Hello! I've been testing RCTD for deconvolution of CosMx data from a single cell reference using the doublet mode. In roughly a quarter of the cells, the weights for the first_type are lower than that of the second_type. Can you provide guidance on interpreting this?
In the dataset, only about 50% of the cells are in the singlet or doublet_certain categories. So if I use get_doublet_weights to return a weight matrix across all cell types, there will be many 0's. Does this indicate that the doublet mode isn't appropriate?
Thanks!
Hello! I've been testing RCTD for deconvolution of CosMx data from a single cell reference using the doublet mode. In roughly a quarter of the cells, the weights for the
first_typeare lower than that of thesecond_type. Can you provide guidance on interpreting this?In the dataset, only about 50% of the cells are in the
singletordoublet_certaincategories. So if I useget_doublet_weightsto return a weight matrix across all cell types, there will be many 0's. Does this indicate that the doublet mode isn't appropriate?Thanks!