Klebsiella

The HiSeq and MiSeq metagenomes were built using 20 sets of bacterial whole-genome shotgun reads. These reads were found either as part of the GAGE-B project [21 (link)] or in the NCBI Sequence Read Archive. Each metagenome contains sequences from ten genomes (Additional file 1: Table S1). For both the 10,000 and 10 million read samples of each of these metagenomes, 10% of their sequences were selected from each of the ten component genome data sets (i.e., each genome had equal sequence abundance). All sequences were trimmed to remove low quality bases and adapter sequences.
The composition of these two metagenomes poses certain challenges to our classifiers. For example, Pelosinus fermentans, found in our HiSeq metagenome, cannot be correctly identified at the genus level by Kraken (or any of the other previously described classifiers), because there are no Pelosinus genomes in the RefSeq complete genomes database; however, there are seven such genomes in Kraken-GB’s database, including six strains of P. fermentans. Similarly, in our MiSeq metagenome, Proteus vulgaris is often classified incorrectly at the genus level because the only Proteus genome in Kraken’s database is a single Proteus mirabilis genome. Five more Proteus genomes are present in Kraken-GB’s database, allowing Kraken-GB to classify reads better from that genus. In addition, the MiSeq metagenome contains five genomes from the Enterobacteriaceae family (Citrobacter, Enterobacter, Klebsiella, Proteus and Salmonella). The high sequence similarity between the genera in this family can make distinguishing between genera difficult for any classifier.
The simBA-5 metagenome was created by simulating reads from the set of complete bacterial and archaeal genomes in RefSeq. Replicons from those genomes were used if they were associated with a taxon that had an entry associated with the genus rank, resulting in a set of replicons from 607 genera. We then used the Mason read simulator [22 ] with its Illumina model to produce 10 million 100-bp reads from these genomes. First we created simulated genomes for each species, using a SNP rate of 0.1% and an indel rate of 0.1% (both default parameters), from which we generated the reads. For the simulated reads, we multiplied the default mismatch and indel rates by five, resulting in an average mismatch rate of 2% (ranging from 1% at the beginning of reads to 6% at the ends) and an indel rate of 1% (0.5% insertion probability and 0.5% deletion probability). For the simBA-5 metagenome, the 10,000 read set was generated from a random sample of the 10 million read set.

The analyses reported here result from applying Kleborate v2.0.0 (doi:10.5281/zenodo.4923015) to publicly available genome collections. A total of 13,156 Klebsiella WGS assemblies, encompassing non-duplicate isolates with unique BioSample accessions identified from published studies (some deposited as read sets only, which were assembled using Unicycler v0.4.7^{78 (link)}, data sources summarized in Supplementary Data 13) plus any additional genomes designated as Klebsiella in NCBI’s RefSeq repository of genome assemblies (as of 17 July 2020). In order to minimize the impact of sampling bias favoring common MDR and/or virulent lineages and those causing outbreaks, we subsampled the collection into a ‘non-redundant’ dataset of 11,277 genomes (9705 K. pneumoniae) as follows. Pairwise Mash distances were calculated using Mash v2.1, and used to cluster genomes using single-linkage clustering with a threshold of 0.0003. These clusters were further divided into non-redundant groups with unique combinations of (i) Mash cluster, (ii) chromosomal ST, (iii) virulence gene profiles (i.e. presence of ybt/clb/iro/iuc loci and lineage assignment), (iv) AMR profiles, (v) year and country of isolation, and (vii) specimen type where available. For each resulting non-redundant group, one genome was selected at random as the representative for analyses. The full list of genomes, including database accessions, isolate information, cluster/group assignment, and Kleborate results are provided in Supplementary Data 2. The subset of 1624 K. pneumoniae assemblies deposited in RefSeq by the European EuSCAPE surveillance study³³ (out of 1649 reported in original study; Supplementary Data 2) were used for the EuSCAPE analyses reported in Figs. 2 and 3. The Kleborate-Viz web application is pre-loaded with the non-redundant and EuSCAPE WGS datasets reported in this paper, and can be used to reproduce the plots shown in Figs. 1a–c, 2b, c, 3, 6a, b and to further explore the Kleborate results.

We obtained a total of 2600 Kp genomes (2021 publicly available and 579 novel genomes from a diverse set collected in Australia). Sequence reads were generated locally or obtained from the European Nucleotide Archive (accession numbers are listed in Table S1, available with the online Supplementary Material); 916 genomes that were publicly available as assembled contigs only were downloaded from PATRIC (Wattam et al., 2014 ) and the NCTC3000 Project (Wellcome Trust Sanger Institute – http://www.sanger.ac.uk/resources/downloads/bacteria/nctc/). For isolates sequenced in this study (n=579), DNA was extracted and libraries prepared using the Nextera XT 96 barcode DNA kit and 125 bp paired-end reads were generated on the Illumina HiSeq 2500 platform.
All paired-end read sets were filtered for a mean Phred quality score ≥30, then assembled de novo using SPAdes v3 (Bankevich et al., 2012 (link)). Genomes were excluded from the study if they were duplicate samples, or if there was evidence of contamination or mixed culture measured by: (i) <50 % reads mapping to the NTUH-K2044 reference chromosome (accession number: AP006725.1); (ii) the ratio of heterozygous/homozygous single nucleotide polymorphism (SNP)calls compared to the reference chromosome exceeding 20 %; (iii) the total assembly length being >6.5 Mb, or >6.0 Mb with evidence of >1 % non-Klebsiella read contamination as determined by MetaPhlAn (Segata et al., 2012 (link)); or (v) the assembly being low quality, i.e. total length <5 Mb.

One pair of bottles (aerobic + anaerobic) with a result in the microbiology database formed a blood culture set. A positive blood culture was defined as a blood culture set with one or more positive findings. Potential contaminants were bacteria that are part of the normal skin microbiota (e.g. Coagulase-negative staphylococci, Corynebacterium, Cutibacterium), see Supplementary material S2, Classification of potential contaminants for details. These were considered contaminants if only one blood culture set was positive within 48 hours.
The deduplication period, the period during which only one BSI episode was registered, was set to 14 days. As the deduplication period varies between previous studies, sensitivity analyses were performed for 30, 90 and 365 days. A duplicate was defined as a culture for which there was another positive blood culture with the same finding, taken within the deduplication period. The positive blood cultures remaining after removal of contaminations and duplicates were considered relevant findings.
A polymicrobial finding was two or more different relevant findings from the same patient, obtained within the deduplication period. A BSI episode was defined as an episode with at least one relevant finding and where polymicrobial findings are deduplicated. Thus, if a blood culture set simultaneously grew Escherichia coli and Klebsiella spp., this would count as two relevant findings but only one BSI episode. An R classification in the original microbiology report defined antimicrobial resistance.

The anti-microbial efficient acts of iron nanoparticles were calculated using a well-gar dispersion strategy. They were cultured on Mueller-Hinton Agar at a concentration of 108/ml for bacterial growth and potato agar for fungal growth. Three fungi forms, Aspergillus niger, Alternaria sp., Rhizopus sp., and bacteria forms, the gram-positive Staphylococcus sp., Bacillus subtilis, and gram-negative community Klebsiella pneumonia, Shigella sp., Pseudomonas aeruginosa, and Escherichia coli, have been examined (Table 1). To measure antimicrobial efficiency, 100 μL of biosynthesized iron oxide nanoparticles and 100 μL of water were added as a negative agent. Bacterial strains were isolated in 0.2 mL (CFU 2.5*10⁵ mL). On every agar plate, sterile swabs are placed at irregular intervals. Three of them were properly separated. The sterile cultivation agar surface was used to create wells (holes) with a diameter of 4 mm per plate. Use a metal cork to bore the hole. The aseptic use of 0.2 mL of extract at every hole was performed in the room. The excerpts can be dispersed and cultivated in the agar medium after 1 hour at room temperature. The negative control used in comparison was aseptic pure water. The plates were kept for 24–48 hours at 37°C and 25°C. Inhibition zones were identified as distinct regions surrounding the pools [30 (link)].

We utilized 6 bacterial and 3 fungal strains, including Staphylococcus sp., Bacillus subtilis gram-positive, Klebsiella pneumonia, Shigella sp., Pseudomonas aeruginosa, Escherichia coli gram-negative, and Aspergillus niger, Alternaria sp., and Rhizopus sp. Fungi, obtained from the biology department faculty of sciences, university of princess Nourha bint Abdulrahman (Table 1).