Assessing cross-genotype contamination in pooled snRNAseq data using per-read donor assignments

[marusakod]

Hi, I am analysing snRNAseq data comprising pooled samples from 8 different donors, with whole-genome sequencing (WGS) data available for each donor. My goal is to evaluate the effectiveness of ambient RNA removal by detecting cross-genotype contamination. To achieve this, I would like to determine the origin of each read i.e. which donor each read belongs to. A similar approach is described in this paper: Genotype-based demultiplexing of single-nucleus RNA-seq data. My expectation is that this should be possible at least for a subset of reads.

I am particularly interested in whether demuxlet can provide per-donor read assignments, or if such assignments are generated as an intermediate step during the processing and can be accessed.

Best,
Maruša