Ion chef workflows

Libraries were prepared using Ion Torrent technology and Ion Chef workflows (Thermo Scientific, Waltham, MA, USA). Sequencing was performed in the S5XLS system, and analysis of the raw sequencing data was conducted by Ion Torrent Suite v.5.10.0, according to the manufacturer’s instructions. Sequencing of the D1644 isolate performed twice to resolve ambiguities related to the presence of an IncN plasmid type harboring bla_NDM-1. The detection of the plasmid type and downstream analysis for D1644 performed on the merged sequencing products. Raw sequencing data were quality-checked and trimmed to keep only the reads that match the minimum length and quality criteria. Taxonomic assignment was performed as a quality step to filter out samples that may have been contaminated by foreign DNA during sample preparation and to verify the presence of the isolated species. Kraken2 was used for the taxonomy classification built on the standard database that includes NCBI taxonomic information, complete RefSeq microbe genomes, human genomes, and a vector collection [25 (link)]. Genomes were de novo assembled by AssemblerSPAdes and SPAdes using the default settings for the k-mers and the Ion Torrent parameter [26 (link)].

The libraries for genomic DNA were prepared using Ion Torrent Technology and Ion Chef Workflows (Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, sequencing of the genomic DNA libraries was performed on the S5XLS system, followed by primary data analysis using Ion Torrent Suite (v.5.10.0). Quality assessment of the reads was conducted using FastQC software (v.0.11.9); assembly of the reads was performed using the SPAdes genome assembler (v3.15.5) with the default settings. The assembled genomes’ quality was evaluated using the Quast version 5.2.0 tool and average coverage was determined for each genome using the mapPacBio tool from BBTools (https://sourceforge.net/projects/bbmap/; accessed on 9 January 2024). The nucleotide alterations of the genes oprD, mexR, nalC, and nalD were confirmed by PCR, followed by sequencing analysis, using primers designed for the purpose of this study.
For the identification of genes associated with antibiotic resistance, ResFinder-4.4.2 was employed with the ID threshold set to 90% and the minimum length set to 60%. Comparative analysis of the genome between strains was performed using Blast analysis, using as reference the P. aeruginosa PAO1 genome (accession no AE004091.2).