Scalable single-cell metagenomic analysis with Bascet and Zorn

Abstract

Single-cell metagenomic sequencing (scMetaG) can provide maximum-resolution insights into complex microbial communities. However, existing bioinformatic tools are not equipped to handle the massive amounts of data generated by high-throughput scMetaG methods. Here, we present a bioinformatic toolkit for end-to-end scMetaG analysis: Bascet, a command-line suite designed to scale to massive scMetaG datasets ([≥]1 million cells); and Zorn, an R package/workflow manager that enables reproducible scMetaG data analysis, exploration, and visualization (http://zorn.henlab.org/). Enabled by recent advances in droplet microfluidics, we use Bascet and Zorn to develop and optimize a high-throughput scMetaG method on a ten-species mock community. To showcase their utility on a clinically relevant sample, we apply Bascet and Zorn to human saliva, generating single-amplified genomes from >15k prokaryotic cells, including uncultured, low-abundance taxa undetected by bulk metagenomics. Overall, Bascet and Zorn enable reproducible scMetaG analysis, allowing users to query microbiomes at unprecedented resolution and scale.