Result interpretation
The classify_cells function returns the input object but with additional columns in the metadata table.
# display the additional metadata fields
seurat.obj[[]][c(20:30), c(8:21)]
#> B_cells_p B_cells_class NK_p NK_class
#> Cy80_II_CD45_D04_S904_comb 0.007983681 no 0.3836233 no
#> Cy80_II_CD45_C12_S900_comb 0.007668953 no 0.5054545 yes
#> Cy80_II_CD45_C02_S890_comb 0.026575995 no 0.3231831 no
#> Cy80_II_CD45_F01_S925_comb 0.029493459 no 0.3658388 no
#> Cy80_II_CD45_G10_S946_comb 0.024490415 no 0.3791378 no
#> Cy80_II_CD45_F08_S932_comb 0.010955256 no 0.4696540 no
#> Cy80_II_CD45_E01_S913_comb 0.008845469 no 0.3526443 no
#> Cy80_II_CD45_H11_S959_comb 0.009423056 no 0.4397574 no
#> Cy80_II_CD45_A01_S865_comb 0.045843008 no 0.3768446 no
#> Cy80_II_CD45_E11_S923_comb 0.014868335 no 0.3850901 no
#> Cy80_II_CD45_E08_S920_comb 0.012824167 no 0.3892861 no
#> Plasma_cells_p Plasma_cells_class T_cells_p
#> Cy80_II_CD45_D04_S904_comb NA <NA> 0.10912289
#> Cy80_II_CD45_C12_S900_comb NA <NA> 0.10356344
#> Cy80_II_CD45_C02_S890_comb NA <NA> 0.09935477
#> Cy80_II_CD45_F01_S925_comb NA <NA> 0.52573389
#> Cy80_II_CD45_G10_S946_comb NA <NA> 0.19678523
#> Cy80_II_CD45_F08_S932_comb NA <NA> 0.10993410
#> Cy80_II_CD45_E01_S913_comb NA <NA> 0.07988881
#> Cy80_II_CD45_H11_S959_comb NA <NA> 0.11294201
#> Cy80_II_CD45_A01_S865_comb NA <NA> 0.01969278
#> Cy80_II_CD45_E11_S923_comb NA <NA> 0.11678435
#> Cy80_II_CD45_E08_S920_comb NA <NA> 0.24587206
#> T_cells_class CD4_T_cells_p CD4_T_cells_class
#> Cy80_II_CD45_D04_S904_comb no NA <NA>
#> Cy80_II_CD45_C12_S900_comb no NA <NA>
#> Cy80_II_CD45_C02_S890_comb no NA <NA>
#> Cy80_II_CD45_F01_S925_comb yes 0.4909765 no
#> Cy80_II_CD45_G10_S946_comb no NA <NA>
#> Cy80_II_CD45_F08_S932_comb no NA <NA>
#> Cy80_II_CD45_E01_S913_comb no NA <NA>
#> Cy80_II_CD45_H11_S959_comb no NA <NA>
#> Cy80_II_CD45_A01_S865_comb no NA <NA>
#> Cy80_II_CD45_E11_S923_comb no NA <NA>
#> Cy80_II_CD45_E08_S920_comb no NA <NA>
#> CD8_T_cells_p CD8_T_cells_class Monocytes_p
#> Cy80_II_CD45_D04_S904_comb NA <NA> 0.215263
#> Cy80_II_CD45_C12_S900_comb NA <NA> 0.215263
#> Cy80_II_CD45_C02_S890_comb NA <NA> 0.215263
#> Cy80_II_CD45_F01_S925_comb 0.3240697 no 0.215263
#> Cy80_II_CD45_G10_S946_comb NA <NA> 0.215263
#> Cy80_II_CD45_F08_S932_comb NA <NA> 0.215263
#> Cy80_II_CD45_E01_S913_comb NA <NA> 0.215263
#> Cy80_II_CD45_H11_S959_comb NA <NA> 0.215263
#> Cy80_II_CD45_A01_S865_comb NA <NA> 0.215263
#> Cy80_II_CD45_E11_S923_comb NA <NA> 0.215263
#> Cy80_II_CD45_E08_S920_comb NA <NA> 0.215263
#> Monocytes_class
#> Cy80_II_CD45_D04_S904_comb no
#> Cy80_II_CD45_C12_S900_comb no
#> Cy80_II_CD45_C02_S890_comb no
#> Cy80_II_CD45_F01_S925_comb no
#> Cy80_II_CD45_G10_S946_comb no
#> Cy80_II_CD45_F08_S932_comb no
#> Cy80_II_CD45_E01_S913_comb no
#> Cy80_II_CD45_H11_S959_comb no
#> Cy80_II_CD45_A01_S865_comb no
#> Cy80_II_CD45_E11_S923_comb no
#> Cy80_II_CD45_E08_S920_comb noNew columns are:
predicted_cell_type: The predicted cell type, also containing any ambiguous assignments. In these cases, the possible cell types are separated by a “/”
most_probable_cell_type: Present only if
simplified_resultwas set toTRUE. Contains the most probably cell type ignoring any ambiguous assignments.columns with syntax
[celltype]_p: probability of a cell to belong to a cell type. Unknown cell types are marked as NAs.
Result visualization
The predicted cell types can now simply be visualized using the matching plotting functions. In this example, we use Seurat’s DimPlot function:
With the current number of cell classifiers, we identify cells belonging to 2 cell types (B cells and T cells) and to 2 subtypes of T cells (CD4+ T cells and CD8+ T cells). The other cells (red points) are not among the cell types that can be classified by the predefined classifiers. Hence, they have an empty label.
For a certain cell type, users can also view the prediction probability. Here we show an example of B cell prediction probability:
Cells predicted to be B cells with higher probability have darker color, while the lighter color shows lower or even zero probability of a cell to be B cells. For B cell classifier, the threshold for prediction probability is currently at 0.5, which means cells having prediction probability at 0.5 or above will be predicted as B cells.
The automatic cell identification by scClassifR matches the traditional cell assignment, ie. the approach based on cell canonical marker expression. Taking a simple example, we use CD19 and CD20 (MS4A1) to identify B cells:
We see that the marker expression of B cells exactly overlaps the B cell prediction made by scClassifR.