Follow-up on Bulk RNA-Seq Normalization Methods Using Seurat

[fmbetul]

Hi Seurat Team,

I recently came across the closed issue #826 and have been using Seurat for bulk RNA-seq analysis as well. The discussion in the issue was very helpful, and I wanted to ask a follow-up question.

Initially, I followed the approach suggested in the thread:

combined_seurat <- CreateSeuratObject(counts = combined_tpm_data, project = "combined_TPM", meta.data = combined_tpm_metadata)  
combined_seurat@assays$RNA$data <- log(combined_seurat@assays$RNA$counts + 1)  
combined_seurat <- FindVariableFeatures(combined_seurat)  
combined_seurat <- ScaleData(combined_seurat)  
combined_seurat <- RunPCA(combined_seurat)  

The PCA plot from this approach reflected clustering patterns aligned with our biological expectations.

I then tested using the NormalizeData function, commonly used for single-cell RNA-seq data:

combined_seurat_v2 <- CreateSeuratObject(counts = combined_tpm_data, project = "combined_TPM", meta.data = combined_tpm_metadata)  
combined_seurat_v2 <- NormalizeData(combined_seurat_v2)  
combined_seurat_v2 <- FindVariableFeatures(combined_seurat_v2)  
combined_seurat_v2 <- ScaleData(combined_seurat_v2)  
combined_seurat_v2 <- RunPCA(combined_seurat_v2)  

Interestingly, the PCA plot generated from this approach appears to better separate our sample groups.

I am unsure if using the NormalizeData method, which is intended for single-cell analysis, is appropriate for bulk RNA-seq data. Would this be considered a legitimate approach, or are there concerns I should be aware of?

Below is my sessionInfo() for reference:

R version 4.3.2 (2023-10-31)  
Platform: aarch64-apple-darwin20 (64-bit)  
Running under: macOS Sonoma 14.7  

attached base packages:  
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:  
[1] Seurat_5.0.1       SeuratObject_5.0.1  

loaded via a namespace (and not attached):  
[1] Matrix_1.6-5       ggplot2_3.5.1      dplyr_1.1.4        data.table_1.14.10  
[5] Rcpp_1.0.12        future_1.33.1      lattice_0.21-9     patchwork_1.2.0  

I’d greatly appreciate your insights. Thank you for your time!

Best regards,
Betul