Seurat analysis report of BD Rhapsody TAS-Seq WTA data
Perform DBEC correction (mRNA leak-derived backgroud subtraction of gene expression) For DBEC correction, genes of which log2(x+1)-transformed maximum expression over 8 were selected, and biexponential transformation was performed by using FlowTrans package in R version 4.2.1 (2022-06-23). Next, normal mixture components (model E) was detected by using mclust package in R version 4.2.1 (2022-06-23), average expression of each components was calculated, and genes of which the maximum average expression over 5.5 were selected for background subtration. Then, if the components of which difference of the maximum average expression against maximal average expression under 5 were detected, the expression value of the components set as 0 for background subtracion.
Seurat pre-processing
Perform Seurat (version 4.1.0) analysis.
- https://satijalab.org/seurat/
Please see Seurat vignettes (PBMC3K tutorial) for detail explanation of the each step of Seurat analysis.
- https://satijalab.org/seurat/articles/pbmc3k_tutorial.html
Entire Seurat object was saved by using qs package with .qs extension. Please use qs package to read .qs file with R.
We first calculating metrics (percent.mito, percent,ribo) and filtered out mitochondrial gene-high cells (over 0.25), doublets, and not-hashtag-assigned cells.
Because BD Rhapsody TAS-Seq data is non-UMI data (raw count data), we perform global normalization to 1M tags and perform scaling by using Seurat ScaleData function with regressing out library size (total raw read count of each cells) as a confounding factor.
Basic plots of statistics
stats1
Figure 2. Density heatmap (pseudocolor) plot of statistics
stats2
Figure 3. Ridge plot, separate by Tags
(after filtering low-quality cells)
File locations
Plot files were stored at ./result/human_RAILD/Seurat/Seurat_plots/.
PCA and Jackstraw analysis
Next, PCA analysis was performed against 8564 of highly-variable genes identified by FindVariableFeatures (selection.method=mvp, mean.cutoff=c(0.1, Inf), dispersion.cutoff=c(1, Inf)) function in Seurat v4.1.0in R version 4.2.1 (2022-06-23).
Figure 3-1. Jackstraw plot
1:150 PCs were selected because the number of significant PC was over 150.
Jackstraw plot file was stored at ./result/human_RAILD/Seurat/Seurat_plots/human_RAILD_Jackstraw.png.
Clustering
Cell clustering was performed by using FindNeighbors and FindClusters function in Seurat v4.1.0 in R version 4.2.1 (2022-06-23).
1:100 PCs were used for clustering analysis.
Clustering resolution parameters were changed sequentially (from 1.0 to 3.5) and evaluate clustering stability by using silhouette score against FIt-SNE space.
Seurat clustering silhouette score plot
We assumed that resolution 1 was optimal because mean silhouette score of resolution 1 was maximal among calculated resolutions.
Figure 4. Mean silhouette scores of clustering in each resolution parameters.
Seurat clustering visualization by FIt-SNE plot
We use python implementation of FIt-SNE 1.1.1 (George C. Linderman et al Nat Methods 2019) through reticulate package in R version 4.2.1 (2022-06-23) for visualizaiton of clustering result, because it is better than commonly-used UMAP method in terms of calculation speed and visualization. FIt-SNE parameters are as follows: perplexity=25 df=0.9. Tag assignment and Clustering results of each resolution parameters were shown separately in tabs.
Figure 5. FIt-SNE visualization of Seurat clustering result
optimal clustering result
