure checks, identity handling, scaling behavior, and expected error cases.

Issues

issue 47 - see checklist

PR Checklist

(~~Strikethrough~~ any points that are not applicable.)

This comment contains a description of changes with justifications, with any relevant issues linked.
Write unit tests for any new features, bug fixes, or other code changes.
~~[ ] Update the docs if there are any API changes (roxygen2 comments, vignettes, readme, etc.).~~
~~[ ] Update NEWS.md with a short description of any user-facing changes and reference the PR number. Follow the style described in https://style.tidyverse.org/news.html~~
Run devtools::check() locally and fix all notes, warnings, and errors.
R-CMD-check succeeds on the most recent user commit.

- Add test-make_bubble_plot.R and test-make_volcano_plot.R - Update test-split_featurePlot.R - Remove source() calls from test-make_volcano_plot.R - Rename avg_log2FC to avg_log_2_fc in make_volcano_plot.R and tests - Hardcode fixed math expectations (log10 transforms) - Regenerate man/make_volcano_plot.Rd Generated using AI

Add missing test helper that defines selectData(), needed by test-make_bubble_plot.R and test-split_featurePlot.R at load time.

…n, seurat_clustering, and cluster_metrics

kelly-sovacool

Significant work is needed to improve these tests. Some patterns I noticed:

Make sure you're testing the behavior of SCOT functions, not functions from base R or other packages, and not merely the function definitions.
Make sure each test actually runs the SCOT functions mentioned in the test_that description, or tests the results of a previously-run SCOT function.
Try to make the tests for functions that return ggplot objects do more than just check that the result is a ggplot. You can test that the various function arguments modify the plots as expected by extracting the plot data, layers, and other plot attributes.

kelly-sovacool · 2026-05-01T15:25:47Z

-    resolution = 0.8,
-    algorithm = 3,
+    resolution = resolution,
+    algorithm = algorithm,


good catch! using parameters is important

kelly-sovacool · 2026-05-01T16:02:30Z

+  ncl_low <- nlevels(SeuratObject::Idents(low_res))
+  ncl_high <- nlevels(SeuratObject::Idents(high_res))
+
+  expect_true(ncl_high >= ncl_low)


since you set seeds above (good job!), you can use the actual values rather than this conditional

kelly-sovacool · 2026-05-01T16:03:42Z

+  brca.data = selectData("wu_et_al_BRCA")
+  input_cells <- colnames(brca.data)
+
+  set.seed(42)
+  brca.clustered = seurat_clustering(so_in = brca.data, npcs_in = 30)


follow the tidyverse style guide

Suggested change

brca.data = selectData("wu_et_al_BRCA")

input_cells <- colnames(brca.data)

set.seed(42)

brca.clustered = seurat_clustering(so_in = brca.data, npcs_in = 30)

brca_data <- selectData("wu_et_al_BRCA")

input_cells <- colnames(brca.data)

set.seed(42)

brca_clustered <- seurat_clustering(so_in = brca.data, npcs_in = 30)

kelly-sovacool · 2026-05-01T16:06:31Z

+test_that("split_featurePlot returns NULL silently with return_list = FALSE", {
+  so <- brca_data
+  res <- split_featurePlot(
+    so = so,
+    features = features[1],
+    split_ident = "orig.ident",
+    plot_image = FALSE,
+    return_list = FALSE,
+    reduction = "umap"
+  )
+  expect_null(res)
+})
+
+test_that("split_featurePlot runs when provided with only ncol", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    ncol = 2,
+    nrow = NA,
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot works with scalar numeric cutoff values", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    min.cutoff = 0,
+    max.cutoff = 5,
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot runs with label = TRUE", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    label = TRUE,
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot works with order = TRUE", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    order = TRUE,
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot works with slot = 'data'", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    slot = "data",
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot works with single split group", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  # Create metadata with single group value
+  so$single_group <- "group_1"
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "single_group",
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+
+test_that("split_featurePlot works with numeric split_ident values", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  # Create numeric metadata
+  so$numeric_group <- sample(1:3, ncol(so), replace = TRUE)
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "numeric_group",
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})
+
+test_that("split_featurePlot handles empty feature list", {
+  so <- brca_data
+
+  res <- split_featurePlot(
+    so = so,
+    features = character(0),
+    split_ident = "orig.ident",
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_equal(length(res), 0)
+})
+
+test_that("split_featurePlot with asymmetric cutoff pair", {
+  so <- brca_data
+  single_feature <- rownames(so@assays$SCT@scale.data)[1]
+
+  res <- split_featurePlot(
+    so = so,
+    features = single_feature,
+    split_ident = "orig.ident",
+    min.cutoff = "q10",
+    max.cutoff = 3.5,
+    plot_image = FALSE,
+    return_list = TRUE,
+    reduction = "umap"
+  )
+
+  expect_true(is.list(res))
+  expect_true(inherits(res[[single_feature]], "ggplot"))
+})


perhaps you could add expectations that do more than only check that the function was able to successfully return lists of ggplots, and actually test that the different function arguments you provide have an effect on the ggplot objects themselves, e.g. testing the layers, labels, plot data, etc;

added some additional expectations for checking layers and plot data

kelly-sovacool · 2026-05-01T16:12:59Z

@@ -15,6 +15,25 @@
 #' @return The updated Seurat single cell object with recalculated PCA,
 #' neighbors, UMAP and clusters
 seurat_clustering <- function(so_in, npcs_in, resolution = 0.8, algorithm = 3) {


I think we should consider renaming this function to a verb. see https://style.tidyverse.org/functions.html @wong-nw

Co-authored-by: Kelly Sovacool, PhD <kelly-sovacool@users.noreply.github.com>

bianjh-cloud and others added 4 commits March 23, 2026 10:05

rewrote calc_sc_gsea

ed645ff

chore: Merge branch 'main' into iss45_rewrite_calc_sc

747e403

test(calc_sc_gsea_score): update snake_case and hardcoded expectations

88bde28

bianjh-cloud linked an issue Apr 17, 2026 that may be closed by this pull request

Unit Tests for Visualization and Clustering Functions #47

Open

github-actions Bot added the SCOT RepoName label Apr 17, 2026

bianjh-cloud and others added 7 commits April 21, 2026 08:31

test: add helper-selectData.R with selectData() helper function

f55dab0

Add missing test helper that defines selectData(), needed by test-make_bubble_plot.R and test-split_featurePlot.R at load time.

test: add unit tests for convert_human_gene_list, run_batch_correctio…

50c328b

…n, seurat_clustering, and cluster_metrics

test: add fixture file and selectData helper for split_featurePlot tests

5abb667

feat: set FindClusters() to use function parameters

b7898e7

feat: update seurat_clustering to use parameters, add error catching

b31562a

fix: set replace gns with genes

89bf1c5

ci: 🤖 auto-format

0faf7b8

bianjh-cloud requested review from kelly-sovacool and wong-nw April 29, 2026 14:15

kelly-sovacool changed the title ~~Visualization Unit Tests~~ test: add Visualization Unit Tests Apr 29, 2026

kelly-sovacool and others added 2 commits May 1, 2026 11:13

Merge branch 'main' into 47-unit-tests-for-split_featureplot

1de3773

ci: 🤖 auto-format

7fde883

kelly-sovacool requested changes May 1, 2026

View reviewed changes

kelly-sovacool reviewed May 1, 2026

View reviewed changes

kelly-sovacool mentioned this pull request May 1, 2026

test: add unit tests for seurat_clustering and cluster_metrics #52

Open

4 tasks

bianjh-cloud and others added 4 commits May 7, 2026 15:10

feat: downsample data, addressed comments

d6c1fbb

ci: 🤖 auto-format

ca42d15

Update tests/testthat/test-seurat_clustering.R

92d67ba

Co-authored-by: Kelly Sovacool, PhD <kelly-sovacool@users.noreply.github.com>

feat: modified convert_human_gene_list.R

9ad9411

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

test: add Visualization Unit Tests#49

test: add Visualization Unit Tests#49
bianjh-cloud wants to merge 17 commits into
mainfrom
47-unit-tests-for-split_featureplot

bianjh-cloud commented Apr 17, 2026 •

edited by kelly-sovacool

Loading

Uh oh!

kelly-sovacool left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Uh oh!

kelly-sovacool May 1, 2026

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Uh oh!

bianjh-cloud Jun 3, 2026

Uh oh!

kelly-sovacool May 1, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

bianjh-cloud commented Apr 17, 2026 • edited by kelly-sovacool Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

Issues

PR Checklist

Uh oh!

kelly-sovacool left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Choose a reason for hiding this comment

Uh oh!

kelly-sovacool May 1, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

kelly-sovacool May 1, 2026

Choose a reason for hiding this comment

Uh oh!

bianjh-cloud Jun 3, 2026

Choose a reason for hiding this comment

Uh oh!

kelly-sovacool May 1, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

bianjh-cloud commented Apr 17, 2026 •

edited by kelly-sovacool

Loading