Added

  • MAS-Seq tutorial (still under construction)

Changes

  • Bugfix in ModuleConnectivity and ReassignModules.

Added

  • ReassignModules function.

Changes

  • New option in ModuleConnectivity to use corSparse to compute the correlation, which greatly reduces runtime and memory usage.
  • New option in ModuleConnectivity to automatically reassign features to different modules if kME is negative.

Added

  • None

Changes

  • Fixed a bug in ResetModuleNames.

Added

  • None

Changes

  • New error checking in MetaspotsByGroups
  • MetacellsByGroups now keeps track of which cells were merged.

Added

  • None

Changes

  • New warning in SetupForWGCNA if the user selects a very small number of genes.
  • MetaspotsByGroups uses sparse matrix format internally.

Added

  • None

Changes

  • Bugfix in MetaspotsByGroups.

Added

  • MetaspotsByGroups to aggregate neighboring ST spots prior to network analysis.
  • Tutorial for applying hdWGCNA to spatial transcriptomics datasets.

Changes

  • None

Added

  • None

Changes

  • networkType option in TestSoftPowers.

Added

  • Differential Module Eigengene (DME) tutorial
  • FindDMEs function
  • FindAllDMEs function
  • PlotDMEsVolcano function

Changes

  • None

Added

  • None

Changes

  • New data format check in MetacellsByGroups to ensure that the selected slot is present in the selected assay.
  • SetDatExpr now backs up to the full dataset if the metacell dataset isn’t found.
  • Changed some text to clarify some points in the Module Preservation tutorial.

Added

  • None

Changes

  • Bugfix in ModuleConnectivity that caused kMEs to be out of order.

Added

  • Tutorial for using SCTransform normalized data.

Changes

  • None

Added

  • None

Changes

  • Now includes options for other types of correlations in ModuleConnectivity.

Added

  • None

Changes

  • Reverted the exclude_grey flag back to doing nothing in the ModuleEigengenes function because it messed up some downstream tasks, will resolve in a future update.
  • ProjectModules now excludes modules with too many missing genes in the query dataset, tunable by the overlap_proportion parameter.

Added

  • None

Changes

  • By default, ModuleEigengenes does not compute MEs for the grey module. User can change behavior with the exclude_grey flag.

Added

  • None

Changes

  • Bugfix in ProjectModules.
  • Bugfix in MetacellsByGroups

Added

  • New tutorial for consensus co-expression network analysis.

Changes

  • ModuleNetworkPlot and RunModuleUMAP now checks if ModuleConnectivity has been computed in order to throw a more informative error.
  • GetTOM checks if the TOM file exists in order to throw a more informative error.

Added

  • None

Changes

  • New warning message in MetacellsByGroups if there are any groups that are excluded by min_cells.
  • Assay in Metacell seurat object is now the same as the assay supplied to MetacellsByGroups, instead of the default “RNA”.
  • ModuleEigengenes takes assay as an argument, clears up some issues with RunHarmony.
  • ModuleEigengenes doesn’t require a “counts” slot to be present in the given assay, but now it throws an error if the normalized data slot is missing.

Added

  • None

Changes

  • ConstructNetwork now checks if the TOM file already exists, and whether the user wants to overwrite the existing TOM.

Added

  • None

Changes

  • Added new arguments to MetacellsByGroups and ConstructMetacells to exclude very small groups (min_cells), to reach a target number of metacells (target_metacells), and to exclude metacells with too much overlap (max_shared).

Added

  • None

Changes

  • Bug fix so ModuleTraitCorrelation can be run with a single trait.

Added

  • GetHubGenes function to extract the top hub genes from the module assignment table.

Changes

  • ConstructMetacells stores run statistics as a table, and has a new option to exclude metacells that overlap with each other.

Added

  • ModuleEigengenes checks to make sure the data has been scaled.

Changes

  • None

Added

  • Wrote docstring for PlotKMEs so it would actually be included in the package.

Changes

  • None

Added

  • PlotKMEs function to visualize the genes in a module ranked by kME.

Changes

  • Changed the name of the package from scWGCNA to hdWGCNA, since we plan to accommodate data types beyond single cell alone.
  • Updated tutorial to recommend computing kMEs in a specific cell population.