Build SummarizedExperiment using multiple sfaira entries — dataset_sfaira

You can apply different filters on the whole data-zoo of sfaria; the resulting single-cell datasets will be combined into a single dataset which you can use for simulation Note: only datasets in sfaira with annotation are considered!

Usage

dataset_sfaira_multiple(
  organisms = NULL,
  tissues = NULL,
  assays = NULL,
  sfaira_setup,
  name = "SimBu_dataset",
  spike_in_col = NULL,
  additional_cols = NULL,
  filter_genes = TRUE,
  variance_cutoff = 0,
  type_abundance_cutoff = 0,
  scale_tpm = TRUE
)

Arguments

organisms: (mandatory) list of organisms (only human and mouse available)
tissues: (mandatory) list of tissues
assays: (mandatory) list of assays
sfaira_setup: (mandatory) the sfaira setup; given by setup_sfaira
name: name of the dataset; will be used for new unique IDs of cells
spike_in_col: which column in annotation contains information on spike_in counts, which can be used to re-scale counts
additional_cols: list of column names in annotation, that should be stored as well in dataset object
filter_genes: boolean, if TRUE, removes all genes with 0 expression over all samples & genes with variance below variance_cutoff
variance_cutoff: numeric, is only applied if filter_genes is TRUE: removes all genes with variance below the chosen cutoff
type_abundance_cutoff: numeric, remove all cells, whose cell-type appears less then the given value. This removes low abundant cell-types
scale_tpm: boolean, if TRUE (default) the cells in tpm_matrix will be scaled to sum up to 1e6

Value

dataset object

Examples

# \donttest{
setup_list <- SimBu::setup_sfaira(tempdir())
#> AttributeError: module 'pandas.arrays' has no attribute 'ArrowStringArray'
ds_human_lung <- SimBu::dataset_sfaira_multiple(
  sfaira_setup = setup_list,
  organisms = "Homo sapiens",
  tissues = "lung parenchyma",
  assay = "10x 3' v2",
  name = "human_lung"
)
#> Warning: You need to setup sfaira first; please use setup_sfaira() to do so.
# }