Dataset Information

Antimicrobial Resistance Prediction for Gram-Negative Bacteria via Game Theory-Based Feature Evaluation.

ABSTRACT: The increasing prevalence of antimicrobial-resistant bacteria drives the need for advanced methods to identify antimicrobial-resistance (AMR) genes in bacterial pathogens. With the availability of whole genome sequences, best-hit methods can be used to identify AMR genes by differentiating unknown sequences with known AMR sequences in existing online repositories. Nevertheless, these methods may not perform well when identifying resistance genes with sequences having low sequence identity with known sequences. We present a machine learning approach that uses protein sequences, with sequence identity ranging between 10% and 90%, as an alternative to conventional DNA sequence alignment-based approaches to identify putative AMR genes in Gram-negative bacteria. By using game theory to choose which protein characteristics to use in our machine learning model, we can predict AMR protein sequences for Gram-negative bacteria with an accuracy ranging from 93% to 99%. In order to obtain similar classification results, identity thresholds as low as 53% were required when using BLASTp.

SUBMITTER: Chowdhury AS

PROVIDER: S-EPMC6785542 | biostudies-literature | 2019 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Antimicrobial Resistance Prediction for Gram-Negative Bacteria via Game Theory-Based Feature Evaluation.

Chowdhury Abu Sayed AS Call Douglas R DR Broschat Shira L SL

Scientific reports 20191009 1

The increasing prevalence of antimicrobial-resistant bacteria drives the need for advanced methods to identify antimicrobial-resistance (AMR) genes in bacterial pathogens. With the availability of whole genome sequences, best-hit methods can be used to identify AMR genes by differentiating unknown sequences with known AMR sequences in existing online repositories. Nevertheless, these methods may not perform well when identifying resistance genes with sequences having low sequence identity with k ...[more]

PMID: 31597945

Similar Datasets

Project description:BackgroundAntimicrobial resistance (AMR) in Gram-negative bacteria-causing bloodstream infections (BSIs), such as Klebsiella pneumoniae and non-typhoidal Salmonella (NTS), is a major public health concern. Nonetheless, AMR surveillance remains scarce in sub-Saharan Africa, where BSI treatment is largely empirical. The aim of the study was to determine the distribution and AMR patterns of BSI-causing NTS, K. pneumoniae, and other Gram-negative bacteria in Ghana.MethodsA cross-sectional study was conducted between April and December 2021 at eleven sentinel health facilities across Ghana as part of a pilot study on the feasibility and implementation of the human sector AMR surveillance harmonized protocol in sub-Saharan Africa. Gram-negative bacteria recovered from blood specimens of febrile patients were identified using MALDI-TOF and evaluated for antimicrobial resistance using the BD Phoenix M50 analyzer and Kirby-Bauer disc diffusion. The Department of Medical Microbiology at the University of Ghana served as the reference laboratory.ResultsOut of 334 Gram-negative blood isolates, there were 18 (5.4%) NTS, 85 (25.5%) K. pneumoniae, 88 (26.4%) Escherichia coli, 40 (12.0%) Acinetobacter baumannii, 25 (7.5%) Pseudomonas aeruginosa, and 77 (23.1%) other Gram-negative bacteria. As a composite, the isolates displayed high resistance to the antibiotics tested-amoxicillin (89.3%), tetracycline (76.1%), trimethoprim-sulfamethoxazole (71.5%), and chloramphenicol (59.7%). Resistance to third-generation cephalosporins [ceftriaxone (73.7%), cefotaxime (77.8%), and ceftazidime (56.3%)] and fluoroquinolones [ciprofloxacin (55.3%)] was also high; 88% of the isolates were multidrug resistant, and the rate of extended-spectrum beta-lactamase (ESBL) production was 44.6%. Antibiotic resistance in K. pneumoniae followed the pattern of all Gram-negative isolates. Antibiotic resistance was lower in NTS blood isolates, ranging between 16.7-38.9% resistance to the tested antibiotics. Resistance rates of 38.9%, 22.2%, and 27.8% were found for cefotaxime, ceftriaxone, and ceftazidime, respectively, and 27.8% and 23.8% for ciprofloxacin and azithromycin, respectively, which are used in the treatment of invasive NTS. The prevalence of multidrug resistance in NTS isolates was 38.9%.ConclusionsMulticenter AMR surveillance of Gram-negative blood isolates from febrile patients was well-received in Ghana, and the implementation of a harmonized protocol was feasible. High resistance and multidrug resistance to first- or second-choice antibiotics, including penicillins, third-generation cephalosporins, and fluoroquinolones, were found, implying that these antibiotics might have limited effectiveness in BSI treatment in the country. Continuation of AMR surveillance in Gram-negative blood isolates is essential for a better understanding of the extent of AMR in these pathogens and to guide clinical practice and policymaking.

Dataset Information

Antimicrobial Resistance Prediction for Gram-Negative Bacteria via Game Theory-Based Feature Evaluation.

Publications

Antimicrobial Resistance Prediction for Gram-Negative Bacteria via Game Theory-Based Feature Evaluation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets