Skip to content

Unit Tests for Visualization and Clustering Functions #47

Open

Description

@bianjh-cloud
No description provided.

Activity

  1. bianjh-cloud commented on Mar 24, 2026

    @bianjh-cloud
    CollaboratorAuthor

    split_featurePlot

    • returns arranged ggplots in list format (for available genes)
    • computes ncol/nrow when NA
    • runs when provided with nrow = 1
    • errors on invalid inputs (seurat object, unavailable gene(s), slot, metadata columns)
    • returns NULL silently with return_list = FALSE
    • works with scalar numeric cutoff values
    • runs with (isolated cases) label = TRUE, order = TRUE, slot = 'data', single split / numeric metadata variable
    • handles empty feature list
    • runs when cutoffs are asymmetric (e.g. when user inputs 3.5 and q10)
  2. changed the title [-]Unit Tests for split_featurePlot, make_bubble_plot, make_volcano_plot[/-] [+]Unit Tests for split_featurePlot[/+] on Mar 24, 2026
  3. self-assigned this
    on Apr 17, 2026
  4. changed the title [-]Unit Tests for split_featurePlot[/-] [+]Unit Tests for split_featurePlot, make_bubble_plot, make_volcano_plot.R[/+] on Apr 17, 2026
  5. bianjh-cloud commented on Apr 17, 2026

    @bianjh-cloud
    CollaboratorAuthor

    make_bubble_plot.R

    • - bubble plot is successful for (BRCA)
    • - bubble plot includes all feature and identity combinations
    • - bubble plot supports scaled average expression
    • - bubble plot respects custom ident columns
    • - bubble plot errors on invalid ident
    • - bubble plot errors when requested features are absent
  6. changed the title [-]Unit Tests for split_featurePlot, make_bubble_plot, make_volcano_plot.R[/-] [+]Unit Tests for Visualization Functions[/+] on Apr 17, 2026
  7. bianjh-cloud commented on Apr 17, 2026

    @bianjh-cloud
    CollaboratorAuthor

    make_volcano_plot.R

    Confirms multiple parameter combinations (significant, logfc, pval) all still return ggplot objects.

    • single-gene input
    • zero p-value input

    Confirms error handling for missing required columns (e.g., missing p_val_adj or missing expected fold-change column).

  8. linked a pull request that will close this issuetest: add Visualization Unit Tests #49on Apr 17, 2026
  9. bianjh-cloud commented on Apr 22, 2026

    @bianjh-cloud
    CollaboratorAuthor

    seurat_clustering.R and cluster_metrics.R

    • clustering produces expected reductions (PCA with 1-50 dimensions, UMAP with 3 dimensions)
    • cluster assignment and nearest-neighbor graphs
    • resolution and algorithm parameters propagate correctly to FindClusters
    • clear error messages for invalid inputs (invalid npcs, resolution, algorithm values).
  10. changed the title [-]Unit Tests for Visualization Functions[/-] [+]Unit Tests for Visualization and Clustering Functions[/+] on Apr 22, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

SCOTRepoName

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions