Format

Send to

Choose Destination
Sci Rep. 2018 Jan 11;8(1):421. doi: 10.1038/s41598-017-18972-w.

Developing an in silico minimum inhibitory concentration panel test for Klebsiella pneumoniae.

Nguyen M1,2,3, Brettin T2,3, Long SW4,5, Musser JM4,5, Olsen RJ4,5, Olson R2,3, Shukla M2,3, Stevens RL2,3,6, Xia F2,3, Yoo H2,3, Davis JJ7,8.

Author information

1
Northern Illinois University, Computation Science, DeKalb, IL, 60115, USA.
2
University of Chicago, Computation Institute, Chicago, IL, 60637, USA.
3
Argonne National Laboratory, Computing Environment and Life Sciences, Argonne, IL, 60439, USA.
4
Center for Molecular and Translational Human Infectious Diseases Research, Department of Pathology and Genomic Medicine, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, 77030, USA.
5
Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, New York, 10065, USA.
6
University of Chicago, Department of Computer Science, Chicago, IL, 60439, USA.
7
University of Chicago, Computation Institute, Chicago, IL, 60637, USA. jjdavis@anl.gov.
8
Argonne National Laboratory, Computing Environment and Life Sciences, Argonne, IL, 60439, USA. jjdavis@anl.gov.

Abstract

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.

PMID:
29323230
PMCID:
PMC5765115
DOI:
10.1038/s41598-017-18972-w
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Nature Publishing Group Icon for PubMed Central
Loading ...
Support Center