Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 8 (1), 421

Developing an in Silico Minimum Inhibitory Concentration Panel Test for Klebsiella Pneumoniae


Developing an in Silico Minimum Inhibitory Concentration Panel Test for Klebsiella Pneumoniae

Marcus Nguyen et al. Sci Rep.


Antimicrobial resistant infections are a serious public health threat worldwide. Whole genome sequencing approaches to rapidly identify pathogens and predict antibiotic resistance phenotypes are becoming more feasible and may offer a way to reduce clinical test turnaround times compared to conventional culture-based methods, and in turn, improve patient outcomes. In this study, we use whole genome sequence data from 1668 clinical isolates of Klebsiella pneumoniae to develop a XGBoost-based machine learning model that accurately predicts minimum inhibitory concentrations (MICs) for 20 antibiotics. The overall accuracy of the model, within ±1 two-fold dilution factor, is 92%. Individual accuracies are ≥90% for 15/20 antibiotics. We show that the MICs predicted by the model correlate with known antimicrobial resistance genes. Importantly, the genome-wide approach described in this study offers a way to predict MICs for isolates without knowledge of the underlying gene content. This study shows that machine learning can be used to build a complete in silico MIC prediction panel for K. pneumoniae and provides a framework for building MIC prediction models for other pathogenic bacteria.

Conflict of interest statement

The authors declare that they have no competing interests.


Figure 1
Figure 1
The pipeline used to optimize and train the XGBoost model using known data (blue), and to predict the MIC values for a new genome (yellow).
Figure 2
Figure 2
The accuracy of the XGBoost model for individual MICs. The X-axis of the heatmap shows the actual MIC (μ g/ml) for a bin and the Y-axis lists the antibiotics. The within ±1-tier accuracy of a particular antibiotic-MIC bin is denoted by color, with red and orange being least accurate and bright yellow and green being most accurate. The number within each cell represents the number of samples (genomes with the MIC) within the bin.

Similar articles

See all similar articles

Cited by 14 PubMed Central articles

See all "Cited by" articles


    1. Chen, L., Todd, R., Kiehlbauch, J., Walters, M. & Kallen, A. Notes from the Field: Pan-Resistant New Delhi Metallo-Beta-Lactamase-Producing Klebsiella pneumoniae –” Washoe County, Nevada, 2016. MMWR Morb Mortal Wkly Rep 66 (2017). - PMC - PubMed
    1. Navon-Venezia S, Kondratyeva K, Carattoli A. Klebsiella pneumoniae: a major worldwide source and shuttle for antibiotic resistance. FEMS Microbiology Reviews. 2017;41:252. doi: 10.1093/femsre/fux013. - DOI - PubMed
    1. Long SW, et al. Population Genomic Analysis of 1,777 Extended-Spectrum Beta-Lactamase-Producing Klebsiella pneumoniae Isolates, Houston, Texas: Unexpected Abundance of Clonal Group 307. mBio. 2017;8:e00489–17. doi: 10.1128/mBio.00489-17. - DOI - PMC - PubMed
    1. Snitkin ES, et al. Tracking a hospital outbreak of carbapenem-resistant Klebsiella pneumoniae with whole-genome sequencing. Science translational medicine. 2012;4:148ra116–148ra116. doi: 10.1126/scitranslmed.3004129. - DOI - PMC - PubMed
    1. Endimiani A, et al. Emergence of blakpc-containing Klebsiella pneumoniae in a long-term acute care hospital: a new challenge to our healthcare system. Journal of Antimicrobial Chemotherapy. 2009;64:1102–1110. doi: 10.1093/jac/dkp327. - DOI - PMC - PubMed

Publication types

MeSH terms