Progressively overloaded single-cell RNA-seq with cell hashing

Barcode-based multiplexing methods can be used to increase throughput and reduce batch effects in large single-cell genomics studies. To evaluate methods for demultiplexing barcode-multiplexed data, we generated a dataset by labeling samples separately with barcode-tagged antibodies, mixing those samples, and progressively overloading a droplet-based scRNA-seq system. An aliquot of Ficoll-purified human PBMCs was sorted into 3 distinct populations by FACS(Naive T cells, Memory T cells, and Non-T cells), each population was divided into two replicates, and each replicate was stained with a different barcoded antibody (BioLegend TotalSeq-A). Samples were then mixed, and loaded into a Chromium Next GEM Single Cell 3' v3.1 scRNA-seq chip (10x Genomics) at 16,000 to 80,000 cells per well. After library prepraration, well libraries were mixed based on overloading level to obtain an increased number of reads with increased overloading. NOTE FROM SUBMITTER: Raw files from human subjects will be submitted to dbGaP