ABSTRACT
A possible way to slow down the antibiotic resistance crisis is to be more strict when it comes to antibiotics prescriptions. For accurate antibiotic prescriptions, antibiotic susceptibility data are needed. With the increasing availability of next-generation sequencing (NGS), bacterial whole genome sequencing (WGS) is becoming a feasible alternative to traditional phenotyping for the detection and surveil-lance of AMR.
This work proposes a machine learning approach that can predict the minimum inhibitory concentration (MIC) for a given antibiotic, here ciprofloxacin, on the basis of genome-wide mutation profiles alongside profiles of acquired resistance genes. We analyzed 704 Escherichia coli WGS samples coming from different countries along with their MIC measurements for ciprofloxacin. The four most important predictors found by the model, mutations in gyrA and parC and the presence of any qnrS gene, have been experimentally validated before (van der Putten BCL et al, J Antimicrob Chemother. 2019 Feb 1;74(2):298-310. doi: 10.1093/jac/dky417). Using only these four predictors with a linear regression model 65% and 92% of the test samples’ MIC were correctly predicted within a two- and a four-fold dilution range, respectively. The presented work goes further than the typical predictions using machine learning as a black box model concept. The recent progress in WGS technology in combination with machine learning analysis approaches indicates that in the near future WGS of bacteria might be cheaper and faster than a MIC measurement.
IMPORTANCE Whole genome sequencing has become the standard approach to study molecular epidemiology of bacteria. However, uptake of WGS in the clinical microbiology laboratory as part of individual patient diagnostics still requires significant steps forward, in particular with respect to prediction of antibiotic susceptibility based on DNA sequence. Whilst the majority of studies of prediction of susceptibility have used a binary outcome (susceptible/resistant), a quantitative prediction of susceptibility, such as MIC, will allow for earlier detection of trends in increasing resistance as well as the flexibility to follow potential adjustments in definitions of susceptible and resistant categories (breakpoints).