Pheno pseudobulk

[KevinMLanderos]

This section allows the pipeline to run 'SAMPLE' and 'COMPARE' modules on samples pseudo-bulked by phenotype.

[github-actions]

Unit Test Results

10 tests  ±0   10 ✅ ±0   2m 45s ⏱️ -1s
 2 suites ±0    0 💤 ±0 
 1 files   ±0    0 ❌ ±0 

Results for commit 9113490. ± Comparison against base commit b862c22.

♻️ This comment has been updated with latest results.

[Copilot]

Pull request overview

This PR adds phenotype-based pseudobulking functionality to the TCR toolkit pipeline, allowing samples to be pseudo-bulked by phenotype annotations from Seurat objects and then analyzed through the existing SAMPLE and COMPARE modules.

Changes:

• Added phenotype pseudobulking workflow that processes samples by phenotype when a Seurat GEX object is provided
• Extended the AIRR_CONVERT subworkflow to generate phenotype-specific pseudobulk files from CellRanger data
• Created parallel analysis paths (SAMPLE_PHENO and COMPARE_PHENO) to process phenotype-segregated data
• Enhanced Python scripts to support phenotype metadata extraction and improved data handling in compare_calc.py

Reviewed changes

Copilot reviewed 12 out of 15 changed files in this pull request and generated 13 comments.

Show a summary per file

File	Description
workflows/tcrtoolkit.nf	Main workflow integration of phenotype analysis with conditional execution based on sobject_gex parameter
subworkflows/local/sample_pheno.nf	New subworkflow for sample-level analysis of phenotype-pseudobulked data
subworkflows/local/compare_pheno.nf	New subworkflow for comparison analysis of phenotype-pseudobulked data
subworkflows/local/resolve_samplesheet_pheno.nf	New subworkflow to collect phenotype sample files for downstream processing
subworkflows/local/map_phenotypes.nf	New subworkflow to transform phenotype files and generate phenotype-specific samplesheet
subworkflows/local/airr_convert.nf	Extended to support phenotype pseudobulking for CellRanger input format
subworkflows/local/sample.nf	Removed commented-out emit statements
subworkflows/local/compare.nf	Added collectFile operations for similarity matrices and removed commented code
modules/local/samplesheet/generate_pheno_samplesheet.nf	New module to generate phenotype-specific samplesheet from metadata
modules/local/airr_convert/pseudobulk_phenotype_cellranger.nf	New module to run phenotype-based pseudobulking on CellRanger data
modules/local/airr_convert/pseudobulk_cellranger.nf	Added container directive
conf/modules.config	Added publishDir configurations for phenotype analysis outputs
bin/pseudobulk.py	Added phenotype processing functions and command-line options
bin/create_pheno_samplesheet.py	New script to generate phenotype samplesheet from JSON metadata
bin/compare_calc.py	Improved data handling with better validation and numeric type handling

Comments suppressed due to low confidence (1)

subworkflows/local/compare.nf:56

• The collectFile operations are collecting outputs from COMPARE_CALC.out but then COMPARE_PLOT is still using the original COMPARE_CALC.out channels (lines 50-52), not the collected files. This creates redundant file collection. Either use the collected files in COMPARE_PLOT or remove the collectFile operations if they're not needed.

    COMPARE_CALC.out.jaccard_mat
        .collectFile(name: 'jaccard_mat.csv', sort: true, 
                     storeDir: "${params.outdir}/compare")
        .set { jaccard_mat }

    COMPARE_CALC.out.sorensen_mat
        .collectFile(name: 'sorensen_mat.csv', sort: true, 
                     storeDir: "${params.outdir}/compare")
        .set { sorensen_mat }

    COMPARE_CALC.out.morisita_mat
        .collectFile(name: 'morisita_mat.csv', sort: true, 
                     storeDir: "${params.outdir}/compare")
        .set { morisita_mat }

    COMPARE_PLOT( samplesheet_resolved,
                  COMPARE_CALC.out.jaccard_mat,
                  COMPARE_CALC.out.sorensen_mat,
                  COMPARE_CALC.out.morisita_mat,
                  file(params.compare_stats_template),
                  params.project_name,
                  all_sample_files
                  )

---

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