RCTD deconvolution using an `screference`
rctd_deconvolute.RdRCTD deconvolution using an `screference`
Usage
rctd_deconvolute(
spatial_obj,
scref,
ncores = 4,
gene_cutoff = 0.000125,
fc_cutoff = 0.5,
gene_cutoff_reg = 2e-04,
fc_cutoff_reg = 0.75,
UMI_min = 100,
UMI_max = 2e+07,
counts_MIN = 10,
UMI_min_sigma = 300,
CELL_MIN_INSTANCE = 25,
MAX_MULTI_TYPES = 4,
CONFIDENCE_THRESHOLD = 5,
DOUBLET_THRESHOLD = 20,
cache_path = "rctd"
)Arguments
- spatial_obj
an `Seurat` object with a Spatial assay
- scref
an object of `screference`
- gene_cutoff
minimum normalized gene expression for genes to be included in the platform effect normalization step.
- fc_cutoff
minimum log-fold-change (across cell types) for genes to be included in the platform effect normalization step.
- gene_cutoff_reg
minimum normalized gene expression for genes to be included in the RCTD step.
- fc_cutoff_reg
minimum log-fold-change (across cell types) for genes to be included in the RCTD step.
- UMI_min
minimum UMI per pixel included in the analysis
- UMI_max
maximum UMI per pixel included in the analysis
- counts_MIN
minimum total counts per pixel of genes used in the analysis.
- UMI_min_sigma
minimum UMI per pixel for the choose_sigma_c function
- CELL_MIN_INSTANCE
minimum number of cells required per cell type. Default 25, can be lowered if desired.
- MAX_MULTI_TYPES
max number of cell types per pixel
- CONFIDENCE_THRESHOLD
the minimum change in likelihood (compared to other cell types) necessary to determine a cell type identity with confidence
- DOUBLET_THRESHOLD
the penalty weight of predicting a doublet instead of a singlet for a pixel
- cache_path
a path to cache the results
- max_cores
for parallel processing, the number of cores used. If set to 1, parallel processing is not used. The system will additionally be checked for number of available cores.