Weird bug, when exporting from R (in this case a Seurat object), NA's in specifically the metadata data frame (though it begs the question what happens if there are NA's in the assay), these are automatically converted to True values in python data frames.
For a more explicit example, I generated metadata with True/False conditions in a column on a subset of cells, wherein the set of cells that were excluded were assigned NA values. When this was imported to python, the cells with NA values in said columns were automatically populated with True values.
I confirmed this by converting NA's to a character string and then the values were preserved.
Weird bug, when exporting from R (in this case a Seurat object), NA's in specifically the metadata data frame (though it begs the question what happens if there are NA's in the assay), these are automatically converted to True values in python data frames.
For a more explicit example, I generated metadata with True/False conditions in a column on a subset of cells, wherein the set of cells that were excluded were assigned NA values. When this was imported to python, the cells with NA values in said columns were automatically populated with True values.
I confirmed this by converting NA's to a character string and then the values were preserved.