Number of cores/cpus to use. Default: 1

Dimensionality to use in UMAP projection and when constructing nearest-neighbor graph before clustering (from 1 to 50). If single value N is provided, use from 1 to N dimensions. If multiple values are provided, subset to only selected dimensions. Default: from 1 to 10

Clustering resolution applied to the constructed nearest-neighbor graph. Can be set as an array. Default: 0.3, 0.5, 1.0

For putative gene markers identification include only those genes that are detected in not lower than this fraction of cells in either of the two tested clusters. Ignored if '--diffgenes' is not set. Default: 0.1

Statistical test to use for putative gene markers identification. Ignored if '--diffgenes' is not set. Default: wilcox

UMAP implementation to run. If set to 'umap-learn' use --umetric 'correlation' Default: uwot

The metric to use to compute distances in high dimensional space for UMAP. Default: cosine

The effective scale of embedded points on UMAP. In combination with '--mindist' it determines how clustered/clumped the embedded points are. Default: 1

Regex pattern to identify mitochondrial genes. Default: '^Mt-'

Controls how tightly the embedding is allowed compress points together on UMAP. Larger values ensure embedded points are moreevenly distributed, while smaller values allow the algorithm to optimise more accurately with regard to local structure. Sensible values are in the range 0.001 to 0.5. Default: 0.3

Path to the headerless TSV/CSV file with the list of barcodes to select cells of interest (one barcode per line). Prefilters input feature-barcode matrix to include only selected cells. Default: use all cells.

Path to the TSV/CSV file to define datasets grouping. First column - 'library_id' with the values and order that correspond to the 'library_id' column from the '--identity' file, second column 'condition'. Default: each dataset is assigned to its own group.

Include cells with the number of genes not bigger than this value. If multiple values provided, each of them will be applied to the correspondent dataset from the '--mex' input based on the '--identity' file. Default: 5000 (applied to all datasets)

Include cells where at least this many genes are detected. If multiple values provided, each of them will be applied to the correspondent dataset from the '--mex' input based on the '--identity' file. Default: 250 (applied to all datasets)

For putative gene markers identification include only those genes that on average have log fold change difference in expression between every tested pair of clusters not lower than this value. Ignored if '--diffgenes' is not set. Default: 0.25

Regress genes per cell counts as a confounding source of variation. Default: false

Distance metric used when constructing nearest-neighbor graph before clustering. Default: euclidean

Determines the number of neighboring points used in UMAP. Larger values will result in more global structure being preserved at the loss of detailed local structure. In general this parameter should often be in the range 5 to 50. Default: 30

Path to the TSV/CSV file with the information for cell cycle score assignment. First column - 'phase', second column 'gene_id'. If loaded Seurat object already includes cell cycle scores in 'S.Score' and 'G2M.Score' metatada columns they will be removed. Default: skip cell cycle score assignment.

Regress UMI per cell counts as a confounding source of variation. Default: false

Include cells where at least this many UMI (transcripts) are detected. If multiple values provided, each of them will be applied to the correspondent dataset from the '--mex' input based on the '--identity' file. Default: 500 (applied to all datasets)

Genes of interest to build genes expression plots. Default: None

Include cells with the percentage of transcripts mapped to mitochondrial genes not bigger than this value. Default: 5 (applied to all datasets)

Regress cell cycle scores as a confounding source of variation. Ignored if --cellcycle is not provided. Default: false

Regress the percentage of transcripts mapped to mitochondrial genes as a confounding source of variation. Default: false

Include only genes detected in at least this many cells. Default: 5 (applied to all datasets)

Integration method used for joint analysis of multiple datasets. Automatically set to 'none' if loaded Suerat object includes only one dataset. Default: seurat

Identify differentially expressed genes (putative gene markers) between each pair of clusters for all resolutions. Default: false

Maximum vector memory in GB allowed to be used by R. Default: 128

Path to the metadata TSV/CSV file to set the datasets identities. If '--mex' points to the Cell Ranger Aggregate outputs, the aggregation.csv file can be used. If input is not provided, the default dummy_metadata.csv will be used instead.

Normalization method applied to genes expression counts. If loaded Seurat object includes multiple datasets, normalization will be run independently for each of them, unless integration is disabled with --ntgr set to 'none' Default: sct

Include cells with the novelty score not lower than this value, calculated for as log10(genes)/log10(UMI). If multiple values provided, each of them will be applied to the correspondent dataset from the '--mex' input based on the '--identity' file. Default: 0.8 (applied to all datasets)

Maximum memory in GB allowed to be shared between the workers when using multiple --cpus. Default: 32

Number of highly variable genes used in datasets integration, scaling and dimensionality reduction. Default: 3000

For putative gene markers identification return only positive markers. Ignored if '--diffgenes' is not set. Default: false

Path to the compressed folder with feature-barcode matrix from Cell Ranger Count/Aggregate experiment in MEX format.