scRNA gene expression without control/comparator

Question

[yakumo]

Hi everyone,

I recently had a patient with a very rare lymphoma type and was able to perform scRNA-Seq of a fresh lymph node biopsy. Is there a way to check for differential gene expression based on one single sample, as I do not have any further (non-lymphoma) material from this patient to compare with his lymphoma sample?

I am grateful for any input!

Thanks a lot & best wishes

Answer 1

[jared.andrews07]

Sure, you can compare between clusters. But in terms of "malignant vs non-malignant" type comparisons, that depends on whether you can separate those populations with any confidence. If malignant tissues/cells have known markers, you can try to use those to classify cells/clusters as malignant or non-malignant cells. Immune profiling is useful for this in many lymphomas as well, which is something to keep in mind for future experiments.

[yakumo]

Thanks for your reply! Can you by any chance recommend a tool/package for this type of analysis? With DESeq2 I find it rather difficult, as it seems to require two sets of values to compare in order to perform differential analysis (at least that's the error message I receive when I try to run it on a single column of gene expression counts)... Many thanks!

[jared.andrews07]

What are you using to analyze your scRNA-seq data? Seurat, scran, monocle3 all have differential testing capabilities. I'd recommend reading the Orchestrating Single Cell Analysis book, as it has a lot of good info.