Table 1.
Summary of the tested deconvolution methods
| Method | Input | Type | Availability |
|---|---|---|---|
| Rodeo | C | expression | R/Rodeo |
| cs-lsfit | C | expression | R/CellMix |
| cs-qprog | C | expression | R/CellMix |
| LRCDE | C | expression | R/LRCDE |
| csSAM | C | expression | R/csSAM |
| Deblender | C | complete | Matlab |
| CDSeq | #T | complete | Matlab, Octave, R |
| LinSeed | #T | complete | R/linseed |
| Deconf | #T | complete | R/CellMix |
Column ’Input’ describes required input data from the user and C refers to cell type proportion matrix and #T to number of cell types. Notably, Deblender is a complete method, but the partial deconvolution can be done separately and in the context of this study, it is used as expression deconvolution method.