Challenges of the Unknown: Clinical Application of Microbial Metagenomics

Abstract

Availability of fast, high throughput and low cost whole genome sequencing holds great promise within public health microbiology, with applications ranging from outbreak detection and tracking transmission events to understanding the role played by microbial communities in health and disease. Within clinical metagenomics, identifying microorganisms from a complex and host enriched background remains a central computational challenge. As proof of principle, we sequenced two metagenomic samples, a known viral mixture of 25 human pathogens and an unknown complex biological model using benchtop technology. The datasets were then analysed using a bioinformatic pipeline developed around recent fast classification methods. A targeted approach was able to detect 20 of the viruses against a background of host contamination from multiple sources and bacterial contamination. An alternative untargeted identification method was highly correlated with these classifications, and over 1,600 species were identified when applied to the complex biological model, including several species captured at over 50% genome coverage. In summary, this study demonstrates the great potential of applying metagenomics within the clinical laboratory setting and that this can be achieved using infrastructure available to nondedicated sequencing centres.

Efficient and unbiased metagenomic recovery of RNA virus genomes from human plasma samples

“…A Φ29-based Multiple Displacement Amplification (MDA) system has been successful in generating whole genomes of HIV from low copy number samples. However, these were prepared by diluting high titre clinical HIV samples in PBS, such that the impact of high host background was mitigated 38 , and in general, MDA displays target amplification biases that limits its potential in metagenomics to detection and identification rather than whole genome reconstruction [39][40][41] .In this study, we have established a sequence-independent RNA library preparation method suitable for the detection and characterization of blood-borne RNA viruses. The method is focused on increasing the relative abundance of viral RNA within the sample, during and after the RNA extraction process, with a specialised library preparation step able to process ultra-low RNA inputs.…”

FastQ Screen: A tool for multi-genome mapping and quality control

“…This has numerous applications, most typically to remove DNA contaminants, as exemplified by a recent clinical microbial metagenomics study in which nucleic acids were extracted from porcine faeces 9 . FastQ Screen was then used to filter-out host sequences, and the remaining reads were then mapped, leading to the identification of over 1,600 bacterial and Archaea species and strains of virus.…”

High potential for temperate viruses to drive carbon cycling in chemoautotrophy‐dominated shallow‐water hydrothermal vents

“…The resulting coverage information (seq count) for each proviral genome detected in mat and non-mat sediment metagenomes was used to construct taxonomic tables based on i) the number of seq matches to the proviruses identified, classified at the viral family level and ii) based on the number of proviral genomes displaying positive matches with mat and non-mat metagenomic sequences. The first was used to provide information on the relative abundance of the different proviral families in mat and non-mat sediments, the second as an estimate of the diversity of proviral DNA sequences in the different samples (Rose et al, 2015;Bendall et al, 2016). BBMap was preferred for the analysis of proviral DNA sequences in mat and non-mat metagenomes owing to ease in mapping with low-percent identity reads, as also reported by Bendall et al, 2016.…”