Ribosomal cluster

Question

[Chris]

Hi Biostars,

I work on single cell data and got this issue hope you can help. I got a cluster has many ribosomal genes. So what can we say about this? I think error with sample preparation. Thank you so much! I found a relevant post here, josegarciamanteiga's comment: https://github.com/satijalab/seurat/issues/759

[seidel]

We don't know the organism, we don't know the context, we don't know the experimental conditions....why is it a problem if you have a cluster enriched for a certain kind of biology? Why do you think an error with sample preparation would cause this? Are there any aspects of your data set that you can verify? i.e. any clusters that do make sense, or that indicate cell types you know to exist?

[Chris]

I don't say it is a problem just want to know about it because normally in single cell, I see each cluster according to a cell type but not a cluster with many ribosomal genes. In sample preparation, we remove ribosome RNA because it has the most abundance but not the thing we interested in. Yes, other clusters make sense because I expected those cell type. It is the human sample, iPSC differential into other cell type.

[Ram]

1. What was the exact subset call?
2. How did you determine that your cluster has "many" MT genes?
3. Does removing cells with high percent_mt_genes ensure that clusters will be depleted of all MT genes?

[Chris]

seurat_mtx <- subset(seurat_mtx, subset = nCount_RNA < 100000 & 
                          nFeature_RNA > 100 &
                          percent.mt <35)

My mistake, ribosomal genes but not mitochondria gene. A cluster has these genes with ave_log2FC > 2: RPL39, TPT1, RPS12, RPS29, RPL26, RPS21, RPL12, RPS15A, RPL32, RPL17, FTL, FTH1, RPS13, RPS14, RPL29, RPL24.

[Ram]

Please edit your post content and change "mitochondrial" to "ribosomal". I don't know how to interpret the fact that some of your clusters have a large fraction of ribosomal genes, maybe someone with more knowledge of biology can help.

[Chris]

Yes, I did. Thank you

[Ram]

Doesn't look like it - I'm still seeing: "After using subset() to remove cells have high mitochondria percentage, I still got a cluster has many mitochondria genes"