Filter cell meta data in ReadPNA_Seurat

[ludvigla]

Description

This PR adds a meta data filter to ReadPNA_Seurat to keep the meta.data slot clean. Full meta data can be obtained by setting detailed_meta_data = TRUE.

The columns to keep are defined in the CELL_META_COLS variable. Currently, these are:

CELL_META_COLS <- c(
  "n_umi", "n_edges", "n_antibodies", "isotype_fraction",
  "average_k_core", "reads_in_component", "tau", "tau_type",
  "n_umi1", "n_umi2", "sample"
)

Fixes: PNA-3078

Type of change

• Bug fix (non-breaking change which fixes an issue).
• New feature (non-breaking change which adds functionality).
• Breaking change (fix or feature that would cause existing functionality to not work as expected).
• This change requires a documentation update.

How Has This Been Tested?

Added new test to tests/testthat/test-ReadPNA_Seurat.R

PR checklist:

• My changes generate no new warnings.
• I have added tests that prove my fix is effective or that my feature works.
• I have documented any significant changes to the code in CHANGELOG.md

---

Note

Medium Risk
Default Seurat metadata is slimmer and may omit columns downstream code assumed were present; behavior is opt-in restorable via detailed_meta_data = TRUE.

Overview
ReadPNA_Seurat now loads only a QC-focused subset of PXL cell metadata into meta.data by default, using the new internal CELL_META_COLS list (e.g. n_umi, n_edges, sample). Set detailed_meta_data = TRUE to restore the full __adata__obs table for troubleshooting.

This is a behavior change for workflows that expect extra columns (such as tau_type) without opting in—tests were updated accordingly (e.g. MoleculeRankPlot grouping uses a synthetic sample_id column). Changelog and roxygen/man docs document the new argument; coverage asserts that detailed mode yields more metadata columns than the default.

^{Reviewed by Cursor Bugbot for commit 58788c3. Bugbot is set up for automated code reviews on this repo. Configure here.}

[Copilot]

Pull request overview

Adds an opt-in mechanism to retain full per-cell metadata when loading PNA PXL files into Seurat, while defaulting to a curated subset of QC-relevant columns to keep meta.data smaller/cleaner (Fixes: PNA-2926).

Changes:

• Added detailed_meta_data argument to ReadPNA_Seurat(); when FALSE it filters meta.data to CELL_META_COLS.
• Introduced CELL_META_COLS as a central definition of which cell metadata columns are retained by default.
• Updated documentation, changelog, and added a test to cover the new behavior.

Reviewed changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
R/load_data_pna.R	Adds detailed_meta_data parameter, validates it, and filters cell metadata by default.
R/aaa.R	Defines CELL_META_COLS constant used for default metadata filtering.
tests/testthat/test-ReadPNA_Seurat.R	Adds a test asserting detailed metadata loads more columns than the default.
man/ReadPNA_Seurat.Rd	Documents the new detailed_meta_data argument.
CHANGELOG.md	Notes the new default metadata filtering behavior and how to opt into detailed metadata.

Files not reviewed (1)

• man/ReadPNA_Seurat.Rd: Generated file

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[maxkarlsson]

I think we should drop tau!

[maxkarlsson]

Looks good!

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

Want fixes drafted automatically? Bugbot Autofix can create code changes for findings. A team admin can enable Autofix in the Cursor dashboard.

^{Reviewed by Cursor Bugbot for commit 58788c3. Configure here.}

[cursor]

Test accidentally duplicated, losing vector collapse coverage

Medium Severity

The first test ("fails when X collapses to a vector") was changed from quote(X <- X[, 1]) to quote(X <- X[, 1, drop = FALSE]), making it identical to the second test ("fails when X is a 1-column matrix"). The original test without drop = FALSE exercised the is.null(ncol(X)) branch of the guard; now both tests only exercise the ncol(X) == 1 branch. The vector-collapse code path is no longer covered.

Additional Locations (1)

• tests/testthat/test-ReadPNA_Seurat.R#L49-L50

 

^{Reviewed by Cursor Bugbot for commit 58788c3. Configure here.}