how to get immundeconv cibersort to match standalone cibersort #138
Replies: 3 comments
-
Hello @RichardCorbett Can you try the command
And check if the results are the same? |
Beta Was this translation helpful? Give feedback.
-
Thanks for the suggestion, @LorenzoMerotto! In my analysis I am providing 1 sample's TPMs duplicated across 2 columns (with unique sample names) so I don't expect differences related to QN. I tried running with your suggested command and still got the |
Beta Was this translation helpful? Give feedback.
-
Your code did work? I try the same and get the following error:
and get the error: Error in (function (sig_matrix, mixture_file, perm = 0, QN = TRUE, saveLocation) : unused arguments (absolute = FALSE, abs_method = "sig.score") What can I do here? |
Beta Was this translation helpful? Give feedback.
-
Hi all,
I thought this would be simple, but so far the solution has eluded me.
I have STAR/RSEM ens100 genes.results files. When I pull out the HGNC and TPM columns and supply them to standalone CIBERSORT (in absolute mode) I get a CD8+ score of 0.07.
When I load my genes.results files into R and then run CIBERSORT through immunedeconv I get a CD8+ estimate of 0.14.
in my standalone CIBERSORT analysis I use the following params:
and then through immunedeconv I use the following line:
I've been at this for days and haven't found the cause of the difference. The gene list, LM22, and TPM values are the same going into both commands. I'm sure I'm missing something obvious. Has anyone done this themselves and figured it out?
Beta Was this translation helpful? Give feedback.
All reactions