NEWS
nipalsMCIA 1.3.2 (2024-12-10)
Major changes
Minor improvements and bugfixes
- Added
harmonize = FALSE option to nipals_multiblock() to allow disabling (slow) sample harmonization for large datasets.
- Fixed a bug in
nipals_multiblock() to allow the numPCs =1 argument (i.e. ability to compute a single global score).
nipalsMCIA 1.2.1 (2024-08-31)
Major changes
Minor improvements and bugfixes
- Updated 'Single-Cell-Analysis' vignette to remove potential issue with BiocFileCache having two copies of data.
- Updated author list and citation.
- Updated references in readme.
nipalsMCIA 1.2.0 (2024-04-26)
Major changes
Minor improvements and bugfixes
- Clarified the dataset used in the single-cell vignette and reformatted the workflow diagram
- Removed the final log transformation in the single-cell vignette when saving for MCIA
- Minor plotting improvements (removed redundancies in legend names, consistency of axis text, etc.)
nipalsMCIA 1.1.0 (2024-03-21)
Major changes
- Added
BiocFileCache to replace previous method of downloading large SC datasets.
Minor improvements and bugfixes
- Updated
nipals_iter to only use var(gs) to compute the significance of global scores.
- Added to documentation for
predict_gs() warning against use with CPCA deflation.
nipalsMCIA 0.99.7 (2023-10-07)
Minor improvements and bug fixes
- Updated README to reflect MAE changes.
- Updated the citation.
- Fixed bug in documentation for
nipals_multiblock.
- Stopped downloading one of the files for the single cell analysis vignette.
nipalsMCIA 0.99.6 (2023-09-09)
Major changes
- Switched primary input to
nipals_multiblock() to a MultiAssayExperiment object.
nipals_multiblock() now outputs an object of the NipalsResult class.- Converted all downstream analysis functions to work with the
NipalsResult class.
Minor improvements and bug fixes
- Added
simple_mae() function to convert a list of dataframes to a MultiAssayExperiment object.
- Fixed missing
\value fields in man page for NipalsResult class.
nipalsMCIA 0.99.5 (2023-06-22)
Major changes
- Bumping version number to trigger re-build following bug in BiocCheck.
nipalsMCIA 0.99.4 (2023-06-01)
Major changes
- Fixed data corruption issues from v0.99.3.
nipalsMCIA 0.99.3 (2023-06-01)
Major changes
- Changed the eigenvalue calculation in
nipals_iter() to compute the variance of the global score at each deflation step. Prior versions used an SVD method to compute the singular values of the deflated data matrix directly.
- Fixed bug in
projection_plot() where there was a mismatch between color labels and plotting order.
- Added parameter to
nipals_multiblock that specifies whether the samples are in the rows or the columns.
Minor improvements and bug fixes
- Changed vignette styling from
rmdformats to BiocStyle and added installation sections to all of the vignettes.
- Removed empty helper.R file.
- Added significantly more unit testing.
- Fixed bug in
projection_plot() when metadata was provided but no color_col was selected.
- Renamed the associated output of
col_preproc_method in nipals_multiblock. The metadata field is also now available in the output independent of whether metadata is provided in the input.
- Added checks for consistency in sample names across data blocks and metadata.
nipalsMCIA 0.99.2 (2023-03-25)
Major changes
- Shrank the vignettes sizes (especially the single cell analysis vignette).
Minor improvements and bug fixes
- Restructured the single cell analysis vignette to be more streamlined and have more explanations.
- Add an additional single cell data file to the repository using
piggyback.
nipalsMCIA 0.99.1 (2023-02-26)
Major changes
- Included support for
MultiAssayExperiment in nipals_multiblock.
- Improved access to existing data objects.
Minor improvements and bug fixes
- Made
get_colors() more flexible for different color palette options.
nipalsMCIA 0.99.0 (2022-10-21)
- Added single cell data to the repository using
piggyback.