Releases: saeyslab/nichenetr
Releases · saeyslab/nichenetr
v2.2.0
- Fixed
make_line_plot()bugs (#273) - Fixed
get_exprs_avg()when cell types have underscores in their names (#274) - Bug fixes for
assign_ligands_to_celltype()(#275, #277) - Fixed mistakes in
generate_info_tables()(grouping metadata and filtered lr network) (#279) - Replaced the GitHub Actions workflow file with something simpler (
.github/workflows/check-standard.yaml) (#290) - Removed limma from the list of dependencies (#290)
- Updated
alias_to_symbol_seurat()(#299) - Changed defaults of
generate_prioritization_tables(): activity_scaled weight is set to 1 by default, and the cell type specificity weights are downweighted (#300) - Modified
make_mushroom_plot(see #300)
v2.1.0
- tidyverse is now a dependency
- New helper functions:
- General:get_weighted_ligand_receptor_links,prepare_ligand_receptor_visualization,make_line_plot
- Circos plot/chord diagram:assign_ligands_to_celltype,get_ligand_target_links_oi,prepare_circos_visualization,make_circos_plot
- Prioritization:generate_info_tables - Changes to existing functions:
-get_lfc_celltype: removedexpression_pctparameter, replaced with...where arguments can directly be provided toSeurat::FindMarkers
-get_ligand_signaling_path: addedminmax_scalingparameter to scale the weights of signaling and regulatory networks
-make*_heatmap_ggplot: returned plot no longer replaces hyphens with dots
-make_mushroom_plot: additional...parameter to pass arguments toggplot2::theme
-make_circos_lr: addedseparate_legendparameter to determine whether the plot and legend are returned as separate objects
- Added nsga2R optimization functions, renamed old model evaluation functions of mlrMBO
-calculate_de: default values changed from NA to NULL
-get_exprs_avg:default values changed from NA to NULL; additional...parameter to pass arguments toSeurat::AverageExpression
-generate_prioritization_tables: addedscenarioparameter - Revised vignettes for readability
parameter_optimizationvignette now contains code on how to submit the job to a computing cluster- Added new vignette on using LIANA to construct the lignad-receptor network part of the prior model