Large-scale estimation of bacterial and archaeal DNA prevalence in metagenomes reveals biome-specific patterns

Abstract Metagenomes often contain many reads derived from eukaryotes. However, there is usually no reliable method for estimating the prevalence of non-microbial reads in a metagenome, forcing many analysis techniques to make the often-faulty assumption that all reads are microbial. For instance, the success of metagenome-assembled genome (MAG) recovery efforts is assessed by the number of reads mapped to recovered MAGs, a procedure which will underestimate the true fidelity if eukaryotic reads are present. Here we present “SingleM microbial_fraction” (SMF), a scalable algorithm that robustly estimates the number of bacterial and archaeal reads in a metagenome, and the average microbial genome size. SMF does not use eukaryotic reference genome data and can be applied to any Illumina metagenome. Based on SMF, we propose the “Domain-Adjusted Mapping Rate” (DAMR) as an improved metric to assess microbial genome recovery from metagenomes. We benchmark SMF on simulated and real data, and demonstrate how DAMRs can guide genome recovery. Applying SMF to 136,284 publicly available metagenomes, we report substantial variation in microbial fractions and biome-specific patterns of microbial abundance, providing insights into how microorganisms and eukaryotes are distributed across Earth. Finally, we show that substantial amounts of human host DNA sequence data have been deposited in public metagenome repositories, possibly counter to ethical directives that mandate screening of these reads prior to release. As the adoption of metagenomic sequencing continues to grow, we foresee SMF being a valuable tool for the appraisal of genome recovery efforts, and the recovery of global patterns of microorganism distribution.