Comparing the impact of sample multiplexing approaches for single-cell RNA-sequencing on downstream analysis using cerebellar organoids

Abstract

Sample multiplexing provides a solution to limited sample throughput in single-cell RNA sequencing (scRNA-seq) experiments. Different strategies for multiplexing are commercially provided by Parse Biosciences combinatorial barcoding (Parse) and 10x Genomics CellPlex combined with microfluidic cell capture (10x). However, the extent to which these two techniques differ when characterizing complex tissues such as regionalized neural organoids and whether data generated from the two techniques can be readily integrated is unknown. Cerebellar organoids are a highly relevant model for understanding evolutionary differences, developmental trajectories, and disease mechanisms of this brain region. However, they have not been extensively characterized through scRNA-seq. Therefore, we compared the two multiplexing techniques, 10x and Parse, using cerebellar organoids derived from three stem cell lines. While both strategies demonstrated technical reproducibility and revealed comparable cellular diversity including the main lineages of cerebellar neurons, we found more stressed cells in 10x than in Parse. Additionally, we observed differences in transcript capture, with Parse covering a higher gene biotype diversity and less mitochondrial and ribosomal protein coding transcripts. In summary, we demonstrate that both techniques provide similar insight into cerebellar organoid biology, but flexibility of experimental design, capture of long transcripts, and the level of cell stress caused by the workflow differ.