VAMPr: VAriant Mapping and Prediction of antibiotic resistance via explainable features and machine learning

Abstract

Antimicrobial resistance (AMR) is an increasing threat to public health. Current methods of determining AMR rely on inefficient phenotypic approaches, and there remains incomplete understanding of AMR mechanisms for many pathogen-antimicrobial combinations. Given the rapid, ongoing increase in availability of high-density genomic data for a diverse array of bacteria, development of algorithms that could utilize genomic information to predict phenotype could both be useful clinically and assist with discovery of heretofore unrecognized AMR pathways. To facilitate understanding of the connections between DNA variation and phenotypic AMR, we developed a new bioinformatics tool, variant mapping and prediction of antibiotic resistance (VAMPr), to (1) derive gene ortholog-based sequence features for protein variants; (2) interrogate these explainable gene-level variants for their known or novel associations with AMR; and (3) build accurate models to predict AMR based on whole genome sequencing data. We curated the publicly available sequencing data for 3,393 bacterial isolates from 9 species that contained AMR phenotypes for 29 antibiotics. We detected 14,615 variant genotypes and built 93 association and prediction models. The association models confirmed known genetic antibiotic resistance mechanisms, such as blaKPC and carbapenem resistance consistent with the accurate nature of our approach. The prediction models achieved high accuracies (mean accuracy of 91.1% for all antibiotic-pathogen combinations) internally through nested cross validation and were also validated using external clinical datasets. The VAMPr variant detection method, association and prediction models will be valuable tools for AMR research for basic scientists with potential for clinical applicability.

Integrative Analysis of Whole Genome Sequencing and Phenotypic Resistance Toward Prediction of Trimethoprim-Sulfamethoxazole Resistance in Staphylococcus aureus

“…Until then, phenotypic AST will remain as a reference method. The advances in WGS associated with the classical phenotypic AST will help build an accurate database by feeding both the draft genomes and the resistance phenotype in a machine-learning algorithm to highlight the genes variants and hot-spot genomic region associated with the AMR ( Macesic et al, 2017 ; Aytan-Aktug et al, 2020 ; Kim et al, 2020 ). Of course, the harmonization of phenotypic AST is essential to minimize interpretation or technical errors and deviations, which may be associated with specific testing methods.…”

Accelerating surveillance and research of antimicrobial resistance – an online repository for sharing of antimicrobial susceptibility data associated with whole-genome sequences

“…It can be used for standardized open-access data sharing, for example for published data, thus creating an ever-growing source of AST metadata available to researchers worldwide. The large volume of data made available will make it easier to use advanced statistical methods such as machine-learning to predict AMR phenotypes from genomic data and discover new AMR determinants as was recently demonstrated [ 13 ].…”

The ‘thanato-resistome’ - The funeral industry as a potential reservoir of antibiotic resistance: Early insights and perspectives

“…The analysis, synthesis and visualization of such dataset using conventional statistical tools are non-trivial tasks. Big data analytical tools such as machine learning, artificial intelligence and artificial neutral networks are ideal for that purpose ( Hyun et al, 2020 ; Kim et al, 2020 ). Moreover, in silico techniques or computational (eco)toxicology and epidemiology, and network analysis can be used for analysis, integration and synthesis of large dataset from various sources ( Vuorinen et al, 2013 ; Raies and Bajic, 2016 ).…”