Skip to contents

Changelog

Source: NEWS.md

multiDEGGs (development version)

multiDEGGs 1.2.1

CRAN release: 2026-04-10

nestedcv and qvalue are no longer required in examples, tests, and vignettes, in compliance with CRAN’s --no-suggests check.

multiDEGGs 1.2.0

CRAN release: 2026-03-24

Updated reference network

  • The reference network internally used by multiDEGGs is now updated with 28,352 additional links obtained from Omnipath. The internal omic_network object contains now a total of 53,035 literature-reported biological links. Running get_diffNetworks() on the same data might now result in differential networks containing new additional links.
  • New citation added.

multiDEGGs 1.1.3

CRAN release: 2026-02-02

Vignettes improvements.

multiDEGGs 1.1.2

CRAN release: 2026-01-15

Patch: get_diffNetworks_singleOmic() now double checks that assayData and metadata are aligned for the layer

multiDEGGs 1.1.1

CRAN release: 2025-10-24

Minor fixes in documentation

multiDEGGs 1.1.0

CRAN release: 2025-07-29

New features for feature augmentation in ML

Two new functions are provided for nested feature engineering. To use them in combination with the nestedcv package their name must be passed to the modifyX parameter of nestcv.glmnet() or nestcv.train().

  • The multiDEGGs_filter() function performs feature selection based entirely on differential network analysis.
  • The multiDEGGs_combined_filter() function combines traditional statistical feature selection (5 options) with differential network analysis.
  • Internally the two predict.multiDEGGs_filter() and predict.multiDEGGs_combined_filter() S3 methods generate predictions by creating a dataset with single and combined predictors based on the filtering results of a multiDEGGs_filter model.
  • The vignette has been updated to showcase the new feature

multiDEGGs 1.0.0

CRAN release: 2025-06-05

Initial Release

  • First public version of multiDEGGs
  • Provides tools for differential network analysis.
  • Can be easily integrated in machine learning pipelines as feature selection method.
  • Supports both single and multi omic analyses.
  • Compatible with R >= 4.4.