NEWS

---

Seurat 2020-12-14

Added

• Titles added to DimPlot when specifying group.by parameter
• keep.scale parameter added to FeaturePlot to control scaling across multiple features and/or splits.

Changes

• Same deprecated in favor of base::identity
• Fix in DietSeurat to work with specialized Assay objects
• Fix p-value return when using the ape implementation of Moran's I
• Fix bug in FindMarkers when using MAST with a latent variable
• Updates to Key<-.DimReduc that allow handling of empty reduction column names
• Allow setting ctrl in CellCycleScoring
• Modify subset.Seurat to allow specialized Assay subsetting methods
• Fix image selection in interactive spatial plots
• Update Rcpp functions with export(rng=FALSE) to avoid potential future warnings
• Fix RenameCells bug for integrated SCT assays
• Fix highlight order with proper factor levels when using SetHighlight in plots
• Small change in CellRanger version detection logic of h5 file to improve robustness to outside tools.
• do.cpp deprecated and will default to true

Seurat 5.6.0 (2026-10-03)

Additions

• Added setThreads and getThreads to control threading in supported C++ kernels and uwot UMAP backends
• Updated NormalizeData to use threaded sparse log-normalization and reduce matrix-copying overhead
• Updated FindVariableFeatures to share a sparse VST implementation across assay versions with threaded standardized-variance calculations
• Updated ScaleData to center, scale, and clip default inputs in a single threaded C++ call
• Updated SCTransform to use threaded C++ calculations for supported Pearson residuals and corrected counts
• Updated the v5 SCTransform workflow to avoid intermediate assay objects and defer default residual calculation until feature selection is complete
• Updated RunPCA to use a threaded dense Gram-matrix backend by default and reuse feature variances
• Updated FindNeighbors to use threaded C++ Annoy searches and SNN graph construction
• Updated FindClusters to reuse graph preparation across resolutions and parallelize modularity-optimization random starts
• Updated RunUMAP to use the Seurat thread setting for uwot, uwot2, and model projection
• Updated FindMarkers to reuse matrix subsets for fold-change calculations and process sparse inputs efficiently
• Updated FindAllMarkers to test supported Wilcoxon comparisons across all identities in one presto call, and compute fold changes and detection percentages in bulk
• Updated IntegrateLayers with RPCAIntegration to reduce per-layer setup overhead and use threaded integration calculations
• Updated IntegrateLayers with CCAIntegration to use threaded anchor scoring and integration-weight calculations
• Added svd.method parameter to CCAIntegration for use through IntegrateLayers
• Updated large dense CCA comparisons to use an implicit RSpectra backend (svd.method = "rspectra") and avoid materializing the cell-by-cell cross-product matrix
• Updated CCAIntegration to skip gene loadings when anchor filtering is disabled

Fixes

• Fixed AddModuleScore (and CellCycleScoring) on v5 objects with on-disk (e.g. BPCells) assays, where each layer was fully densified to an in-memory dgCMatrix before scoring; scoring now operates directly on the on-disk matrix (#10448)
• Fixed GetResidual() to correctly handle multi-model SCT assays with partial feature overlap (#10541)
• Fixed bug in PercentageFeatureSet where layer data was incorrectly retrieved prior to finding features with requested pattern (#10438)
• Updated as.SingleCellExperiment to address conversion case where an object has both original and sketched assay / reductions (differing numbers of cells) (#10419)
• Fixed bugs in behavior of RidgePlot parameters fill.by, y.max, and same.y.lims (#10424, #10517)
• Fixed naming of combined p-value column in output of FindConservedMarkers when a non-default meta.method is specified (#10429)
• Updated minimum required uwot version to 0.2.1 (#10447)
• Registered Radius.VisiumV2 as an S3 method to fix Radius returning NULL for VisiumV2 images (#10454)
• Fixed FindSpatiallyVariableFeatures() to correctly match FOV coordinates to cells in the requested assay and handle various edge cases, including fewer than two matched cells, a single varying feature, or no varying requested features (#10504)
• Updated argument handling in FindSpatiallyVariableFeatures() by resolving selection.method with match.arg() and restoring the FindSpatiallyVariableFeatures.Assay() default nfeatures value to 2000 (#10504)
• Fixed RunMarkVario() to return one named mark variogram result per feature for single-feature inputs and parallel execution chunks (#10505)
• Updated IntegrateLayers() to throw an error message when less than two groups/layers are provided (#10396)
• Fixed recorrection in SCTransform after subsetting so FindMarkers() works on subsetted SCT assays with observed median UMI lower than the correction depth stored in every SCT model (#10509)
• Fixed FindNeighbors to correctly convert Annoy angular distances to cosine distances
• Fixed RunPCA on v5 assays to exclude unavailable requested features and cap the number of PCs by the available features and cells
• Fixed SCTransform residual retrieval to reuse cached features and respect an explicitly supplied clip.range
• Fixed SCTransform on v5 assays to preserve features selected by variable.features.rv.th when variable.features.n = NULL, including across multiple layers
• Fixed reference-based RPCA/CCA anchor finding to exclude query-query comparisons regardless of verbose

Seurat 5.5.1 (2026-06-26)

Additions

• Added uwot.init parameter to RunUMAP to optionally customize method of UMAP initialization (#10386)

Fixes

• Fixed bug in PrepSCTFindMarkers in the multi-model case that produced zero counts for some valid genes due to an incorrect filter for theta (#10410)
• Fixed issue with FindClusters not setting metadata properly when a vector of resolutions is passed to resolution and a custom cluster name is specified (#10385)
• Updated logic in Parenting to extract function names from the call stack efficiently, avoiding slowdowns when functions are run with do.call (#10161)
• Fixed bug in PrepDR5 affecting RunPCA on v5 objects, where supplied features could be incorrectly dropped when non-supplied features had zero variance (#10398)
• Updated LabelClusters to use fill argument for label background color in geom_label* when specified, instead of the default per-cluster color (#10404)

Seurat 5.5.0 (2026-04-22)

Additions

• Added support for vars.to.regress and latent.data in ScaleData.IterableMatrix (BPCells) (#10300)
• Added warning to SCTransform to notify that do.correct.umi is hard set to F when a reference SCT model is specified (#10329, #10366)
• Added warning to FindMarkers/FindAllMarkers when running on an IterableMatrix (BPCells) with column-major storage order (#10365)
• Added plot_type parameter to ElbowPlot to switch between default (PCs x standard deviation) and two new options:
  • PCs x % variance explained (#8929)
  • PCs x cumulative % variance explained (#10369)
• Added legend.position parameter DimHeatmap to allow repositioning of legend when combine = T (#9813)
• Added "selected cells" indicator in InteractiveSpatialPlot (#10254)
• Added stroke.alpha parameter to SpatialPlot to be able to control transparency for fill and stroke separately (#10345)
• Added support for unnamed cols vectors in SpatialPlot (#10298)
• Added assay parameter to FeaturePlot to specify assay to pull features from (#10349)

Fixes

• Fixed issue with UMAP version parsing in RunUMAP (#10246)
• Updated HTODemux to handle HTO assays containing a subset of cells (#10362)
• Updated Rcpp error handling (#10317, #10353)
• Fixed issue with NaN values being propagated in PrepSCTFindMarkers (#10361)
• Fixed umi.assay labeling for model list in SCTransform to match the assay specified (#10330)
• Fixed factor-level ordering for metadata columns resulting from FindClusters (#10253)
• Fixed error in FindSpatiallyVariableFeatures from use of slot in function cascade (#10285)
• Fixed classification of gRNA with less than 5 cells in RunMixscape for proper label as "NP" (#10299)
• Fixed issue with checking object scope in as.Seurat.SingleCellExperiment (#10339)
• Fixed SketchData to pass the provided argument for verbose to LeverageScore (#10145)
• Fixed FindConservedMarkers to pass the provided argument for min.cells.group to FindMarkers (#8910)
• Updated IntegrateEmbeddings to respect dims.to.integrate when reference is specified (#10172)
• Fixed list structure of neighbors in FindNeighbors when SNN is not computed (#10364)
• Updated DimHeatmap to add PC labels as individual titles (#9813)
• Updated layer-building logic in Single/MultiExIPlot for ggplot2 v4+ compatibility (#10333)
• Fixed order = T in rasterized Dim/FeaturePlots via ggrastr-based rasterization (#10307, #10344)
• Updated SpatialPlot coordinate system check when plotting VisiumV1 images (#10267)
• Updated SpatialPlot for Visium data to speed up plot construction (#10345)
• Fixed interactive functionality of LinkedDim/LinkedFeaturePlot and interactive mode in SpatialPlot (#10368)
• Updated coordinate systems used for plot display in InteractiveSpatialPlot (#10254)
• Fixed coordinate ranges when spatial plots are generated with crop = F (#10278, #10284)

Seurat 5.4.0 (2025-12-14)

Additions

• Added support for 10x Space Ranger 4.0 outputs (Visium data with segmentations)
  • Updated data loading functions Load10X_Spatial, Read10X_Image,
  • Updated GetTissueCoordinates.VisiumV2
  • Updated spatial visualization functions SpatialPlot, SingleSpatialPlot, SpatialDimPlot, SpatialFeaturePlot
  • Added helper functions Read10X_Segmentations, Read10X_HD_GeoJson, Format10X_GeoJson_CellID
• Added function InteractiveSpatialPlot to allow users to interactively lasso-select cells from a spatial Seurat object (Visium, SlideSeq, or Vizgen data)

Fixes

• Updated loading & visualization functions (see above for list) for Visium objects -- see #10125 for details
  • For binned Visium data, x now correctly corresponds to imagecol from tissue positions; y now correctly corresponds to imagerow from tissue positions; these are now consistent with 10X's coordinate system (with the origin being the top left).
• Reverted #10062 in favor of fetching both grouping variables and dimensionality reduction embeddings with FetchData as previously; added warning to alert users when column names of metadata and dimensionality reduction embeddings conflict
• Added color retrieval logic in LabelClusters for consistent cluster label coloring (#10198)

Seurat 5.3.1 (2025-10-29)

Changes

Additions

• Added option to use cluster_leiden from igraph when running Leiden clustering (#9931)
• Added option to use umap2 from uwot when running UMAP (#9918)
• Added option to use approx_pow parameter from uwot when running UMAP (#9449)
• Updated LoadXenium to add support for Xenium protein data (#10024)
• Added option to specify stroke size in FeaturePlot (#10053)
• Added label.size.cutoff parameter to DimPlot to allow users to label only clusters above a certain size (#10092)

Fixes

• Fixed deprecated calls to Seurat functions GetAssayData and PackageCheck (#9924), (#10102)
• Fixed errors in RunAzimuth relating to FindTransferAnchors (#9924)
• Fixed errors in LeverageScore relating to handling of (variable) features (#9924)
• Updated PseudobulkExpression to acccount for the case that a category passed to group.by has only one value (#10092)
• Updated VlnPlot to use assay passed to the function if specified (#10092)
• Updated FeatureScatter to allow gene names with parentheses to be plotted (#10092)
• Updated SingleRasterMap to store object metadata correctly (#10092)
• Updated FindClusters to order cluster factor levels correctly when assigning a custom cluster name (#10092)
• Updated FindConservedMarkers to remove NA values (#9917)
• Updated visualization functions to avoid ggplot2 guides, aes_string, and facet_grid deprecation warnings (#9409, #10116)
• Fixed DimPlot bug where metadata columns named 'PC_1', 'UMAP_1' etc override reduction embeddings (#10062)

Seurat 5.3.0 (2025-04-23)

Changes

• Fixed PseudobulkExpression to forward relevant arguments to NormalizeData (#9840)
• Fixed bugs in FindSpatiallyVariableFeatures; deprecated the slot parameter in favor of layer (#9836)
• Extended FindTransferAnchors's reference argument to accept SCT inputs containing more than one SCT model; in this case, the reference model that was fit against the largest number of cells is used (#9833)
• Extended FindTransferAnchors's query argument to accept multi-layer inputs; updated MappingScore to support multi-layer query inputs (#9832)
• Updated LeverageScore.default to convert BPCells::IterableMatrix inputs with less than 7500 cells into a sparse matrix before performing the calculation (#9831)
• Dropped VariableFeatures setter from SketchData (#9830)
• Extended Cells.SCTAssay's layer argument accept slot names: "counts", "data", "scale.data"; enabled compatibility with SketchData/LeverageScore(#9830)
• Updated SCTransform.StdAssay to simplify and speed up the method (#9828)
• Updated AddModuleScore to support multi-layer inputs (#9826)
• Fixed PseudobulkExpression to work with Seurat inputs containing more than one assay (9824)
• Added RunICA.StdAssay (#9825)
• Updated HarmonyIntegration to call harmony::RunHarmony in favor of the deprecated harmony::HarmonyMatrix (#9789)
• Added raster.dpi parameter to VlnPlot (#9665)
• Fixed SpatialDimPlot(..., interactive = TRUE) to support SlideSeq and STARmap input types; added ScaleFactors.SlideSeq & ScaleFactors.STARmap generics (#9691)
• Fixed PercentageFeatureSet so that the assay parameter is always respected; fixed PercentageFeatureSet to raise a warning if any features are absent in the specified assay instead of throwing an error (#9686)
• Fixed GroupCorrelation and GroupCorrelationPlot to be compatible with SeuratObject >= 5.0.0 (#9625)

Seurat 5.2.1 (2025-01-24)

Changes

• Fixed test_find_clusters.R to accommodate variability in label assignments given by FindClusters across different systems (#9641)

Seurat 5.2.0 (2025-01-13)

Changes

• Added group.by parameter to FindAllMarkers, allowing users to regroup their data using a non-default identity class prior to performing differential expression (#9550)
• Added image.type parameter to Read10X_Image enabling VisiumV1 instances to be populated instead of instances of the default VisiumV2 class (#9556)
• Fixed IntegrateLayers to respect the dims.to.integrate parameter
• Added stroke.size parameter to DimPlot (#8180)
• Updated RunLeiden to use the leidenbase package instead of leiden; deprecated the method parameter for RunLeiden and FindClusters; updated RunLeiden to reset random.seed to 1 if the value is 0 or less (#6792)
• Updated RunPCA to use the BPCells-provided SVD solver on BPCells matrices; updated JackStraw to support BPCells matrices (#8271)
• Fixed RunPCA to avoid converting BPCells matrices into dense matrices - significantly reduces the function's memory usage when running on BPCells matrices (#8966)
• Updated RunSLSI to support BPCells matrices
• Fixed DietSeurat so that the layers parameter is respected (#8197)
• Updated RunUMAP to support umap-learn version >= 0.5.0 (#9559)
• Updated LoadXenium and ReadXenium to accommodate the output from XOA v3.0; updated LoadXenium to provide more fine-grained control over the datatypes parsed in, including nucleus segmentation masks, segmentation methods, and other experimental metadata; updated ReadXenium to load cell_feature_matrix.h5 when present in favor of the MEX format files; updated ReadXenium to load .parquet files using arrow instead of .csv.gz files to support XOA 3.0 (#8604)
• Fixed LoadXenium to accommodate datasets without "Blank Codeword" or "Unassigned Codeword" matrices(#9135)
• Fixed ReadXenium to properly parse multiple molecular outputs at once (#8265)
• Added features parameter to LeverageScore and SketchData
• Updated SketchData's ncells parameter to accept integer vector

Seurat 5.1.0 (2024-05-10)

Changes

• Updated interactive SpatialDimPlots to display spatial coordinates
• Updated BuildClusterTree to PseudobulkExpression(..., method = "aggregate") instead of AverageExpression
• Fixed SpatialPlot to properly handle images with shared cells
• Added cluster.name parameter to BuildNicheAssay
• Added VisiumV2 class, inherits SeuratObject::FOV, returned by Load10X_Spatial
• Updated Load10X_Spatial to support loading Visium HD datasets - added bin.size parameter
• Updated Read10X_Coordinates to handle new tissue_coordinates.parquet files
• Added shape parameter to SpatialPlot and wrappers: SpatialFeaturePlot and SpatialDimPlot
• Added image.scale parameter to SpatialPlot and related functions: SpatialFeaturePlot/ISpatialFeaturePlot and SpatialDimPlot/ISpatialDimPlot
• Restored image.name parameter to Read10X_Image
• Added scale parameter to Radius.VisiumV1
• Fixed calculation of spot.radius attribute for VisiumV1 instance returned by Read10X_Image — replace scale.factors$fiducial with scale.factors$spot
• Added Read10X_Coordinates and Read10X_ScaleFactors
• Fixed SpatialPlot to properly handle mutli-assay Seurat instances

Seurat 5.0.3 (2024-03-18)

Changes

• Fixed PercentAbove to discount null values (#8412)
• Added log parameter to FeatureScatter
• Fixed handling of clip.range for SCTransform when ncells is less than the size of the passed dataset

Seurat 5.0.2 (2024-02-28)

Changes

• Fixed AverageExpression to handle features vector when layer="scale.data" and return.seurat=TRUE
• Added fc.slot parameter to FindMarkers
• Fixed SCTransform to handle vars.to.regress (#8148) and (#8349)
• Fixed SCTransform to handle fetching residuals (#8185)

Seurat 5.0.1 (2023-11-17)

Changes

• Fixed SCTransform.StdAssay to pass extra arguments to sctransform::vst(). Fixes #875
• Fixed PercentageFeatureSet Layer calling (#8009)
• Fixed cell highlighting (#7914)
• Updated marker sorting to be by p-value with ties broken by absolute difference in percent expression
• Fixed issue with replicated barcodes in MappingScore (#7922)
• Improved PseudobulkExpression by adding 'g' to cell names that started with numeric values
• Improved PseudobulkExpression by adding each variable specified in group.by as columns in the object metadata when return.seurat=TRUE
• Fixed DimPlot and FeatureScatter which were breaking when using the split.by argument with a variable that contained NAs

Seurat 5.0.0 (2023-11-04)

Added

• Add BridgeCellsRepresentation to construct a dictionary representation for each unimodal dataset.
• Add BuildNicheAssay to construct a new assay where each feature is a cell label. The values represent the sum of a particular cell label neighboring a given cell.
• Add CalcDispersion to calculate the dispersion of features.
• Add CCAIntegration to perform Seurat-CCA Integration.
• Add CountSketch to generate a CountSketch random matrix.
• Add CreateCategoryMatrix to create a one-hot matrix for a given label.
• Add DISP to find variable features based on dispersion.
• Add FastRPCAIntegration as a convenience wrapper function around the following three functions that are often run together when performing integration.
• Add FetchResiduals_reference as a temporary function to get residuals from the reference.
• Add FetchResiduals to call sctransform::get_residuals.
• Add FetchResidualSCTModel to calculate Pearson residuals of features not in the scale.data.
• Add FindBridgeAnchor to find bridge anchors between two unimodal datasets.
• Add FindBridgeIntegrationAnchors to find a set of anchors for integration between unimodal query and the other unimodal reference using a pre-computed BridgeReferenceSet.
• Add FindBridgeTransferAnchors to find a set of anchors for label transfer between unimodal query and the other unimodal reference using a pre-computed BridgeReferenceSet.
• Add GaussianSketch to perform Gaussian sketching.
• Add HarmonyIntegration to perform Harmony integration.
• Add IntegrateLayers to integrate layers in an assay object.
• Add JointPCAIntegration to perform Seurat-Joint PCA Integration.
• Add LeverageScore to compute the leverage scores for a given object.
• Add LoadCurioSeeker to load Curio Seeker data.
• Add MVP to find variable features based on mean.var.plot.
• Add NNtoGraph to convert the Neighbor class to an asymmetrical Graph class.
• Add PrepareBridgeReference to preprocess the multi-omic bridge and unimodal reference datasets into an extended reference.
• Add ProjectCellEmbeddings to project query data onto the reference dimensional reduction.
• Add ProjectData to project high-dimensional single-cell RNA expression data from a full dataset onto the lower-dimensional embedding of the sketch of the dataset.
• Add ProjectDimReduc to project query data to reference dimensional reduction.
• Add ProjectIntegration to integrate embeddings from the integrated sketched.assay.
• Add PseudobulkExpression to normalize the count data present in a given assay.
• Add Read10X_probe_metadata to read the probe metadata from a 10x Genomics probe barcode matrix file in HDF5 format.
• Add RPCAIntegration to perform Seurat-RPCA Integration.
• Add RunGraphLaplacian to run a graph Laplacian dimensionality reduction.
• Add SelectIntegrationFeatures5 to select integration features for v5 assays.
• Add SelectSCTIntegrationFeatures to select SCT integration features.
• Add SketchData to use sketching methods to downsample high-dimensional single-cell RNA expression data for help with scalability for large datasets.
• Add TransferSketchLabels to transfer cell type labels from a sketched dataset to a full dataset based on the similarities in the lower-dimensional space.
• Add UnSketchEmbeddings to transfer embeddings from sketched cells to the full data.
• Add VST to apply a variance stabilizing transformation for selection of variable features.

Changes

• Change FindTransferAnchors so that anchor filtering is not performed by default
• Change merge so that layers will be added to a single Seurat object instead of combining raw count matrices
• Deprecate slot parameter in favor of layers in accessor and set methods

Seurat 4.4.0 (2023-09-28)

Added

• Add parallelization support with speed improvements for PrepSCTFindMarkers
• Fix bug in LoadNanostring(#7566)

Changes

• Fix bug in as.Seurat.SingleCellExperiment() (#6692)
• Support for Visium probe information introduced in Spaceranger 2.1 (#7141)
• Add LoadCurioSeeker to load sequencing-based spatial datasets generated using the Curio Seeker
• Fix fold change calculation for assays (#7095)
• Fix pt.size bug when rasterization is set to true (#7379)
• Fix FoldChange and FindMarkers to support all normalization approaches (#7115,#7110,#7095,#6976,#6654,#6701,#6773, #7107)
• Fix for handling newer ParseBio formats in ReadParseBio (#7565)
• Fix for handling rasterization by default (#7842)
• Fix bug in ReadMtx() to add back missing parameters
• Fix SCTransform() for V5 assays to retain gene attributes (#7557)
• Fix LeverageScore() for objects with few features (#7650

Seurat 4.3.0 (2022-11-18)

Added

• Add support for imaging-based spatial datasets

Changes

• Fix bug in FindMarkers() when run post Integration/Transfer (#6856)

Seurat 4.2.1 (2022-11-07)

Changes

• Replaced import from spatstat.core with spatstat.explore
• Fix bug in FindMarkers() when using SCT assay (#6856)

Seurat 4.2.0 (2022-09-21)

Changes

• Fix legend color in DoHeatmap() (#5783)
• Fix bug in ScaleData() when regressing out one gene (#5970)
• Fix name pulling in PlotPerturbScore() (#6081)
• Support spaceranger 2.0 (#6208)
• Fix bug in SpatialDimPlot() when using group.by (#6179)
• Add add.noise parameter in VlnPlot() (#5756)
• Fix uwot model backwards compatibility (#6345)
• Allow pseudocount.use in differential expression functions to be set at the Assay level

Seurat 4.1.1 (2022-05-02)

Changes

• Fix giveCsparse related warnings in Read10X_h5
• Fix ident labeling for SpatialPlot (#5774)
• Fix ReadMtx on Windows (#5687)
• Fix VlnPlot to switch on rasterization only when required (#5846)
• Fix ncol behavior in SpatialPlot (#5774)
• Set jitter to FALSE in FeatureScatter (#5876)
• Update Cells methods to new signature (x, ...)
• Replace use of default.stringsAsFactors() with getOption("stringsAsFactors")

Seurat 4.1.0 (2022-01-14)

Added

• Add raster.dpi parameter to DimPlot/FeaturePlot to optionally rasterize individual points (#5392)
• Add support for sctransform v2, differential expression with SCT assay

Changes

• Update ReadParseBio to support split-pipe 0.9.6p (#5446)
• Fixes for MAST differential expression (#5441)
• Fix scaling options when using split.by in FeaturePlot() (#5243)

Seurat 4.0.6 (2021-12-16)

Added

• Implement supervised LSI

Changes

• Add raster parameter to VlnPlot to optionally rasterize individual points (#5076)
• Add min.cells.group parameter to FindConservedMarkers (#5079)
• Set do.center to FALSE for lsiproject in FindTransferAnchors
• Fix error message in ReadMtx() (#5158)
• Add label.color parameter to FeaturePlot (#5314)
• Fix issues in ProjectUMAP (#5257, #5104, #5373)

Seurat 4.0.5 (2021-10-17)

Changes

• Update documentation for to.upper parameter in Load10X_Spatial() (#4576)
• Update concept tags for RunSPCA() (#4978)
• Conditionally run tests/packages that use suggested packages (#5160)
• Set random state in RunUMAP() when using the umap-learn method (#5194)

Seurat 4.0.4 (2021-08-20)

Added

• Add reduction parameter to BuildClusterTree() (#4598)
• Add DensMAP option to RunUMAP() (#4630)
• Add image parameter to Load10X_Spatial() and image.name parameter to Read10X_Image() (#4641)
• Add ReadSTARsolo() function to read output from STARsolo
• Add densify parameter to FindMarkers()
• Add ReadParsebio() function to read output from Parse Biosciences
• Add the image.alpha parameter to SpatialDimPlot() and SpatialFeaturePlot()
• Add support for the correlation metric in RunUMAP (#4972)

Changes

• Warn and continue rather than erroring if not all features are available in FindSpatiallyVariableFeatures() (#4611)
• Bug fix for SCT-based integration in selecting proper reference model (#4355)
• Bug fix for reading from absolute paths in ReadMtx (#4723)
• Bug fix in SingleCellExperiment conversion (#4633)
• Bug fix in FindVariableFeatures() when using selection.method = "mvp" and binning.method = "equal_frequency" (#4712)
• Bug fix in DoHeatmap() to remove random characters from plot legend(#4660)
• Fix cell renaming in RunCCA()
• Fix issue in SingleCellExperiment conversion where the mainExp would not be set properly
• Fix for default dispersion info displayed in VariableFeaturePlot()

Seurat 4.0.3 (2021-06-10)

Added

• Add jitter parameter to FeatureScatter()

Changes

• Fix issues with as.SingleCellExperiment.Seurat() for the latest verion of SingleCellExperiment (#4532)
• Ensure proper reference.reduction is used in MapQuery()
• Fix to UpdateSymbolList(), no longer searches aliases and exposes the search.types parameter in GeneSymbolThesarus() (#4545)
• Transfer scale.data slot as well when converting with as.SingleCellExperiment.Seurat()
• Enable alpha parameter for SpatialDimPlot()
• Fix as.SingleCellExperiment.Seurat() conversion for atypical reducedDim components

Seurat 4.0.2 (2021-05-20)

Added

• New AddAzimuthScores() and AddAzimuthResults() functions
• Add shuffle parameter to FeatureScatter() (#4280)
• Add lsiproject and rpca options for FindTransferAnchors()
• Add rlsi option for FindIntegrationAnchors()

Changes

• Preserve feature metadata when converting from SingleCellExperiment to SeuratObject class (#4205)
• Preserve multiple assays when converting from SingleCellExperiment to SeuratObject class (#3764)
• Fix passing of score.thresh parameter in ScoreJackStraw() (#4268)
• Fix FC calculation in FindMarkers() non-log transformed data.
• Add umap-learn version >= 0.5.0 compatibility for RunUMAP()
• Fix DotPlot to use log1p when scale=False (#4298)
• Fix split and shuffled DimPlot
• Disallow NULL or another length 0 vector for ident.1 in FindMarkers()
• Fix range shift when labeling clusters on a GeomSpatial plot
• Fix SpatialPlot distortion for non-square images.
• Fix future-related warnings in FindIntegrationAnchors()
• Fix fc.name parameter in FindMarkers() (#4474)
• Deprecate group.by parameter in PlotPerturbScore() in favor of mixscape.class.

Seurat 4.0.1 (2021-03-18)

Added

• Add direction option to PlotClusterTree()
• Add cols parameter to JackStrawPlot()
• Add ReadMtx() to read local and remote mtx files with associated cell and feature name files

Changes

• Equality added to differential expression thresholds in FindMarkers (e.g, >= logfc.threshold rather than >)
• Read10X() now prepends dataset number for first dataset when reading multiple datasets
• Bug fix for subset.AnchorSet()
• Bug fix for fold change values in FindMarkers() when setting a different pseudocount (#4111)
• Bug fix for RunLDA() related to proper passing of assay parameter.
• When using order=TRUE in SingleDimPlot(), print NA points under all others.
• Remove default parameter value for data.dir in Read10X()
• Import spatstat fxns from subpackages (spatstat.core, spatstat.geom)
• RunUMAP now checks for graph/neighbor consistency

Seurat 4.0.0 (2021-01-29)

Added

• Expose FoldChange() component in FindMarkers().
• Add the merge.DimReduc method
• Add IntegrateEmbeddings() to correct embeddings of DimReducs
• Add ProjectUMAP() to project query cells into a reference UMAP space
• Add MapQuery() as a wrapper around IntegrateData(), IntegrateEmbeddings(), and ProjectUMAP()
• Add MappingScore to compute a per-cell mapping score used in Azimuth
• Add AggregateExpression() for summation based pseudobulk calculations
• Add mixscape functionality via CalcPerturbSig(), PrepLDA(), RunLDA(), DEenrichRPlot(), MixscapeHeatmap(), MixscapeLDA(), PlotPerturbScore(), RunMixscape()
• Add FindSubCluster() to further cluster existing clusters
• Add supervised PCA functionality via RunSPCA()
• Add functionality to enable weighted nearest neighbor analyses via FindMultiModalNeighbors()
• Add neighbor visualization plot via NNPlot().
• Add PredictAssay() to impute expression or embeddings from nearest neighbors
• Add Graphs() function to access the names of the stored Graph objects or pull a specific one
• Add checks for NA, NaN, logical, non-integer, and infinite values during CreateAssayObject and NormalizeData.default
• Add AnnotateAnchors() to aid in AnchorSet interpretation as well as subset.AnchorSet()
• Add flexibility of choice for cell column in Read10X()
• Add rasterization option to FeatureScatter() and VariableFeaturePlot()
• Add step1 feature parameters in the SCTModel via PrepVSTResults()

Changes

• Default neighbor finding algorithm changed from "rann" to "annoy"
• Default ncells parameter in SCTransform() changed to 5000
• Default fold change in FindMarkers() changed from ln to log2
• Implementation improvements to AverageExpression()
• AnchorSet class re-implemented as a virtual class from which IntegrationAnchorSet and TransferAnchorSet now inherit.
• Point size in VlnPlot() now set automatically if not specified
• Return the sample.tree properly when integrating with a single reference dataset
• Replace as.character.quosure usage with as_label due to deprecation
• Minor changes to the exact calculation of the anchor weight matrix
• Default rasterization limit in DimPlot() and FeaturePlot() changed from 50,000 to 100,000
• SCTransform() now returns a formalized Assay subclass SCTAssay()
• When using normalization.method='SCT' in FindTransferAnchors(), normalize query using reference SCT model when possible.
• Change default Neighbor name in FindNeighbors to Assay.nn

Removed

• CreateGeneActivityMatrix replaced by Signac::GeneActivity()
• RunLSI replaced by by Signac::RunTFIDF() and Signac::RunSVD()
• ReadAlevin and ReadAlevinCsv moved to SeuratWrappers
• ExportToCellbrowser and StopCellbrowser moved to SeuratWrappers

Seurat 3.2.2 (2020-09-26)

Changes

• Set the seed in WhichCells regardless of whether or not idents is passed
• Retain Graph and Neighbor objects when subsetting only on features
• Fix data.frame input to CreateAssayObject() when data.frame has no rownames.
• Default annoy search to sequential if not using multicore future plans.
• Require sctransform >= 0.3.0

Seurat 3.2.1 (2020-09-07)

Added

• Added support for nearest neighbor input and return.model parameter in RunUMAP()
• Enable named color vectors in DoHeatmap()
• Add label.color and label.box parameters to DimPlot
• Added shuffle and seed parameters to DimPlot() to help with overplotting
• Added new stacked violin plot functionality

Changes

• Allow setting slot parameter in RunUMAP
• Added support for FIt-SNE v1.2+
• Fix for Spatial*Plot when running with interactive=TRUE
• Set max for number of items returned by Top and remove duplicate items when balanced=TRUE
• Fix logging bug when functions were run via do.call()
• Fix handling of weight.by.var parameter when approx=FALSE in RunPCA()
• Fix issue where feature names with dashes crashed CellSelector
• Fix issue where errors in subsetting were being swallowed
• Fix issue where labeling uncropped spatial plots was broken

Deprecated

• CreateActivityMatrix deprecated in favor of Signac::GeneActivity
• ReadAlevin and ReadAlevinCsv deprecated in favor of SeuratWrappers::ReadAlevin
• ExportToCellbrowser and StopCellbrowser deprecated in favor of SeuratWrappers::ExportToCellbrowser and SeuratWrappers::StopCellbrowser
• ReadH5AD and WriteH5AD deprecated in favor of h5Seurat/H5AD functionality found in SeuratDisk
• as.loom and as.Seurat.loom deprecated in favor of functionality found in SeuratDisk

Seurat 3.2.0 (2020-07-16)

Added

• Added ability to create a Seurat object from an existing Assay object, or any object inheriting from the Assay class
• Added ability to cluster idents and group features in DotPlot
• Added ability to use RColorBrewer plaettes for split DotPlots
• Added visualization and analysis functionality for spatially resolved datasets (Visium, Slide-seq).

Changes

• Removed add.iter parameter from RunTSNE function
• Fixed integer overflow error in the WilcoxDETest function
• Minor visual fixes in DoHeatmap group bar + labels
• Efficiency improvements in anchor scoring (ScoreAnchors)
• Fix bug in FindClusters() when the last node has no edges
• Default to weighted = TRUE when constructing igraph objects in RunLeiden. Remove corresponding weights parameter from FindClusters().
• Fix handling of keys in FeatureScatter()
• Change CellSelector to use Shiny gadgets instead of SDMTools
• Mark PointLocator as defunct
• Remove SDMTools
• Fixed data slot return in AverageExpression when subsetting features and returning a Seurat object

Seurat 3.1.5 (2020-04-16)

Added

• New scale parameter in DotPlot
• New keep.sparse parameter in CreateGeneActivityMatrix` for a more memory efficient option
• Added ability to store model learned by UMAP and project new data
• New strip.suffix option in Read10X. This changes the default behavior of Read10X. A trailing -1 present in all cell names will not be removed by default.
• Added group.by parameter to FeatureScatter

Changes

• Replace wilcox.test with limma implementation for a faster FindMarkers default method
• Better point separation for VlnPlots when using the split.by option
• Efficiency improvements for anchor pairing
• Deprecate redundant sort.cell parameter in FeaturePlot
• Fixes to ensure correct class of Matrix passed to c++ functions
• Fixes for underscores in ident labels for DotPlot
• Ensure preservation of matrix dimnames in SampleUMI
• Fix non-standard evaluation problems in subset and WhichCells
• Default split violin option is now a multi group option
• Preserve alpha in FeaturePlot when using blend
• Update assay.used slot for DimReducs when Assay is renamed

Seurat 3.1.4 (2020-02-26)

Changes

• Fixes to DoHeatmap to remain compatible with ggplot2 v3.3
• Adoption of patchwork framework to replace CombinePlots

Seurat 3.1.3 (2020-02-11)

Added

• New system agnostic Which function to address problems with FItSNE on Windows

Changes

• Export CellsByIdentities and RowMergeSparseMatrices functions
• nCount and nFeature metadata variables retained after subset and updated properly with UpdateSeuratObject
• Fix uwot support for running directly on feature matrices
• Fixes for keys with underscores
• Fix issue with leiden option for FindClusters
• Fix for data transfer when using sctransform
• SDMTools moved to Suggests as package is orphaned

Seurat 3.1.2 (2019-12-12)

Added

• New silent slot updater
• New random seed options to RunCCA, RunTSNE, WhichCells, HTODemux, AddModuleScore, VlnPlot, and RidgePlot
• Enhancements for dealing with Assay-derived objects

Changed

• Only run CalcN (generates nFeatures and nCounts) when counts changes
• Fix issue regarding colons in feature names
• Change object class testing to use inherits or is.* for R 4.0 compatability

Seurat 3.1.1 (2019-10-03)

Added

• New RegroupIdents function to reassign idents based on metadata column majority
• UpdateSymbolList function to pull new gene names from HGNC
• Added support for H5AD layers as additional assays in a Seurat object

Changed

• Fix rownames issue when running UMAP on dist object
• Add support for new H5AD obsm and varm stucture
• Fix issue when trying to read non-existent feature-level metadata from an H5AD file
• Fix in integration workflow when using SCTransform
• Improved error checking for AddModuleScore
• cbind fix in reference-based integration (MapQuery)
• Fix for convenience plots error hanging
• Ensure Seurat objects aren't stored in the command logs

Seurat 3.1.0 (2019-08-20)

Added

• New PrepSCTIntegration function to facilitate integration after SCTransform
• Reference-based integration with the reference parameter in FindIntegrationAnchors
• Reciprocal PCA as a reduction option in FindIntegrationAnchors
• New CollapseEmbeddingOutliers function
• Enable FindTransferAnchors after SCTransform
• Added back ColorDimSplit functionality
• Include a code of conduct
• Added uwot support as new default UMAP method
• Added CheckDots to catch unused parameters and suggest updated names
• Reductions and Assays assays functions to list stored DimReducs and Assays

Changed

• Fix regex in LogSeuratCommand
• Check for NAs in feature names in Read10X
• Prevent dimnames for counts/data/scale.data matrices from being arrays
• Updates ReadH5AD to distinguish FVF methods
• Fixes to UpdateSeuratObject for v2 objects
• Sink all output from stdout to stderr
• Fix to scale.data cell ordering after subsetting
• Enable Assay specification in BuildClusterTree
• Fix FeaturePlot when using both blend and split.by
• Fix to WhichCells when passing cells and invert
• Fix to HoverLocator labels and title
• Ensure features names don't contain pipes (|)
• Deprecation of RunLSI and RunALRA
• Fix legend bug when sorting in ExIPlot

Seurat 3.0.2 (2019-06-14)

Added

• Flag to skip singleton grouping in FindClusters
• New custom colors for blended FeaturePlots
• New GetResidual function
• New Seurat/Monocle converters

Changed

• Fix issue where certain assays weren't being shown in the Seurat object
• Fix issue where we weren't updating DimReduc object column names
• Fix line spacers in DoHeatmap
• Fix uninformative labels in FeaturePlot
• Fix unset identities when converting from SCE to Seurat
• Fix single colors being interpreted as palettes in SingleDimPlot
• Ensure factor levels are always numerically increasing after FindClusters
• Better cell highlighting colors for DimPlot
• Fix to levels<-.Seurat
• Add ability to use counts/scaled data in BuildClusterTree
• Minor fix to split ScaleData

Seurat 3.0.1 (2019-05-16)

Added

• Add global option (Seurat.memsafe) to skip gc() calls
• Restore draw.lines to DoHeatmap, maintain size of color bar with different number of features (#1429)
• Enable split.by parameter for ScaleData
• Add slot parameter to FeaturePlot (#1483)
• Add assay parameter to DotPlot (#1404)

Changed

• Fix to color options for VlnPlot with split.by option (#1425)
• Improvements to conversion functions (loom, SCE)
• Fix for cluster tree reordering (#1434)
• Fix PercentageFeatureSet for single feature case
• Fix to fold change calculation and filtering for other slots in FindMarkers (#1454)
• Keep title vectorized in AugmentPlot (#1515)
• Export LogSeuratCommand function
• Fix for FindConservedMarkers when one ident is missing from a group (#1517)

Seurat 3.0.0 (2019-04-15)

Added

• New method for identifying anchors across single-cell datasets
• Parallelization support via future
• Additional method for demultiplexing with MULTIseqDemux
• Support normalization via sctransform
• New option for clustering with the Leiden algorithm
• Support for reading 10X v3 files
• New function to export Seurat objects for the UCSC cell browser
• Support for data import from Alevin outputs
• Imputation of dropped out values via ALRA

Changed

• Significant code restructuring
• Most occurances of "gene(s)" in function names/arguments renamed to "feature(s)"
• Changes to the Seurat object class to facilitate multimodal data
• New BlendPlot implementation

Seurat 2.3.4 (2018-07-17)

Added

• GetIdent function added to pull identity info

Changed

• DiffusionMap dependency replaced with destiny to avoid archival
• Java dependency removed and functionality rewritten in Rcpp
• Speed and efficiency improvements for Rcpp code
• More robust duplicate handling in CellCycleScoring

Seurat 2.3.3 (2018-07-02)

Added

• New HTOHeatmap function
• Support for custom PNG arguments for vector-friendly plotting
• Fix for 'NA'-labeled cells disappearing with custom color scale

Changed

• Replaced FNN with RANN
• Removed unused compiler flags
• Moved several lightly-used packages from 'imports' to 'suggests'

Seurat 2.3.2 (2018-06-12)

Added

• RenameCells added for easy renaming of all cells
• Read10X_h5 added to read in 10X formatted h5 files
• SetAssayData ensures cell order is the same between assay objects and the Seurat object
• Compatability updates for ggplot2 v2.3.0

Seurat 2.3.1 (2018-05-05)

Added

• Support for UMAP dimensional reduction technique
• New conversion functions for SingleCellExperiment and anndata

Changed

• FetchData preserves cell order
• Require Matrix 1.2-14 or higher
• AddModuleScore no longer densifies sparse-matrices
• Various visualization fixes and improvements
• Default value for latent.vars in FindMarkers/FindAllMarkers changed to NULL.

Seurat 2.3.0 (2018-03-23)

Added

• Support for HTO demultiplexing
• Utility functions: TransferIdent, CombineIdent, SplitObject, vector.friendly
• C++ implementation for parts of BuildSNN
• Preliminary parallelization support (regression and JackStraw)
• Support for FItSNE

Changed

• MetaDE replaced with metap for combining p-values (MetaDE was removed from CRAN)
• NMF heatmaps replaced (NMF to be archived by CRAN)

Seurat 2.2.1 (2018-02-14)

Changed

• MetaDE replaced with metap for combining p-values (MetaDE was removed from CRAN)
• NMF heatmaps replaced (NMF to be archived by CRAN)

Seurat 2.2.0 (2018-01-10)

Added

• Multiple alignment functionality with RunMultiCCA and AlignSubspace extended to multiple datasets
• CalcAlignmentScore added to evaluate alignment quality
• MetageneBicorPlot added to guide CC selection
• Change cluster order in DoHeatmap with group.order parameter
• Ability to change plotting order and add a title to DimPlot
• do.clean and subset.raw options for SubsetData

Changed

• JoyPlot has been replaced with RidgePlot
• FindClusters is now more robust in making temp files
• MetaDE support for combining p-values in DE testing

Seurat 2.1.0 (2017-10-12)

Added

• Support for using MAST and DESeq2 packages for differential expression testing in FindMarkers
• Support for multi-modal single-cell data via @assay slot

Changed

• Default DE test changed to Wilcoxon rank sum test

Seurat 2.0.1 (2017-08-22)

Added

• Now available on CRAN
• Updated documentation complete with examples
• Example datasets: pbmc_small and cc.genes
• C++ implementation for parts of FindVariableGenes
• Minor bug fixes

Seurat 2.0.0

Added

• New method for aligning scRNA-seq datasets
• Significant code restructuring
• New methods for scoring gene expression and cell-cycle phases
• New visualization features (do.hover, do.identify)