We expect a custom_selections
HDF5 group at the root of the file, containing information about the custom selections.
The group itself contains the parameters
and results
subgroups.
Definitions:
num_cells
: number of cells remaining after QC filtering. This is typically determined from thequality_control
step.num_genes
: number of genes in the dataset. This is typically determined from theinputs
step.
parameters
should contain:
selections
: a group defining the custom selections. Each child is named after a user-created selection. Each child is an integer dataset of arbitrary length containing the indices of the selected cells. Note that indices refer to the dataset after QC filtering and should be less thannum_cells
.
results
should contain:
markers
: a group containing the marker results for each selection after a comparison to a group containing all other cells. Each child is named after its selection and is a group containing:means
: a float dataset of length equal tonum_genes
, containing the mean expression of each gene in the selection.detected
: a float dataset of length equal tonum_genes
, containing the proportion of cells with detected expression of each gene in the selection.lfc
: a float dataset of length equal tonum_genes
, containing the log-fold change in the selection compared to all other cells.delta_detected
: same aslfc
, but for the delta-detected (i.e., difference in the percentage of detected expression).cohen
: same aslfc
, but for Cohen's d.auc
: same aslfc
, but for the AUCs.
Added in version 1.0.