DEN-IM: dengue virus genotyping from amplicon and shotgun metagenomic sequencing.

Dengue virus (DENV) represents a public health threat and economic burden in affected countries. The availability of genomic data is key to understanding viral evolution and dynamics, supporting improved control strategies. Currently, the use of high-throughput sequencing (HTS) technologies, which can be applied both directly to patient samples (shotgun metagenomics) and to PCR-amplified viral sequences (amplicon sequencing), is potentially the most informative approach to monitor viral dissemination and genetic diversity by providing, in a single methodological step, identification and characterization of the whole viral genome at the nucleotide level. Despite many advantages, these technologies require bioinformatics expertise and appropriate infrastructure for the analysis and interpretation of the resulting data. In addition, the many software solutions available can hamper the reproducibility and comparison of results. Here we present DEN-IM, a one-stop, user-friendly, containerized and reproducible workflow for the analysis of DENV short-read sequencing data from both amplicon and shotgun metagenomics approaches. It is able to infer the DENV coding sequence (CDS), identify the serotype and genotype, and generate a phylogenetic tree. It can easily be run on any UNIX-like system, from local machines to high-performance computing clusters, performing a comprehensive analysis without the requirement for extensive bioinformatics expertise. Using DEN-IM, we successfully analysed two types of DENV datasets. The first comprised 25 shotgun metagenomic sequencing samples from patients with variable serotypes and genotypes, including an in vitro spiked sample containing the four known serotypes. The second consisted of 106 paired-end and 76 single-end amplicon sequences of DENV 3 genotype III and DENV 1 genotype I, respectively, where DEN-IM allowed detection of the intra-genotype diversity. The DEN-IM workflow, parameters and execution configuration files, and documentation are freely available at https://github.com/B-UMMI/DEN-IM).

Author Correction: Critical steps in clinical shotgun metagenomics for the concomitant detection and typing of microbial pathogens.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

Streptococcus pneumoniae Serotype 3 in Mexico (1994 to 2017): Decrease of the Unusual Clonal Complex 4909 Lineage following PCV13 Introduction.

Streptococcus pneumoniae expressing serotype 3 has a high virulence and a high case fatality ratio. Most studies of serotype 3 pneumococci have focused on a single lineage, the widespread sequence type 180 (ST180). To evaluate the serotype 3 lineages causing infections in Mexico, we characterized 196 isolates recovered from 1994 to 2017. The isolates were mostly susceptible to all antimicrobials tested. A single meningitis isolate was resistant to penicillin, and the resistance to erythromycin was 5.2%. The isolates represented the widely disseminated clonal complex 180 (CC180; n = 140), the unusual CC4909 (n = 42), CC260 (n = 11), and a few singletons (n = 3). CC260 was less frequent among pneumococcal invasive disease isolates than CC180 and CC4909 (P = 0.015). There was a decrease of CC4909 (P < 0.001) following PCV13 introduction (2012 to 2017). The CC4909 isolates were represented mostly by ST1119 (n = 40), seemingly having a restricted geographic origin, with isolates in the PubMLST database having been recovered only in Mexico, the United States, and Germany. A genomic analysis of publicly available genomes showed that ST1119 isolates have less than 32% similarity with ST180 isolates, indicating that these lineages are more separated than revealed by traditional multilocus sequence typing. Considering the suggestions of a lower efficacy of the 13-valent pneumococcal conjugate vaccine against serotype 3, the different dynamics of the two major serotype 3 lineages in Mexico following the introduction of PCV13 should be closely monitored.

Critical steps in clinical shotgun metagenomics for the concomitant detection and typing of microbial pathogens.

High throughput sequencing has been proposed as a one-stop solution for diagnostics and molecular typing directly from patient samples, allowing timely and appropriate implementation of measures for treatment, infection prevention and control. However, it is unclear how the variety of available methods impacts the end results. We applied shotgun metagenomics on diverse types of patient samples using three different methods to deplete human DNA prior to DNA extraction. Libraries were prepared and sequenced with Illumina chemistry. Data was analyzed using methods likely to be available in clinical microbiology laboratories using genomics. The results of microbial identification were compared to standard culture-based microbiological methods. On average, 75% of the reads corresponded to human DNA, being a major determinant in the analysis outcome. None of the kits was clearly superior suggesting that the initial ratio between host and microbial DNA or other sample characteristics were the major determinants of the proportion of microbial reads. Most pathogens identified by culture were also identified through metagenomics, but substantial differences were noted between the taxonomic classification tools. In two cases the high number of human reads resulted in insufficient sequencing depth of bacterial DNA for identification. In three samples, we could infer the probable multilocus sequence type of the most abundant species. The tools and databases used for taxonomic classification and antimicrobial resistance identification had a key impact on the results, recommending that efforts need to be aimed at standardization of the analysis methods if metagenomics is to be used routinely in clinical microbiology.

Beta-hemolytic Streptococcus dysgalactiae strains isolated from horses are a genetically distinct population within the Streptococcus dysgalactiae taxon.

The pathogenic role of beta-hemolytic Streptococcus dysgalactiae in the equine host is increasingly recognized. A collection of 108 Lancefield group C (n = 96) or L (n = 12) horse isolates recovered in the United States and in three European countries presented multilocus sequence typing (MLST) alleles, sequence types and emm types (only 56% of the isolates could be emm typed) that were, with few exceptions, distinct from those previously found in human Streptococcus dysgalactiae subsp. equisimilis. Characterization of a subset of horse isolates by multilocus sequence analysis (MLSA) and 16S rRNA gene sequence showed that most equine isolates could also be differentiated from S. dysgalactiae strains from other animal species, supporting the existence of a horse specific genomovar. Draft genome information confirms the distinctiveness of the horse genomovar and indicates the presence of potentially horse-specific virulence factors. While this genomovar represents most of the isolates recovered from horses, a smaller MLST and MLSA defined sub-population seems to be able to cause infections in horses, other animals and humans, indicating that transmission between hosts of strains belonging to this group may occur.