scRNA and stemness score

Question

[a511512345]

Hello guys

I am learning single-cell RNA sequencing analysis using the Seurat package. Here, I have a Seurat object with 9 cell clusters. How to calculate the stemness score for each cluster.

Can anyone tell me a relevant R package or example?

Thank you very much

[ATpoint]

These scores are basically just an aggregate of expression of a set of genes. In your case you would need to define stemcell-related genes and then get such a score, e.g. using something like AUCell or SingleR. The former uses a spearman correlation and the latter an "area under curve" approach. Check the manuals of both, they're at Bioconductor. Eventually you will get a score per gene for a given gene set which you can use to color your UMAP (or similar types of plots) with. The most naive approach would be to get the median or mean for the expression of the gene set per cell.

Answer 1

[fracarb8]

I recently worked on a dataset of progenitors T cells, and I started playing with STEMNET. It is a dimensionality reduction algorithm that was developed to obtain a detailed view on the transition from stem cells to lineage-restricted progenitors in the continuous HSPC (hematopoietic stem- and progenitor cells) landscape. It basically uses hierarchical clustering to identify the most mature populations, and it then uses these populations as a training set to classify priming in the less mature populations. At the end you have a similarity score of each cells to the progenitor classes.