A Genome-Based Model to Predict the Virulence of Pseudomonas aeruginosa Isolates

Abstract

Variation in the genome of Pseudomonas aeruginosa, an important pathogen, can have dramatic impacts on the bacterium’s ability to cause disease. We therefore asked whether it was possible to predict the virulence of P. aeruginosa isolates based on their genomic content. We applied a machine learning approach to a genetically and phenotypically diverse collection of 115 clinical P. aeruginosa isolates using genomic information and corresponding virulence phenotypes in a mouse model of bacteremia. We defined the accessory genome of these isolates through the presence or absence of accessory genomic elements (AGEs), sequences present in some strains but not others. Machine learning models trained using AGEs were predictive of virulence, with a mean nested cross-validation accuracy of 75% using the random forest algorithm. However, individual AGEs did not have a large influence on the algorithm’s performance, suggesting instead that virulence predictions are derived from a diffuse genomic signature. These results were validated with an independent test set of 25 P. aeruginosa isolates whose virulence was predicted with 72% accuracy. Machine learning models trained using core genome single-nucleotide variants and whole-genome k-mers also predicted virulence. Our findings are a proof of concept for the use of bacterial genomes to predict pathogenicity in P. aeruginosa and highlight the potential of this approach for predicting patient outcomes. IMPORTANCE Pseudomonas aeruginosa is a clinically important Gram-negative opportunistic pathogen. P. aeruginosa shows a large degree of genomic heterogeneity both through variation in sequences found throughout the species (core genome) and through the presence or absence of sequences in different isolates (accessory genome). P. aeruginosa isolates also differ markedly in their ability to cause disease. In this study, we used machine learning to predict the virulence level of P. aeruginosa isolates in a mouse bacteremia model based on genomic content. We show that both the accessory and core genomes are predictive of virulence. This study provides a machine learning framework to investigate relationships between bacterial genomes and complex phenotypes such as virulence.

Phenotypic and Genomic Variability of Serial Peri-Lung Transplantation Pseudomonas aeruginosa Isolates From Cystic Fibrosis Patients

“…Additionally, we saw the spread of the SPLT isolates throughout the phylogenetic tree based on core genes, still showing patient specific clustering. This observation was found to be coherent with the noted difficulties in discriminating CF isolates based on their geography and origin ( Jeukens et al, 2019 ; Pincus et al, 2020 ).The latter demands further understanding of the emergence of resistance and virulence among these P. aeruginosa to avoid or manage lung infections ( Cho et al, 2014 ; Jaillard et al, 2017 ).…”

Single nucleotide variants in Pseudomonas aeruginosa populations from sputum correlate with baseline lung function and predict disease progression in individuals with cystic fibrosis

“…By including these features in our prediction models, we noted a moderate increase across all the measured metrics relative to using only AmpliSeq data. These results are in line with previous studies showing the improvement of ML-based phenotype prediction by adding relevant clinical data (MacFadden, Melano et al 2019, Pincus, Ozer et al 2020). We note that clinical factors only modestly improved the performance of the models (∼5%), highlighting the rich information and predictive value of the Pa AmpliSeq data alone.…”

“…This suggests that genetic variation in dominant pathogens can significantly complement and improve upon predictions of disease status based on the microbiome. Along these lines, another recent study showed that the Pa genomic data can predict pathogenicity in mouse models (Pincus, Ozer et al 2020).…”

Genomic Features Associated with the Degree of Phenotypic Resistance to Carbapenems in Carbapenem-Resistant Klebsiella pneumoniae

“…Machine learning analyses were performed on the ST258 CR-Kp isolates ( n = 123) using the sci-kit learn library (v0.23.2) ( https://www.jmlr.org/papers/v12/pedregosa11a.html ) in Python v3.6.12 following the pipeline described by Pincus et al ( 58 ) For the curated analysis, β-lactamase presence/absence, outer membrane porin phenotype, transposon isoform, and bla KPC copy number were used as features and converted to binary variables. For example, bla KPC copy number was considered “high” if ≥ 4 copies were present and “low” if < 4 copies were present.…”