skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Antimicrobial resistance prediction in PATRIC and RAST

Abstract

The emergence and spread of antimicrobial resistance (AMR) mechanisms in bacterial pathogens, coupled with the dwindling number of effective antibiotics, has created a global health crisis. Being able to identify the genetic mechanisms of AMR and predict the resistance phenotypes of bacterial pathogens prior to culturing could inform clinical decision-making and improve reaction time. At PATRIC (http://patricbrc.org/), we have been collecting bacterial genomes with AMR metadata for several years. In order to advance phenotype prediction and the identification of genomic regions relating to AMR, we have updated the PATRIC FTP server to enable access to genomes that are binned by their AMR phenotypes, as well as metadata including minimum inhibitory concentrations. Using this infrastructure, we custom built AdaBoost (adaptive boosting) machine learning classifiers for identifying carbapenem resistance in Acinetobacter baumannii, methicillin resistance in Staphylococcus aureus, and beta-lactam and co-trimoxazole resistance in Streptococcus pneumoniae with accuracies ranging from 88–99%. We also did this for isoniazid, kanamycin, ofloxacin, rifampicin, and streptomycin resistance in Mycobacterium tuberculosis, achieving accuracies ranging from 71–88%. Lastly, this set of classifiers has been used to provide an initial framework for species-specific AMR phenotype and genomic feature prediction in the RAST and PATRIC annotation services.

Authors:
 [1];  [2];  [1];  [3];  [3];  [1];  [4];  [5];  [1];  [3];  [3];  [1];  [1]
  1. Univ. of Chicago, Chicago, IL (United States); Argonne National Lab. (ANL), Lemont, IL (United States)
  2. Gydle Inc., Chanoine Morel Quebec, QC (Canada)
  3. Biocomplexity Institute of Virginia Tech., Blacksburg, VA (United States)
  4. Argonne National Lab. (ANL), Lemont, IL (United States); The Fellowship for Interpretation of Genomes, Burr Ridge, IL (United States)
  5. Univ. of Chicago, Chicago, IL (United States)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1258659
Grant/Contract Number:
AC02-06CH11357
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Scientific Reports
Additional Journal Information:
Journal Volume: 6; Journal ID: ISSN 2045-2322
Publisher:
Nature Publishing Group
Country of Publication:
United States
Language:
English
Subject:
59 BASIC BIOLOGICAL SCIENCES; adaptive boosting; genome annotation; machine learning; random forest; support vector machines; computational biology and bioinformatics; genetic databases

Citation Formats

Davis, James J., Boisvert, Sebastien, Brettin, Thomas, Kenyon, Ronald W., Mao, Chunhong, Olson, Robert, Overbeek, Ross, Santerre, John, Shukla, Maulik, Wattam, Alice R., Will, Rebecca, Xia, Fangfang, and Stevens, Rick. Antimicrobial resistance prediction in PATRIC and RAST. United States: N. p., 2016. Web. doi:10.1038/srep27930.
Davis, James J., Boisvert, Sebastien, Brettin, Thomas, Kenyon, Ronald W., Mao, Chunhong, Olson, Robert, Overbeek, Ross, Santerre, John, Shukla, Maulik, Wattam, Alice R., Will, Rebecca, Xia, Fangfang, & Stevens, Rick. Antimicrobial resistance prediction in PATRIC and RAST. United States. doi:10.1038/srep27930.
Davis, James J., Boisvert, Sebastien, Brettin, Thomas, Kenyon, Ronald W., Mao, Chunhong, Olson, Robert, Overbeek, Ross, Santerre, John, Shukla, Maulik, Wattam, Alice R., Will, Rebecca, Xia, Fangfang, and Stevens, Rick. Tue . "Antimicrobial resistance prediction in PATRIC and RAST". United States. doi:10.1038/srep27930. https://www.osti.gov/servlets/purl/1258659.
@article{osti_1258659,
title = {Antimicrobial resistance prediction in PATRIC and RAST},
author = {Davis, James J. and Boisvert, Sebastien and Brettin, Thomas and Kenyon, Ronald W. and Mao, Chunhong and Olson, Robert and Overbeek, Ross and Santerre, John and Shukla, Maulik and Wattam, Alice R. and Will, Rebecca and Xia, Fangfang and Stevens, Rick},
abstractNote = {The emergence and spread of antimicrobial resistance (AMR) mechanisms in bacterial pathogens, coupled with the dwindling number of effective antibiotics, has created a global health crisis. Being able to identify the genetic mechanisms of AMR and predict the resistance phenotypes of bacterial pathogens prior to culturing could inform clinical decision-making and improve reaction time. At PATRIC (http://patricbrc.org/), we have been collecting bacterial genomes with AMR metadata for several years. In order to advance phenotype prediction and the identification of genomic regions relating to AMR, we have updated the PATRIC FTP server to enable access to genomes that are binned by their AMR phenotypes, as well as metadata including minimum inhibitory concentrations. Using this infrastructure, we custom built AdaBoost (adaptive boosting) machine learning classifiers for identifying carbapenem resistance in Acinetobacter baumannii, methicillin resistance in Staphylococcus aureus, and beta-lactam and co-trimoxazole resistance in Streptococcus pneumoniae with accuracies ranging from 88–99%. We also did this for isoniazid, kanamycin, ofloxacin, rifampicin, and streptomycin resistance in Mycobacterium tuberculosis, achieving accuracies ranging from 71–88%. Lastly, this set of classifiers has been used to provide an initial framework for species-specific AMR phenotype and genomic feature prediction in the RAST and PATRIC annotation services.},
doi = {10.1038/srep27930},
journal = {Scientific Reports},
number = ,
volume = 6,
place = {United States},
year = {Tue Jun 14 00:00:00 EDT 2016},
month = {Tue Jun 14 00:00:00 EDT 2016}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 10works
Citation information provided by
Web of Science

Save / Share:
  • Shotgun metagenomics creates millions of fragments of short DNA reads, which are meaningless unless analyzed appropriately. The Metagenomics RAST server (MG-RAST) is a web-based, open source system that offers a unique suite of tools for analyzing these data sets. After de-replication and quality control, fragments are mapped against a comprehensive nonredundant database (NR). Phylogenetic and metabolic reconstructions are computed from the set of hits against the NR. The resulting data are made available for browsing, download, and most importantly, comparison against a comprehensive collection of public metagenomes. A submitted metagenome is visible only to the user, unless the user makesmore » it public or shares with other registered users. Public metagenomes are available to all.« less
  • Sewage effluent and outfall confluence samples were collected at the Barceloneta Regional Treatment Plant in Barceloneta, Puerto Rico; outfall confluence samples at Ocean City, Md., were also collected. Samples from uncontaminated open ocean areas served as clean-water controls. Bacteria were enriched in marine broth 2216 amended with 1 ..mu..g of one of a set of chemical selected for study per ml: nitrobenzene, dibutyl phthalate, m-cresol, o-cresol, 4-nitroaniline, bis(tributyltin) oxide, and quinone. MICs of the chemicals were determined individually for all isolates. Bacterial isolates were evaluated for resistance to nine different antibiotics and for the presence of plasmid DNA. Treated sewagemore » was found to contain large numbers of bacteria simultaneously possessing antibiotic resistance, chemical resistance, and multiple bands of plasmic DNA. Bacteria resistant to penicillin, erythromycin, nalidixic acid, ampicillin, m-cresol, quinone, and bis(tributyltin) oxide were detected in nearly all samples, but only sewage outfall confluence samples yielded bacterial isolates that were resistant to streptomycin. Bacteria resistant to a combination of antibiotics, including kanamycin, chloramphenicol, gentamicin, and tetracycline, were isolated only from sewage effluent samples. It is concluded that bacterial isolates derived from toxic chemical wastes more frequently contain plasmid DNA and demonstrate antimicrobial resistance than do bacterial isolates from domestic sewage-impacted waters or from uncontaminated open ocean sites.« less
  • Heavy metals are regularly found in liquid pig manure, and might interact with bacterial antimicrobial resistance. Concentrations of heavy metals were determined by atomic spectroscopic methods in 305 pig manure samples and were connected to the phenotypic resistance of Escherichia coli (n=613) against 29 antimicrobial drugs. Concentrations of heavy metals (/kg dry matter) were 0.08-5.30 mg cadmium, 1.1-32.0 mg chrome, 22.4-3387.6 mg copper, <2.0-26.7 mg lead, <0.01-0.11 mg mercury, 3.1-97.3 mg nickel and 93.0-8239.0 mg zinc. Associated with the detection of copper and zinc, resistance rates against {beta}-lactams were significantly elevated. By contrast, the presence of mercury was significantly associatedmore » with low antimicrobial resistance rates of Escherichia coli against {beta}-lactams, aminoglycosides and other antibiotics. Effects of subinhibitory concentrations of mercury on bacterial resistance against penicillins, cephalosporins, aminoglycosides and doxycycline were also demonstrated in a laboratory trial. Antimicrobial resistance in the porcine microflora might be increased by copper and zinc. By contrast, the occurrence of mercury in the environment might, due to co-toxicity, act counter-selective against antimicrobial resistant strains.« less
  • Cited by 18