Whole-genome sequencing of Mycobacterium tuberculosis directly from clinical samples for high-resolution genomic epidemiology and drug resistance surveillance: an observational study

Abstract

Background Direct whole-genome sequencing of Mycobacterium tuberculosis from clinical specimens will be a major breakthrough in tuberculosis diagnosis and control. To date, direct whole-genome sequencing has never been used in genomic epidemiology, and its accuracy in transmission inference remains unknown. We investigated the technical challenges imposed by direct whole-genome sequencing, and used it to infer transmission clusters and predict drug resistance.Methods Using an optimised workflow, we did direct whole-genome sequencing for 37 clinical specimens from 23 tuberculosis patients. Nine sputum samples from nine patients who were infected with different non-tuberculous mycobacteria and culture-negative for tuberculosis were used as controls in the qPCR assays and pre-sequencing runs. Additionally, 780 clinical isolates in the region of Comunidad Valenciana (Spain) were whole-genome sequenced between Jan 1, 2014, and Dec 31, 2016. We analysed the genomic variants to build a tuberculosis transmission network for the region, including the clinical specimens, and to predict drug susceptibility profiles.Findings After sequencing 37 clinical specimens, 28 specimens (22 [85%] of 26 smear-positive and six [55%] of 11 smear-negative) met the quality criteria for downstream analysis. All 28 clinical specimens clustered with their matching culture isolates, with a median distance of 0 single nucleotide polymorphisms. Of the 28 clinical specimens, 16 (57%) were accurately assigned to ten transmission clusters in the region, and 12 (43%) were unique cases. Transmission inferences and drug-susceptibility predictions from direct whole-genome sequencing data were concordant with sequences from corresponding cultures and phenotypic drug-susceptibility testing. Complete genomic analysis, within a week of specimen receipt, cost €217 per sample (excluding personnel costs).Interpretation Direct whole-genome sequencing could be used to accurately delineate transmission clusters of tuberculosis and conduct culture-independent surveillance. Compared with conventional approaches, direct wholegenome sequencing allows researchers to do real-time genomic epidemiology and drug resistance surveillance in settings where culture and drug susceptibility testing are not available.

Capture and visualization of live Mycobacterium tuberculosis bacilli from tuberculosis patient bioaerosols

“…Although we did not detect any clear alterations in morphology or DMN-trehalose incorporation profile, the incubation period was limited to two weeks which might be too short to allow for metabolic and/or replicative adaptation in the captured organisms; ongoing work is extending the duration of incubation, and will also explore the use of other compartmentalized capture devices as alternatives to the COC-embossed nanowell format. At the very least, these results support the ability to capture and maintain single Mtb cells in an arrayed format for serial imaging, and hint at the potential to exploit the physical separation of cells into nanowell compartments for clonal propagation of bacilli downstream for genomic and other analyses which require biomass [33].…”

Mortality from drug-resistant tuberculosis in high-burden countries comparing routine drug susceptibility testing with whole-genome sequencing: a multicentre cohort study

“…34 Such tests would allow rapid diagnosis and efficient, individual-based treatment of drug-resistant tuberculosis. 35 Test systems performing WGS on sputum samples, using new laboratory and bioinformatics pipelines are in development. High-burden countries should consider building central, high-throughput sequencing capacities.…”

Comprehensive and accurate genetic variant identification from contaminated and low coverage Mycobacterium tuberculosis whole genome sequencing data

“…It was not possible to represent this endless variety of bacterial contaminants in our simulation experiments hence the most commonly observed bacterial contaminants where used instead [2], NTMs were included because these pose a serious challenge for Mtb variant calling. Contaminant levels were simulated to match levels observed for genuine sputum samples [8]. As such it was possible to study the effect of the various contaminants and their levels on variant calling, this would not have been possible had all the endless contaminants observed for direct-from-sputum samples been used.…”

“…To simulate Mtb WGS data obtained directly from sputum samples, another six datasets were generated, each with a set number of paired-end sequence reads per sample, ranging from 500,000 to 3,000,000 PE reads (Figure 2). Mtb and contamination levels were randomly sampled from a beta distribution around 0.01 to 78.63% Mtb DNA to match levels observed for direct-from-sputum WGS data [8]. The contamination type was either P. aeruginosa, S. epidermidis, H. sapiens , a mixture of P. aeruginosa, S. epidermidis and H. sapiens or a NTM mixture ( M. intracellulare, M. abcessus or M. avium up to 20% of the Mtb fraction with the remaining contamination consisting of P. aeruginosa, S. epidermidis or H. sapiens ).…”

“…Data published by Roetzer et al [16] was used to test the pipelines’ sensitivity by identifying a set of known variants (Sanger confirmed) in DNA extracted from cultured Mtb samples. Data from Goig et al [8] was used to evaluate the ability to call variants in WGS data from DNA extracted directly from sputum. Only samples with ≥5× genomic coverage depth (S02, S26, S17, S21, S31, S20, S27, S67, S09 and S69) were analysed to ensure sufficient width of coverage and prevent problems with phylogenetic inference.…”