Keep high-quality cells based on QC columns

Usage

keep_quality_cells(
  data,
  empty_droplet_col = "empty_droplet",
  alive_col = "alive",
  doublet_col = "scDblFinder.class"
)

Source

Mangiola et al.,2023

Arguments

data: A data frame or tibble containing single-cell metadata.
empty_droplet_col: A string specifying the column name that indicates empty droplets (default: "empty_droplet"). Expected logical vector
alive_col: A string specifying the column name that indicates whether cells are alive (default: "alive"). Expected logical vector
doublet_col: A string specifying the column name that indicates doublets (default: "scDblFinder.class"). Expected character vector: "doublet" and/or "singlet" and/or "unknown".

Value

A filtered data frame containing only cells that pass all QC checks.

Examples

get_metadata(cloud_metadata = SAMPLE_DATABASE_URL, cache_directory = tempdir()) |>
  head(2) |>
  keep_quality_cells()
#> # A query:  ?? x 36
#> # Database: DuckDB 1.5.4 [unknown@Linux 6.17.0-1020-azure:R 4.6.1/:memory:]
#>   cell_id dataset_id            sample_id sample_ experiment___ run_from_cell_id
#>     <dbl> <chr>                 <chr>     <chr>   <chr>         <chr>           
#> 1      15 842c6f5d-4a94-4eef-8… 1119f482… 1119f4… ""            NA              
#> 2      16 842c6f5d-4a94-4eef-8… 1119f482… 1119f4… ""            NA              
#> # ℹ 30 more variables: sample_heuristic <chr>, age_days <int>,
#> #   tissue_groups <chr>, nFeature_expressed_in_sample <int>, nCount_RNA <dbl>,
#> #   empty_droplet <lgl>, cell_type_unified_ensemble <chr>, is_immune <lgl>,
#> #   subsets_Mito_percent <int>, subsets_Ribo_percent <int>,
#> #   high_mitochondrion <lgl>, high_ribosome <lgl>, scDblFinder.class <chr>,
#> #   sample_chunk <int>, cell_chunk <int>, sample_pseudobulk_chunk <int>,
#> #   file_id_cellNexus_single_cell <chr>, file_id_cellNexus_pseudobulk <chr>, …