Dear community members,
Given the widespread and increasing use of sample multiplexing in sequencing experiments, especially in scRNAseq applications, and the potentially data comprising effects of sample index hopping that multiplexing brings about, we felt that sharing our new work with you would potentially alert you to potential problems that you may encounter in your own research, especially if it involves the generation or the analysis of sample-multiplexed sequencing data. The paper is entitled
Statistical modeling, estimation, and remediation of sample index hopping in multiplexed droplet-based single-cell RNA-seq data. (R. Farouni H. S. Najafabadi, 2019)
and is available as a preprint on bioRxiv. Although we show that the read hopping rate is low, the proportion of phantom molecules can vary widely across samples. For example, it ranges from 1% to 18% in a Novaseq 6000 multiplexed Tabula Muris dataset. We hope that once the approach proposed in the paper is vetted by the community that 10x Genomics would adopt or adapt the proposed ideas and methods in their own pipeline. We appreciate any feedback or comments you might have for us, no matter how critical they might be.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.