Staphylococcus aureus is a multi-host pathogen that causes infections in animals and humans globally. The specific genetic loci—and the extent to which they drive cross-species switching, transmissibility, and adaptation—are not well understood. Here, we conducted a population genomic study of 437 S. aureus isolates to identify bacterial genetic variation that determines infection of human and animal hosts through a genome-wide association study (GWAS) using linear mixed models. We found genetic variants tagging φSa3 prophage-encoded immune evasion genes associated with human hosts, which contributed ∼99.9% of the overall heritability (∼88%), highlighting their key role in S. aureus human infection. Furthermore, GWAS of pairs of phylogenetically matched human and animal isolates confirmed and uncovered additional loci not implicated in GWAS of unmatched isolates. Our findings reveal the loci that are critical for S. aureus host transmissibility, infection, switching, and adaptation and how their spread alters the specificity of host-adapted clones.

Bloodstream infections caused by the opportunistic pathogen Klebsiella pneumoniae are associated with adverse health complications and high mortality rates. Antimicrobial resistance (AMR) limits available treatment options, thus exacerbating its public health and clinical burden. Here, we aim to elucidate the population structure of K. pneumoniae in bloodstream infections from a single medical center and the drivers that facilitate the dissemination of AMR. Analysis of 136 short-read genome sequences complemented with 12 long-read sequences shows the population consisting of 94 sequence types (STs) and 99 clonal groups, including globally distributed multidrug resistant and hypervirulent clones. In vitro antimicrobial susceptibility testing and in silico identification of AMR determinants reveal high concordance (90.44–100%) for aminoglycosides, beta-lactams, carbapenems, cephalosporins, quinolones, and sulfonamides. IncF plasmids mediate the clonal (within the same lineage) and horizontal (between lineages) transmission of the extended-spectrum beta-lactamase gene bla_CTX-M-15. Nearly identical plasmids are recovered from isolates over a span of two years indicating long-term persistence. The genetic determinants for hypervirulence are carried on plasmids exhibiting genomic rearrangement, loss, and/or truncation. Our findings highlight the importance of considering both the genetic background of host strains and the routes of plasmid transmission in understanding the spread of AMR in bloodstream infections.

Background: Bloodstream infections due to Staphylococcus aureus cause significant patient morbidity and mortality worldwide. Of major concern is the emergence and spread of methicillin-resistant S. aureus (MRSA) in bloodstream infections, which are associated with therapeutic failure and increased mortality. Methods: We generated high quality draft genomes from 323 S. aureus blood culture isolates from patients diagnosed with bloodstream infection at the Dartmouth-Hitchcock Medical Center, New Hampshire, USA in 2010–2018. Results: In silico detection of antimicrobial resistance genes revealed that 133/323 isolates (41.18%) carry horizontally acquired genes conferring resistance to at least three antimicrobial classes, with resistance determinants for aminoglycosides, beta-lactams and macrolides being the most prevalent. The most common resistance genes were blaZ and mecA, which were found in 262/323 (81.11%) and 104/323 (32.20%) isolates, respectively. Majority of the MRSA (102/105 isolates or 97.14%) identified using in vitro screening were related to two clonal complexes (CC) 5 and 8. The two CCs emerged in the New Hampshire population at separate times. We estimated that the time to the most recent common ancestor of CC5 was 1973 (95% highest posterior density (HPD) intervals: 1966–1979) and 1946 for CC8 (95% HPD intervals: 1924–1959). The effective population size of CC8 increased until the late 1960s when it started to level off until late 2000s. The levelling off of CC8 in 1968 coincided with the acquisition of SCCmec Type IV in majority of the strains. The plateau in CC8 also coincided with the acceleration in the population growth of CC5 carrying SCCmec Type II in the early 1970s, which eventually leveled off in the early 1990s. Lastly, we found evidence for frequent recombination in the two clones during their recent clonal expansion, which has likely contributed to their success in the population. Conclusions: We conclude that the S. aureus population was shaped mainly by the clonal expansion, recombination and co-dominance of two major MRSA clones in the last five decades in New Hampshire, USA. These results have important implications on the development of effective and robust strategies for intervention, control and treatment of life-threatening bloodstream infections.

Klebsiella oxytoca is an opportunistic pathogen causing serious nosocomial infections. Knowledge about the population structure and diversity of healthcare-associated K. oxytoca from a genomic standpoint remains limited. Here, we characterized the phylogenetic relationships and genomic characteristics of 20 K . oxytoca sensu stricto isolates recovered from bloodstream infections at the Dartmouth-Hitchcock Medical Center, New Hampshire, USA from 2017 to 2021. Results revealed a diverse population consisting of 15 sequence types (STs) that together harbored 10 variants of the intrinsic beta-lactamase gene bla _OXY-2 , conferring resistance to penicillins. Similar sets of antimicrobial resistance (AMR) determinants reside in multiple distinct lineages, with no one lineage dominating the local population. To place the New Hampshire K. oxytoca in a broader context, we compared them to 304 publicly available genomes of clinical isolates from 18 countries. This global clinical K. oxytoca sensu stricto population is represented by over 65 STs that together harbored resistance genes against 14 antimicrobial classes, including eight bla _OXY-2 variants. Three dominant STs in the global population (ST2, ST176, ST199) circulate across multiple countries and were also present in the New Hampshire population. The global K. oxytoca population is genetically diverse, but there is evidence for broad dissemination of a few lineages carrying distinct set of AMR determinants. Our findings reveal the clinical diversity of K. oxytoca sensu stricto and its importance in surveillance efforts aimed at monitoring the evolution of this drug-resistant nosocomial pathogen. IMPORTANCE: The opportunistic pathogen Klebsiella oxytoca has been increasingly implicated in patient morbidity and mortality worldwide, including several outbreaks in healthcare settings. The emergence and spread of antimicrobial resistant strains exacerbate the disease burden caused by this species. Our study showed that clinical K. oxytoca sensu stricto is phylogenetically diverse, harboring various antimicrobial resistance determinants and bla _OXY-2 variants. Understanding the genomic and population structure of K. oxytoca is important for international initiatives and local epidemiological efforts for surveillance and control of drug-resistant K. oxytoca .

Staphylococcus aureusin the bloodstream causes high morbidity and mortality, exacerbated by the spread of multidrug-resistant and methicillin-resistantS. aureus(MRSA). We aimed to characterize the circulating lineages ofS. aureusfrom bloodstream infections and the contribution of individual lineages to resistance over time. Here, we generated 852 high-quality short-read draft genome sequences ofS. aureusisolates from patient blood cultures in a single hospital from 2010 to 2022. A total of 80 previously recognized sequence types (ST) and five major clonal complexes are present in the population. Two frequently detected lineages, ST5 and ST8 exhibited fluctuating demographic structures throughout their histories. The rise and fall in their population growth coincided with the acquisition of antimicrobial resistance, mobile genetic elements, and superantigen genes, thus shaping the accessory genome structure across the entire population. These results reflect undetected selective events and changing ecology of multidrug-resistantS. aureusin the bloodstream.

This study introduces a new framework for the artificial intelligence-based characterization of Gram-stained whole-slide images (WSIs). As a test for the diagnosis of bloodstream infections, Gram stains provide critical early data to inform patient treatment in conjunction with data from rapid molecular tests. In this work, we developed a novel transformer-based model for Gram-stained WSI classification, which is more scalable to large data sets than previous convolutional neural network-based methods as it does not require patch-level manual annotations. We also introduce a large Gram stain data set from Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire, USA) to evaluate our model, exploring the classification of five major categories of Gram-stained WSIs: gram-positive cocci in clusters, gram-positive cocci in pairs/chains, gram-positive rods, gram-negative rods, and slides with no bacteria. Our model achieves a classification accuracy of 0.858 (95% CI: 0.805, 0.905) and an area under the receiver operating characteristic curve (AUC) of 0.952 (95% CI: 0.922, 0.976) using fivefold nested cross-validation on our 475-slide data set, demonstrating the potential of large-scale transformer models for Gram stain classification. Results were measured against the final clinical laboratory Gram stain report after growth of organism in culture. We further demonstrate the generalizability of our trained model by applying it without additional fine-tuning on a second 27-slide external data set from Stanford Health (Palo Alto, California, USA) where it achieves a binary classification accuracy of 0.926 (95% CI: 0.885, 0.960) and an AUC of 0.8651 (95% CI: 0.6337, 0.9917) while distinguishing gram-positive from gram-negative bacteria. IMPORTANCE: This study introduces a scalable transformer-based deep learning model for automating Gram-stained whole-slide image classification. It surpasses previous methods by eliminating the need for manual annotations and demonstrates high accuracy and generalizability across multiple data sets, enhancing the speed and reliability of Gram stain analysis.