Nextflow

[dimalvovs]

No description provided.

[Copilot]

Pull Request Overview

This PR adds Nextflow pipeline enhancements to the CoGAPS package, including improved resource configuration, container updates, and preprocessing capabilities. The changes modernize the Nextflow workflow with better resource limits, updated container versions, and a new preprocessing step for data filtering.

• Adds a new COGAPS_PREPROCESS process for data filtering and gene selection
• Updates container versions and resource configurations in nextflow.config
• Fixes resource allocation patterns and adds proper resource limits
• Includes comprehensive unit tests for distributed CoGAPS functionality

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
nextflow.config	Updates resource limits, removes check_max calls, and increases default memory/CPU allocations
main.nf	Adds preprocessing step, updates container versions, reorganizes stub/script sections
tests/testthat/test_DistributedCogaps.R	New unit tests for single-cell and genome-wide distributed modes
R/DistributedCogaps.R	Bug fixes for accessing fixed patterns in distributed processing
Dockerfile	Adds sparseMatrixStats dependency
DESCRIPTION	Version bump and date update

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Both genome-wide and single-cell modes are using @fixedPatterns for different matrices. In genome-wide mode, line 236 should likely use @sampleFactors instead of @fixedPatterns for the P matrix, and in single-cell mode, line 258 should use @featureLoadings for the A matrix.

[dimalvovs]

I don't think so.

[Copilot]

Pull Request Overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[drbergman]

Far as I can tell, it looks all in order. Just the one package version print to check on. Were some of these added checks to ensure that we don't return an array of all 0's?

[drbergman]

I assume this version bump makes sense?

[dimalvovs]

Yes, Bioconductor is strict about versions: in x.y.z, y is always odd in devel and even in release. Current release is 28, so current devel must be 29. Although ir may be 29.0, will find out when pushing to Bioconductor.

[dimalvovs]

Far as I can tell, it looks all in order. Just the one package version print to check on. Were some of these added checks to ensure that we don't return an array of all 0's?

Yes, the added tests check for all 0's