Benchmarking shallow metagenomics using synthetic communities. null

Next-generation sequencing techniques are essential for microbiome research. However, their widespread use contrasts to the low number of studies testing their robustness and accuracy. Shallow metagenomics (sMG) is an emerging technique to overcome the main limitations of the two most common approaches, 16S rRNA gene amplicon (amplification bias, low taxonomic resolution) and deep shotgun sequencing (high costs, demanding bioinformatic analysis). Here we systematically assessed the effects of shotgun sequencing depth on taxonomic and functional readouts using comprehensive datasets from synthetic microbial communities of multiple complexities. Four synthetic communities of varying diversity (24-70 strains) and two abundance profiles (equal vs. staggered) were sequenced at nine depths (0.1-10 Gb/sample), including library preparation in two different facilities (Aachen and Maastricht). Background effects were tested by spiking DNA extracted from the intestinal content of germfree mice. The genomes of all strains were used as references during analysis, enabling both de-novo and targeted analysis.