Danko-Lab
Can the TCGA bulk RNA-seq data be dissected using only macrophage subsets ?
If I only had single-cell RNAseq data from macrophages as a reference， however， I want to analysis the ratio of macrophage subtypes in TCGA bulk RNAseq. Can I use BayesPrism? I have tried to do this analysis, however, the total ratio is 1 in final result. In theory, bulk RNAseq also contains T cells, malignant cells, fibroblast and so on.
Here are the results.
Thank you for your interest in our methods.
Just to make sure I understand. I noticed there are 11 columns in the figure you uploaded. Are they all subtypes of macrophages?
On Thu, Jan 9, 2025 at 8:53 PM gloriamj @.***> wrote:
If I only had single-cell RNAseq data from macrophages as a reference， however， I want to analysis the ratio of macrophage subtypes in TCGA bulk RNAseq. Can I use BayesPrism? I have tried to do this analysis, however, the total ratio is 1 in final result. In theory, bulk RNAseq also contains T cells, malignant cells, fibroblast and so on. Here are the results. image.png (view on web) https://github.com/user-attachments/assets/52ba44d9-796d-4626-9f24-d55ec9e37c00
— Reply to this email directly, view it on GitHub https://github.com/Danko-Lab/BayesPrism/issues/109, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB4NHSZPDX7M7I4TXKHPUY32J4RZZAVCNFSM6AAAAABU5MK642VHI2DSMVQWIX3LMV43ASLTON2WKOZSG43TSMBZGUYTIMA . You are receiving this because you are subscribed to this thread.Message ID: @.***>
Thank you for your interest in our methods. Just to make sure I understand. I noticed there are 11 columns in the figure you uploaded. Are they all subtypes of macrophages? … On Thu, Jan 9, 2025 at 8:53 PM gloriamj @.> wrote: If I only had single-cell RNAseq data from macrophages as a reference， however， I want to analysis the ratio of macrophage subtypes in TCGA bulk RNAseq. Can I use BayesPrism? I have tried to do this analysis, however, the total ratio is 1 in final result. In theory, bulk RNAseq also contains T cells, malignant cells, fibroblast and so on. Here are the results. image.png (view on web) https://github.com/user-attachments/assets/52ba44d9-796d-4626-9f24-d55ec9e37c00 — Reply to this email directly, view it on GitHub <#109>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB4NHSZPDX7M7I4TXKHPUY32J4RZZAVCNFSM6AAAAABU5MK642VHI2DSMVQWIX3LMV43ASLTON2WKOZSG43TSMBZGUYTIMA . You are receiving this because you are subscribed to this thread.Message ID: @.>
Yes, it is all of macrophage subcluters.
Sorry for the delay. The quick answer is no. We need to use a comprehensive scRNA-seq reference. Please refer to the answer in this thread https://github.com/Danko-Lab/BayesPrism/issues/95#issuecomment-2287841909
Sorry for the delay. The quick answer is no. We need to use a comprehensive scRNA-seq reference. Please refer to the answer in this thread #95 (comment)
Thank you for your reply! I got it. Therefore, this problem "Assuming that you have no or imcomplete representation of the tumor cells, yes, reads from the tumor cells in the bulk sample will inflate the cell type with similar transcription profile to the tumor cells" is unavoidable if all cell types in bulk RNAseq are not used as reference?