Hi all,
I have experience with non-hashtagged single-cell data (and with bulk RNA-seq), but it's my first time with cell hashtagging using TotalSeq-C. I have 50 total PBMC samples (hashed 5 samples each). After sequencing, I get 10 total sample fastqs (e.g., "xxx_SB-1_FB_S1_L005_R1_001.fastq.gz" and "xxx_SB-1_S1_L005_R1_001.fastq.gz", etc.). My questions are as follows:
Do I essentially run each fastq pair (containing counts for 5 samples each) through CellRanger with --expect-cells=30000? My understanding is that the fastqs labeled "FB-S1" contain the Feature Barcodes that will ultimately help me demultiplex the sames later in Seurat? The 10X tutorial is not quite as helpful as I was hoping it could be: https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#pattern
Assuming the above is true and I generate essentially 10 BAM files (for the total 50 samples), will the fact that they're hashtagged affect 1) read trimming homopolymers and the template switch oligo and 2) generating spliced/unspliced reads for downstream RNA velocity analyses.
Similarly, I assume that feeding the CellRanger outputs into CellBender and similar programs for ambient RNA correction shouldn't be affected.
Any helpful suggestions or pointing towards tutorials/example code would be much appreciated, especially code for running CellRanger for the raw fastq files. Thanks!