New publication on standards for microbiome analysis by metagenomics in Nature Microbiology

DOME | JMF | CeMESS

Joana Séneca, Petra Pjevac and David Berry from the Joint Microbiome Facility (JMF) and DOME | CeMESS (University of Vienna) contributed to a new study, led by scientists at Uniklinik RWTH Aachen, benchmarking a key method used to characterize the microbial contents of our digestive tracts.

The JMF team worked with Univ.-Prof. Dr. Thomas Clavel, Professor for Intestinal Microbiology at Uniklinik RWTH Aachen, to rigorously assess the potentials and limits of metagenomic sequencing, a key method in microbiome research. The study, carried out in collaboration with further colleagues from Maastricht (Netherlands), is entitled “Benchmarking of shotgun sequencing depth reveals the potential and limitations of shallow metagenomics and strain-level analysis” and was published in the journal Nature Microbiology. 

The team investigated different complex mixtures of bacterial DNA with varying distributions of different bacteria. Sequencing these mixtures, where the composition of microbes was known, allowed the researchers to determine what information can be gained at what sequencing depth for which community complexity. Importantly, determining the taxonomic diversity of a complex community is shown to be possible even at a comparatively shallow sequencing depth (<1 GB per sample) when a reference genome database is available, an important consideration due to the costs of deeper sequencing in large-scale studies. 

Uni Wien/Der Knopfdruecker

Joana Séneca loading the nanopore with samples.

However, reconstructing the entire genetic material of single bacterial strains, also referred to as into metagenome-assembled genomes (MAGs), requires a sequencing depth that is 10 times higher than what is needed to determine the taxonomic diversity, and even at this depth prone to generating artefacts. The analysis also showed what sequencing depths are required to characterize the functional potential of microbiome members, with a 5-fold higher depth required to identify individual proteins compared to simply describing processes occurrence. The authors further found that while different protocols and high amounts of background DNA may confound results, this can be mitigated by sequencing samples at higher depth.

Prof. Clavel states: “Diversity between strains contributes to functional differences in microbiomes between individuals. So, there is great interest in reconstructing the genomes of individual microbiome members. However, based on our results we can conclude that reconstructing MAGs is a complicated endeavour, even at high sequencing depths. In fact, only 50–80% of these sequences are correct, depending on the bioinformatic approach used.”

Links