Direct prediction of carbapenem resistance in Pseudomonas aeruginosa by whole genome sequencing and metagenomic sequencing

Abstract

Carbapenem resistance is a major concern in the management of antibiotic-resistant Pseudomonas aeruginosa infections. The direct prediction of carbapenem-resistant phenotype from genotype in P. aeruginosa isolates and clinical samples would promote timely antibiotic therapy. The complex carbapenem resistance mechanism and the high prevalence of variant-driven carbapenem resistance in P. aeruginosa make it challenging to predict the carbapenem-resistant phenotype through the genotype. In this study, using whole genome sequencing (WGS) data of 1,622 P. aeruginosa isolates followed by machine learning, we screened 16 and 31 key gene features associated with imipenem (IPM) and meropenem (MEM) resistance in P. aeruginosa, including oprD(HIGH), and constructed the resistance prediction models. The areas under the curves of the IPM and MEM resistance prediction models were 0.906 and 0.925, respectively. For the direct prediction of carbapenem resistance in P. aeruginosa from clinical samples by the key gene features selected and prediction models constructed, 72 P. aeruginosa-positive sputum samples were collected and sequenced by metagenomic sequencing (MGS) based on next-generation sequencing (NGS) or Oxford Nanopore Technology (ONT). The prediction applicability of MGS based on NGS outperformed that of MGS based on ONT. In 72 P. aeruginosa-positive sputum samples, 65.0% (26/40) of IPM-insensitive and 63.2% (24/38) of MEM-insensitive P. aeruginosa were directly predicted by NGS-based MGS with positive predictive values of 0.897 and 0.889, respectively. By the direct detection of the key gene features associated with carbapenem resistance of P. aeruginosa, the carbapenem resistance of P. aeruginosa could be directly predicted from cultured isolates by WGS or from clinical samples by NGS-based MGS, which could assist the timely treatment and surveillance of carbapenem-resistant P. aeruginosa.

Keywords: Pseudomonas aeruginosa; carbapenem resistance prediction; machine learning; metagenomic sequencing; whole genome sequencing.

Fig 1

Phylogenetic, geographic, and resistant phenotype distribution of 1622 P. aeruginosa isolates. Five circles from the inside out represent the phylogenetic tree, the country distribution, the antibiotic-resistant phenotype to IPM, the antibiotic-resistant phenotype to MEM, and the training and validation set classification of 1,622 P. aeruginosa isolates included in the study.

Fig 2

PPVs and distributions of GVARs according to HIGH, MODERATE, and LOW effect impact classification. (A) PPVs of the HIGH, MODERATE, and LOW effect impact groups of different GVARs. oprD_PAO1 or oprD_PAO1+ indicates the variants of oprD identified with reference to PAO1 alone or eight strains including PAO1. (B and C) PPVs of each GVAR in the oprD(HIGH) group. (D and E) Distributions of oprD(HIGH) and each GVAR in the oprD(HIGH) group. The short blue line represents the existence of oprD(HIGH) variants in P. aeruginosa isolates. Abbreviations: R, resistant; S, sensitive.

Fig 3

Method selection and sample size evaluation of machine learning. (A and B) Performance of prediction models for the IPM resistance (A) and MEM resistance (B) built by different machine learning methods. (C and D) Performance of prediction models with different sample sizes in the IPM (B) and MEM (C) data sets.

Fig 4

Key gene features and performance of the carbapenem resistance prediction model of P. aeruginosa. (A) Key gene features for the IPM and MEM resistance prediction of P. aeruginosa. GPA features are presented in orange, and GVAR features are presented in black. (B) AUCs of IPM and MEM resistance prediction models in the training and validation sets. PAO1 or PAO1+ represents the AUCs when variants of oprD were identified with reference to PAO1 alone or eight strains including PAO1. (C) AUCs of the IPM and MEM resistance prediction models at different sequencing depths and genome coverages derived from simulated NGS reads and ONT reads.

Fig 5

Heatmap of key gene features associated with carbapenem resistance of P. aeruginosa detected in 30 sputum samples by MGS based on NGS or ONT. The yellow block indicates the detection of key gene features by NGS; the green block indicates the detection of key gene features by ONT; the plum block indicates R/I by AST; and the gray block indicates ND of key gene features or S by AST. Abbreviations: ID, identity; ND, not detected; R, resistant; S, sensitive.