This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Looking for bulk RNA-seq deconvolution papers

I am looking for papers that performed deconvolution on bulk RNA-seq data, preferably using publicly accessible cohorts such as TCGA, with two requirements. First, the cell type reference must include epithelial (tumor or malignant epithelial) cells, not only immune and stromal compartments. Second, the deconvolution results should be publicly available so they can be inspected or downloaded.

For example, Thorsson et al.’s 2018 paper, “The Immune Landscape of Cancer,” is excellent, but it does not include epithelial cells, hence they used CIBERSORT, which is focused on immune cell markers. Given how widely deconvolution is used in cancer bulk RNA-seq, I expected this to be easy to find, but I have not been able to identify relevant papers that meet these criteria.

I would appreciate any help identifying studies that provide bulk RNA-seq deconvolution on public data, include epithelial cells, and make the resulting estimates available for review. Thank you!

Note - it doesn't have to be cancer papers, I mentioned TCGA as a popular example.

rna-seq bulk rnaseq deconvolution python

As an aside, CIBERSORT can be used with any reference data you have. It just has a built-in dataset for immune cells.

1 answer

Driver gene combinations dictate cutaneous squamous cell carcinoma disease continuum progression https://www.nature.com/articles/s41467-023-40822-9#Sec31

  • data is published with the study and uses a publicly available 2020 scRNAseq cSCC dataset for CiberSort deconvolution.

Deconvolution isn't high resolution, includes broad categories of malignant and normal epithelial cell populations - no immune cells, but still might be worth a look.

Log in to answer this question.