Skip to contents

Changelog

Source: NEWS.md

ComplexMap 1.1.1 - 2025-12-02

This is a maintenance and stability release that addresses several critical bugs, improves cross-platform compatibility, and streamlines the advanced analysis workflow.

Workflow Enhancements

  • The summarizeThemes() function has been significantly improved for a more intuitive user experience. It now accepts an add_to_object argument (defaulting to TRUE) that adds theme information (themeId, themeLabel, themePurity) directly to the ComplexMap object’s node table. This makes querying by theme using queryMap(..., type = "theme") seamless and removes the need for manual data merging.

Major Improvements

  • Cross-Platform Parallel Processing: The evaluateComplexes function has been refactored to use the future.apply backend, replacing the Unix-specific parallel::mclapply. This ensures that parallel computation now works reliably on all operating systems, including Windows.

Bug Fixes

  • Fixed a critical runtime error in visualizeMapInteractive that caused the function to fail when the color.by argument was not specified. The function now correctly generates tooltips in all use cases.

  • Resolved a package build failure caused by an incorrect function call in the main workflow vignette (vignette_02-workflow.Rmd). The package can now be built and checked successfully.

  • Corrected a dependency issue where evaluateComplexes would fail if the clue package was not installed. clue has been moved from Suggests to Imports to guarantee its availability for this core function.

Documentation

  • Updated the README.md to accurately reflect that "jaccard" is the default similarityMethod, aligning the documentation with the package’s diversity-first refinement strategy.

  • The advanced analysis vignette (vignette_03-advanced-analysis.Rmd) has been updated to demonstrate the new, streamlined workflow for theme analysis using the refactored summarizeThemes() function.

  • Synchronized versioning and dates across all documentation files (DESCRIPTION, README.md) for consistency.

ComplexMap 1.1.0 - 2025-11-15

Added

  • Quantitative Data Visualization: All three visualization functions (visualizeMapDirectLabels, visualizeMapWithLegend, visualizeMapInteractive) now support a color.by argument to map continuous experimental data directly onto node colors.

  • Robust Parameter Benchmarking: Added a new benchmarkParameters() function to systematically test and optimize mergeThreshold values against a reference complex set.

  • New Similarity Metrics: Added dice and matching_score similarity options to refineComplexList for more flexible complex merging.

  • New Vignette: Added a comprehensive vignette (04-quantitative-visualization.Rmd) to demonstrate the new quantitative data workflow.

Changed

  • New Default Metric: The default similarityMethod in refineComplexList is now "matching_score" to better align refinement logic with the standard MMR evaluation metric.

  • Enhanced Traceability: refineComplexList now returns a mergeMap tibble, providing a clear and reproducible map from original to merged complex IDs.

  • Improved QC Report: Overhauled the output of qcComplexList to be a user-friendly, formatted report with clear warnings and visual cues, rather than raw R output.

  • Metric Renaming: Renamed the "overlap" similarity metric to "simpson" for clarity and correctness.

Fixed

  • Resolved R CMD check errors in vignettes caused by excessive parallel core usage in examples (nCores is now limited).

  • Fixed data type mismatches in vignettes that occurred when joining igraph membership objects with dplyr tibbles.

ComplexMap 1.0.0

  • Initial stable release.
  • Complete refactoring of the codebase to adhere to Bioconductor standards.
  • Introduction of the ComplexMap S3 object for a structured workflow.
  • Added high-level wrapper createComplexMap() for end-to-end analysis.
  • Added new functions for analysis and exploration: summarizeThemes(), queryMap(), and exportNetwork().
  • Added three comprehensive vignettes.
  • Numerous bug fixes and improvements to robustness.