Default to cores = 3 if parallel = TRUE is supplied with no cores argument
Add compatability for numeric vectors in compare_seurat()
Updated compatability with Seurat v5. compare_seurat()
and seurat_extract()
automatically detect Seurat object version
Change package dependeny version to improve installation
Ability to specify multiple gmt filepaths in the pathway argument of compare_pathways()
. For
examples see here
Fix downsample so it isn't forcing 500 cells #34
- Parallel processing is now built into into the
compare_pathways()
,compare_seurat()
, andcompare_sce()
functions. Users just need to specifyparallel = TRUE
andcores = x
to use parallel processing instead of callingcompare_pathways_parallel()
- Fix downsample #31
- Parallel implementation for
compare_pathways()
viacompare_pathways_parallel()
. Users can specify number of cores to run in the pathway comparison
- The main function of SCPA has been rewritten to be much more efficient with memory usage. This means that RAM availability is likely not going to be limiting on a typical computer i.e. 8GB RAM, even with a huge number of comparisons.
- The speed of the pathway comparisons is also slightly improved
- Added the option to filter gene set size in
compare_seurat()
- Add functionality to choose assay in Seurat object
- Add functionality to deal with SingleCellExperiment objects
sce_extract
can now extract data from SCE objects based on colDatacompare_sce
can now compare pathways directly within an SCE object
- Add functionality to visualise results from SCPA with
plot_rank()
andplot_heatmap()
- Improve messaging with
compare_pathways()
- Allow usage of multiple file types for pathways e.g. gmt, csv
- Fix CRAN archive with multicross and crossmatch
- Release SCPA for multivariate testing of pathways in scRNA-seq data