Running Compass from Seurat object

[BB8-panda]

Hi All,

I'm very new in bioinformatics. I've done my cell annotation etc using Seurat pipelines and I want to use CompassR for metabolic flux, pathway prediction analysis. However, I'm stuck at not being able to generate reactions.tsv. Could anyone please explain to me or direct me to tutorial on how to get the reactions.tsv in R after Seurat pipelines?

Any leads would be much appreciated, thank you!

[lzh-fugwaa]

I also encountered this problem, May I ask if you have solved it?

[BB8-panda]

I ended up processing my data on their package in python to get the reactions.tsv file. Then came back to R for the downstream analysis just because I’m more familiar with R.

I followed the steps mentioned here:
https://compass-wagnerlab.readthedocs.io/en/latest/

[lzh-fugwaa]

I ended up processing my data on their package in python to get the reactions.tsv file. Then came back to R for the downstream analysis just because I’m more familiar with R.

I followed the steps mentioned here: https://compass-wagnerlab.readthedocs.io/en/latest/

Thank you for your prompt reply! I have another question. Does it take a long time to run in python for compass?

[BB8-panda]

No problem!

I think it depends on your equipment. I used 16GB-RAM desktop and it took me days to run the analysis. I gave that up and rerun the analysis with only a specific cell type (200ish cells) and it took several hours on my desktop computer. I suggest you run the analysis for one cell type at a time (or do one sample at a time, or do pseudobulk, or do other way to lower the cell number) if you have limited computer power as I do.

[lzh-fugwaa]

I am very grateful for your reply. Currently, I am using the server of my research group, so the process should be faster.