Hashing Tutorial Using Feature Barcoding 10x Cellranger Count Pipeline

[erichpeterson]

Hi,
We are trying to perform an analysis on three mice samples, in which each of three mice was tagged with a unique antibody, and then the samples were pooled and sequenced. A gene expression and a "feature barcoding" experiment was run per 10x guidelines. At the time of running, we didn't know that the 10x Cellranger count does not support hashing.

We are trying to get the Seurat tools to work with this data and using the vignette found here as a template: https://satijalab.org/seurat/archive/v3.1/hashing_vignette.html

We were just wanting to see if the steps we are taking below seem appropriate given the situation:

1. Ran the samples via the Cellranger mkfastq as normal
2. Ran Cellranger count using the "feature barcoding" method found here: https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis (providing library CSV with the fastq files for the gene expression and feature barcoding samples, as well as the feature reference CSV).
3. The output directory "filtered_feature_bc_matrix" containing the barcodes.tsv.gz, features.tsv.gz, and matrix.mtx.gz were used as input for the tutorial at https://satijalab.org/seurat/archive/v3.1/hashing_vignette.html
4. The vignette was run as-is, except for one notable change, which is the crux of our question: it appears as though after running Read10X() the gene expression data is placed in the first element of a list and the feature barcoding portion is placed in the second. So, we edited the vignette to look something like this:

htos <- Read10X("./HTO")
umis <- htos[[1]]
htos <- htos[[2]]

Is this the way it should be done?

Thanks

[timoast]

Yes that's correct