This is something that I find myself doing manually often. I'll look at the top differentially expressed genes and then go to the Single Cell tab and see what cell types in a comparable single-cell sample express those genes.
The idea would be that you could select a single cell dataset for "gene-level" deconvolution. This would annotate genes with a cell type(s) whose expression can be accounted for by those cell type(s).
This is something that I find myself doing manually often. I'll look at the top differentially expressed genes and then go to the Single Cell tab and see what cell types in a comparable single-cell sample express those genes.
The idea would be that you could select a single cell dataset for "gene-level" deconvolution. This would annotate genes with a cell type(s) whose expression can be accounted for by those cell type(s).