Selected abstract: clinical metagenomics in endophthalmitis

Selected abstract: clinical metagenomics in endophthalmitis


Endophthalmitis is a complication of eye surgeries and can lead to loss of vision. It is an inflammation that develops in response to microorganisms entering the eye. A rapid determination of the causing agents and their antibiotic resistance would facilitate precise treatments and reduce blindness.



We examine the microbiome of 14 vitreous (intraocular body fluid) samples from patients with endophthalmitis. As controls we include 17 samples (vitreous from endophthalmitis-negative patients; aliquots of basal salt solution injected into the eye during vitrectomy; DNA extraction blanks). We evaluate two DNA isolation protocols, and sequenced the 62 samples using Illumina MiSeq sequencing technology. In comparison, we performed cultivations of vitreous, and analyse isolates using WGS, MALDI-TOF, and phenotypic antimicrobial susceptibility testing.



For metagenomics data analysis of these low-biomass samples we designed a bioinformatics workflow with filtering steps to reduce DNA sequences originating from: i) human hosts, ii) ambiguousness/contaminants in public microbial reference genomes, and iii) the environment. Our metagenomic read classification revealed in nearly all cases the same microorganism than was determined in cultivation- and mass spectrometry-based analyses. We identified the sequence type of the microorganism and antibiotic resistance genes through analyses of WGS assemblies, and metagenomic assemblies. In support of open science we share our detailed methods and results, sequencing data, code for curated database generation, guidelines for metagenomics projects, and video summary.



Our findings suggest that metagenomics analysis together with WGS-based analysis is suitable for the identification of the infectious agents from human ocular body fluid and could guide therapeutic strategies including targeted antimicrobial therapy and the choice of steroids. Prerequisites for a robust data analysis are suitable procedures that facilitate the isolation of nucleic acid from microorganisms residing in complex samples, the analysis of relevant control samples, as well as high-quality genome sequence reference databases.