NEWS

singleCellTK 2.23.6 (2026-10-08)

Help pages link to other packages' topics with package anchors (for example \link[SummarizedExperiment]{colData}), as R CMD check and Bioconductor now require. No change to functionality.

singleCellTK 2.23.5 (2026-10-08)

runFastMNN() hides deprecation warnings raised inside batchelor's fastMNN() and reducedMNN(), as runMNNCorrect() already did for mnnCorrect(). batchelor still calls deprecated scuttle functions. Other warnings are unchanged.

singleCellTK 2.23.4 (2026-10-07)

Fixed feature selection on the Celda tab of the Shiny app. Both the Scran_modelGeneVar and SeuratFindHVG options failed: the first called a function that does not exist, the second was never run because its option value did not match, and both requested the top genes with an argument that getTopHVG() does not accept. With SeuratFindHVG, only the vst method works with Seurat 5. The selected genes are now stored in the featureSubset alternative experiment used by Celda; the main dataset keeps all genes, as with the None option, instead of being reduced to them.
Fixed starting the Shiny app with data, singleCellTK(inSCE = sce), which failed for any object without alternative experiments.
The SoupX help page in the app describes the new clustering method.

singleCellTK 2.23.3 (2026-10-07)

runModelGeneVar() now models gene variances with scrapper::modelGeneVariances() instead of the deprecated scran::modelGeneVar(). Means and total variances are unchanged, but the biological component comes from a different trend fit, so the ranking of highly variable genes changes (about 90-97% of the top 500 genes are the same on the test datasets). Results that depend on modelGeneVar HVGs, such as PCA, UMAP, and clusters, can change.
runSoupX() clusters cells (when no clusters are supplied) with the steps of scran::quickCluster(), computed with scrapper and bluster. The automatic clusters, and therefore SoupX's contamination estimates, can change slightly.

singleCellTK 2.23.2 (2026-10-07)

Replaced functions deprecated in scuttle 1.22 and scran 1.40. Results are unchanged (identical, or within floating-point error):
- runClusterSummaryMetrics(), plotSCEHeatmap(), and the TSCAN functions aggregate with scrapper or base R instead of scuttle::aggregateAcrossCells().
- scaterlogNormCounts() normalizes with scrapper::normalizeRnaCounts.se().
- runPerCellQC() and sampleSummaryStats() compute the per-cell QC metrics with scrapper::computeRnaQcMetrics() and singleCellTK code (for the top-N, gene-set, and altExp columns) instead of scater::addPerCellQC(). This is about 4 times faster on large data. runPerCellQC() uses BPPARAM to set the number of threads. The metadata packageVersion recorded by runPerCellQC() is now singleCellTK's.
- runScranSNN() builds graphs with bluster::makeSNNGraph().
- runWilcox() uses an internal Wilcoxon rank-sum test, the same test scran::pairwiseWilcox() performed.
New dependencies: scrapper, bluster, and BiocSingular. scuttle is no longer imported directly.
runMNNCorrect() hides deprecation warnings raised inside batchelor::mnnCorrect(), which still calls deprecated scuttle functions. Other warnings are unchanged.
Fixed plotSCEHeatmap() with aggregateRow, which always failed.
Fixed runScranSNN() with useAltExp and altExpRedDim, which always failed.

singleCellTK 2.23.1 (2026-10-06)

runDecontX() now calls the decontX package directly instead of going through celda. Since celda 1.23.0, celda::decontX() requires decontX to be installed, which made runDecontX() and runCellQC() fail where it was not. decontX is now imported, and the run metadata records the decontX version. Results are unchanged. Note that decontX 1.11 initializes its cell clusters with scrapper, so contamination estimates can differ slightly from earlier Bioconductor releases.
Moved scMerge from Imports to Suggests. It is only used by runSCMerge(), and it cannot currently be installed on all platforms (an unrelated toolchain problem in one of its dependencies), which blocked installing singleCellTK there entirely. runSCMerge() now stops with an informative message if scMerge is not installed.
Fixed runHarmony() with harmony >= 1.0.0, where HarmonyMatrix() was removed. The function now detects the installed harmony version and calls RunHarmony() on 1.0.0 and above, while keeping the original HarmonyMatrix() code path for older installations. With harmony >= 1.0.0 the PCA for a full-size assay is computed by singleCellTK, since harmony no longer performs it, using all features of the selected assay with scaling so that it matches the PCA the legacy interface performed internally. Corrected embeddings are not expected to match between harmony versions. After upgrading harmony, the legacy tuning arguments (tau, block.size, max.iter.cluster, epsilon.cluster, epsilon.harmony) are no longer accepted through '...'; harmony raises an error for them, and they must be passed through .options = harmony::harmony_options(...) instead.
Changed runScDblFinder() defaults for nNeighbors and simDoublets to NULL, so scDblFinder chooses these values itself

singleCellTK 2.21.1

Updated depreciated parameters in Seurat function calls

singleCellTK 2.20.0 (2025-10-29)

Updated version to match Bioconductor 3.22

singleCellTK 2.18.1 (2025-07-01)

Updated enrichR examples

singleCellTK 2.18.0 (2025-04-15)

Updated call to msigdbr to work with newer version
Updated version to match Bioconductor 3.21
Fixed reproducibility with scDblFinder

singleCellTK 2.16.1 (2024-02-12)

Fixed decontX/soupX webapp error
Fixed error in reportCellQC
Fixed additional bugs with Seurat V5 integration

singleCellTK 2.16.0 (2024-10-29)

Updated version to match Bioconductor 3.20

singleCellTK 2.14.0 (2024-05-03)

Updated version to match Bioconductor 3.19
Update runGSVA function to work with newer GSVA package

singleCellTK 2.12.2 (2024-01-28)

Added support for Seurat V5

singleCellTK 2.12.1 (2024-01-10)

Updates to documentation
Fixes to runTSCAN and plotSeurat Genes
Added support for flat file import into SCTK-QC
Fixed directory issue in importCellRanger
Added Bubble plot to Shiny GUI
Updated Dockerfile

singleCellTK 2.12.0 (2023-10-24)

Updated version to match Bioconductor 3.18

singleCellTK 2.10.1 (2023-07-26)

Added function for bubble plot
In SCTK-QC pipeline, added support for batch processing multiple inputs
In SCTK-QC pipeline, added support for importing and exporting AnnData objects
In SCTK-QC pipeline, fixed a bug causing YAML output files to be empty
Update the SCTK-QC tutorial
Fixed bug in combineSCE causing it to create multiple copies of row or column data

singleCellTK 2.10.0 (2023-04-25)

Updated version to match Bioconductor 3.17

singleCellTK 2.8.1 (2023-03-10)

Added scanpy wrapper functions for use from console
Added scanpy UI curated workflow
Integrated scanpy to a la carte workflow
Fixed a bug in importing fluidigm dataset
Updated downloading features in the Shiny app
Added error checking around Enrichr functions
Minor tweaks to plot defaults

singleCellTK 2.8.0 (2022-12-19)

Updated version to match Bioconductor version

singleCellTK 2.7.3 (2022-10-25)

Fixed bugs related to dependency updates

singleCellTK 2.7.2 (2022-10-19)

Deprecated findMarkerDiffExp(), findMarkerTopTable() and plotMarkerDiffExp(), which are replaced by runFindMarker(), getFindMarkerTopTable() and plotFindMarkerHeatmap(), respectively
Added useReducedDim, detectThresh arguments for find marker functions
Deprecated getUMAP() and getTSNE(), which are replaced by runUMAP() and runTSNE(), respectively
Added runQuickUMAP() and runQuickTSNE() functions which directly compute the proper embedding from raw counts matrices with a simplified argument set
Added arguments aggregateRow and aggregateCol to plotSCEHeatmap()
Updated output metadata structure of QC functions, as well as combineSCE() which merges the new structure properly
Refined batch correction function set
Fixed bugs related to UI and console functions

singleCellTK 2.7.1 (2022-06-29)

Refactored scaling related parts of the workflow, including variable feature detection, dimension reduction and 2D embedding
Redesigned UI landing page and UI running prompt
Added marker table module across UI
Added more unit tests
Fixed bugs in TSCAN UI
Other minor bug fixes

singleCellTK 2.6.0 (2022-04-28)

Updated version to match Bioconductor

singleCellTK 2.5.2 (2022-04-23)

Added Seurat report functions
Added TSCAN trajectory analysis functions
Refactored EnrichR wrapper function (runEnrichR)
Added new cut-offs for DE functions
Other refactors and bug fixes

singleCellTK 2.5.1 (2022-03-31)

Added SoupX method for decontamination (runSoupX)
Added useReducedDim parameter for DE analysis and Heatmap
Added Differential Abundance section to the tutorials
Fixed Mitochondrial gene list
Other refactors and bug fixes

singleCellTK 2.4.1 (2021-12-22)

Added new function for DEG volcano plot (plotDEGVolcano)
Added new function for plotting pathway scores (plotPathway)
Added Pathway Analysis section to the tutorials
Added seed parameter to several functions and UI for reproducibility
Updated R console and GUI tutorials to match each other
Fixed console logging in the GUI

singleCellTK 2.4.0 (2021-10-27)

Updated version to match Bioconductor

singleCellTK 2.3.2 (2021-10-24)

Added summary table into the cellQC report
Improved formatting in QC report
Added functions getDEGTopTable() & plotBatchCorrCompare()
Other refactors and bug fixes

singleCellTK 2.3.1 (2021-10-15)

Several bug fixes

singleCellTK 2.2.2 (2021-10-10)

Several enhancements, refactors, and bug fixes to the UI
Refactor documentation and pkgdown site
Added tutorials for R console analysis
Updates to the UMAP generation in the SCTK-QC pipeline
Addition of VAM to Pathway prediction tab
Bug fix to the mitochondrial gene set functions

singleCellTK 2.1.3 (2021-05-14)

Added diffAbundanceFET and plotClusterAbundance function
Linked Shiny UI help buttons to new online help pages
Several bug fixes

singleCellTK 2.0.2 (2021-05-08)

Expanded convertSCEtoSeurat() function to copy additional data
Updated and merged pkgdown docs
Added HTML reports for Seurat curated workflow
Refactor of Normalization UI
Added generic wrapper function for dimensionality reduction
Added tagging system for matrix type
Several bug fixes
Added missing documentation
Added wrapper functions for normalization, dimensionality reduction and feature selection
Added function seuratReport() to generate a seurat report from input SCE object

singleCellTK 2.0.1 (2021-01-07)

Added cell type labeling functional, wrapping SingleR method
Added cell type labeling UI under differential expression tab
Added marker identification in Seurat workflow

singleCellTK 2.0.0 (2020-10-16)

Added quality control (empty droplet detection, doublet detection, etc) functionality
Ability to import data from varying preprocessing tools
Ability to export SingleCellExperiment object as varying file types (flat file, Python anndata)
Added functions for visualization of data
New CellViewer functionality in UI
Improvements to differential expression, now includes DESeq2, limma, ANOVA
Incorporates Seurat workflow

singleCellTK 1.1.26 (2018-10-23)

New UI design for the Differential Expression tab.
New UI design for the Data Summary & Filtering tab.
Support for additional assay modification including log transforming any assay and renaming assays.
New function visPlot for creating scatterplots, boxplots, heatmaps, and barplots for custom gene sets.
The Downsample tab now works on a generic counts matrix
You can upload a SCtkExperiment object or a SingleCellExperiment object saved in an RDS file on the Upload tab.
Differential Expression results can now be saved in the rowData of the object and loaded for later analysis.
Improved ability to save a biomarker based on user options.
The Differential Expression plot is not automatically created, for more user control with large datasets.

singleCellTK 1.1.3

Improvements to plotting, change text size and hide labels in gsva plots.
MAST violin and linear model plots are now more square when plotting less than 49 facets.
Changed y axis label in plotBatchVariance to "Percent Explained Variation"

singleCellTK 1.1.2

Ability to hide version number in the SCTK GUI.

singleCellTK 1.1.1

Fixed a bug that would cause the diffex color bar to not display when special characters were in the annotation.

singleCellTK 0.99.3

Consistent use of camel case throughout package

singleCellTK 0.6.3

Additional links to help documentation
Example matrices on upload page.

singleCellTK 0.4.7

Ability to download/reupload annotation data frame and convert annotations to factors/numerics

singleCellTK 0.4.5

Documentation updates to fix NOTES and pass BiocCheck