Getting unspliced counts for scRNA-seq data generated from cellranger aggr pipeline

Question

[Luna_P]

Hi everyone,

I have 15 scRNA-seq samples that I have aggregated with cellranger aggr pipeline. My problem is that, I would like to use scVelo on my dataset. However, I will need the unspliced counts for this. I have wanted to run Velocyto but it requires bam files and cellranger aggr pipeline does not have the bam as an output. From the examples I have seen, if I want to get the unspliced counts I either have to use the fastq files for this or simply generate a loom file from all the samples individually and somehow aggregate them. I was wondering if there is a way for me to do this with the output generated from cellranger aggr ?

Thank you !

Answer 1

[fracarb8]

You should add some details on how you ran cellranger aggr.

Usually, before aggregating you run cellranger count, and you then aggregate the outputs together. Doing so, you will have the outs directories with the BAM files in it.

The 10x website has extensive documentation on the matter.

[Luna_P]

I aggregated the outputs from cell ranger count. So all my samples have an individual bam files generated from the cell ranger count. I already have individual .loom files for each sample. However, when you aggregate the outputs from cell ranger count the aggr pipeline doesn't generate a .bam file for all of these samples as far as my outputs and the website suggest if I am not missing anything.

I am simply wondering if instead of creating all these individual loom files I can find a way of calculating unspliced counts from the aggr outputs.

[fracarb8]

If you are running velocyto, you can merge different samples in one loom file by using loompy. Have a look at the proper section of the documentation.