Hello,
Is anyone working on Cell Plex using 10X Genomics reagents?
My question is related to the method of analysis you are using for the analysis of single-cell sequencing using Cell Plex?
Which method are you using?
1. Loupe Brower
2. Seurat
a) Individual tumor
b) Merging the tumors and using the integrated analysis in Seurat.
c) Using the codes from 10X Genomics and using the additional codes from Seurat and running the data?
If you are using the c method, are you also having issues with the RUNPCA command?
Were you able to resolve this?
Thank you so much.
Kind Regards,
Shweta Johari
Hi Shweta,
First, don't use Loupe if you want any deep dive analysis of single cell data. I personally run most data normalization, filtering, and integration in Seurat v4 (make sure you are using the latest version if you want to use Seurat, and check what version of Seurat is being referenced by your guide). RunPCA works fine in Seurat v4. Don't use solution "c" as 10X Genomics and Seurat (or any other package/tool for that matter) developers are not "collaborating" on their guides so if they put anything that does not work, they won't know about it. I am presuming that you have problems after integration since you are asking about RunPCA. Maybe share your error code or what specifically was the problem?
Best wishes on your work!
Hey Wilson,
Thank you so much for your reply.
I completely agree on not running on Loupe.
I am using the latest Seurat version.
For c, I am facing an error as in the combined codes, the RNA assay is not there by the time we reach the RUNPCA code, the reason being in the initial steps default assay is CMO.
Only when I used the c method, I am facing a major issue with c due to the RNA assay not being present and when I try to run the selection.method = "mean.var.plot", it gives me the following error:
'Error in irlba(A = t(x = object), nv = npcs, ...) : max(nu, nv) must be positive'
I wrote to 10X genomics, but they have still not responded with a solution.
I have written on githib as well. A solution was given for the combination of 10X genomics + Seurat, but still, it is not running swiftly.
For Seurat, individual tumors are working great. I am working on the integration of it now, I hope that goes well.
Thank you for your reply.
Kind Regards,
Shweta Johari
10X will not be able to help you when you are mixing codes.
This code:
selection.method = "mean.var.plot",
tells me that you are using a deprecated script. This is no longer needed. I suggest you use the following vignette:
https://satijalab.org/seurat/articles/integration_introduction.html
And use the SCTransform. You should not have any problems for data integratio.
Hey Wilson,
Thank you. I was using that code for the c method but when I do integration, I will use SCT and that vignette.
Thank you again.
Kind Regards,
Shweta Johari
- Badges
- Science method