Direct nanopore sequencing of M. tuberculosis on sputa and rescue of suboptimal results to enhance transmission surveillance

Whole-genome sequencing (WGS) has enhanced precision to predict antimicrobial resistance and track transmission in Mycobacterium tuberculosis (MTB). Due to the slow-growing nature of MTB, genomic results are unavoidably delayed; however, few efforts have been made to accelerate them, by performing WGS directly on respiratory specimens. Most culture-free efforts in the literature have focused on the acceleration of resistance prediction. The present study provides further evidence to the only preceding study aiming to accelerating the precise delineation of transmission, coupling culture-free WGS to a surveillance programme. Our study is distinguished from its predecessor by being the first to apply flexible nanopore sequencing for the purpose of further accelerating the process. A total of 71 sputa were selected for the study, in which we only applied a procedure to deplete human DNA, thus avoiding other costly and cumbersome alternatives based on capture baits. Optimal results (>90% of genome covered, mean coverage >45× and >70% genome covered >20×) were obtained from 33.8% of the cases, which allowed the corresponding cases to be ruled in or out of transmission clusters, close to their diagnosis. A further 12.6% of the samples yielded suboptimal results (15.5% - 90.92% at >10×), which could also be exploited thanks to our rescue pipeline; the rationale was to attempt to identify, from the suboptimal sequences, the informative SNPs that are markers for relevant transmission clusters in our population. The rescue pipeline facilitated pre-allocation of new cases to pre-existing clusters and moreover enabled the specification of genomic relationships between new cases and preceding ones in the cluster. In summary, the present study has demonstrated that epidemiologically valuable information can be obtained directly from sputum in approximately half of the samples analysed. This study represents a new advancement in the pursuit of enhancing faster comparative genomics, at diagnosis, for epidemiological applications.