Explainable Deep Learning Approach for Multilabel Classification of Antimicrobial Resistance With Missing Labels DOI Creative Commons
Mukunthan Tharmakulasingam, Brian Gardner, Roberto M. La Ragione

et al.

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 113073 - 113085

Published: Jan. 1, 2022

Predicting Antimicrobial Resistance (AMR) from genomic sequence data has become a significant component of overcoming the AMR challenge, especially given its potential for facilitating more rapid diagnostics and personalised antibiotic treatments. With recent advances in sequencing technologies computing power, deep learning models have been widely adopted to predict reliably error-free. There are many different types AMR; therefore, any practical prediction system must be able identify multiple AMRs present sequence. Unfortunately, most datasets do not all labels marked, thereby making modelling approach challenging owing reliance on reliability accuracy. This paper addresses this issue by presenting an effective solution, Mask-Loss 1D convolution neural network (ML-ConvNet), with missing labels. The core ML- ConvNet utilises masked loss function that overcomes effect predicting AMR. proposed ML-ConvNet is demonstrated outperform state-of-the-art methods literature 10.5%, according F1 score. model's performance evaluated using degrees label found conventional 76% score when 86.68% missing. Furthermore, was established explainable artificial intelligence (XAI) pipeline, it ideally suited hospital healthcare settings, where model interpretability essential requirement.

Language: Английский

Fighting Antibiotic Resistance in Hospital-Acquired Infections: Current State and Emerging Technologies in Disease Prevention, Diagnostics and Therapy DOI Creative Commons
Ekaterina Avershina,

Valeria Shapovalova,

German A. Shipulin

et al.

Frontiers in Microbiology, Journal Year: 2021, Volume and Issue: 12

Published: July 21, 2021

Rising antibiotic resistance is a global threat that projected to cause more deaths than all cancers combined by 2050. In this review, we set summarize the current state of resistance, and give an overview emerging technologies aimed escape pre-antibiotic era recurrence. We conducted comprehensive literature survey >150 original research review articles indexed in Web Science using “antimicrobial resistance,” “diagnostics,” “therapeutics,” “disinfection,” “nosocomial infections,” “ESKAPE pathogens” as key words. discuss impact nosocomial infections on spread multi-drug resistant bacteria, over existing developing strategies for faster diagnostics infectious diseases, novel approaches therapy finally hospital disinfection prevent MDR bacteria spread.

Language: Английский

Citations

109

Ultrafast and Cost-Effective Pathogen Identification and Resistance Gene Detection in a Clinical Setting Using Nanopore Flongle Sequencing DOI Creative Commons
Ekaterina Avershina, Stephan A. Frye, Jawad Ali

et al.

Frontiers in Microbiology, Journal Year: 2022, Volume and Issue: 13

Published: March 17, 2022

Rapid bacterial identification and antimicrobial resistance gene (ARG) detection are crucial for fast optimization of antibiotic treatment, especially septic patients where each hour delayed prescription might have lethal consequences. This work investigates whether the Oxford Nanopore Technology's (ONT) Flongle sequencing platform is suitable real-time directly from blood cultures to identify bacteria detect resistance-encoding genes. For analysis, we used pure four clinical isolates Escherichia coli Klebsiella pneumoniae two samples spiked with either E. or K. that had been cultured overnight. We sequenced both whole genome plasmids isolated these using different kits. Generally, data allow rapid ID resistome based on first 1,000-3,000 generated sequences (10 min 3 h start), albeit ARG variant did not always correspond ONT MinION Illumina sequencing-based data. sufficient 99.9% coverage within at most 20,000 (clinical isolates) 50,000 (positive cultures) generated. The SQK-LSK110 Ligation kit resulted in higher more accurate than SQK-RBK004 Barcode kit.

Language: Английский

Citations

39

Clinical Diagnostics of Bacterial Infections and Their Resistance to Antibiotics—Current State and Whole Genome Sequencing Implementation Perspectives DOI Creative Commons
Ekaterina Avershina, Abdolrahman Khezri, Rafi Ahmad

et al.

Antibiotics, Journal Year: 2023, Volume and Issue: 12(4), P. 781 - 781

Published: April 19, 2023

Antimicrobial resistance (AMR), defined as the ability of microorganisms to withstand antimicrobial treatment, is responsible for millions deaths annually. The rapid spread AMR across continents warrants systematic changes in healthcare routines and protocols. One fundamental issues with lack diagnostic tools pathogen identification detection. Resistance profile often depends on culturing thus may last up several days. This contributes misuse antibiotics viral infection, use inappropriate antibiotics, overuse broad-spectrum or delayed infection treatment. Current DNA sequencing technologies offer potential develop that can provide information a few hours rather than However, these techniques commonly require advanced bioinformatics knowledge and, at present, are not suited routine lab use. In this review, we give an overview burden healthcare, describe current screening methods, perspectives how be used diagnostics. Additionally, discuss common steps data analysis, currently available pipelines, analysis. Direct, culture-independent has complement culture-based methods clinical settings. there need minimum set standards terms evaluating results generated. machine learning algorithms regarding phenotype detection (resistance/susceptibility antibiotic).

Language: Английский

Citations

31

Short turnaround time of seven to nine hours from sample collection until informed decision for sepsis treatment using nanopore sequencing DOI Creative Commons
Jawad Ali, Wenche Johansen, Rafi Ahmad

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: March 19, 2024

Abstract Bloodstream infections (BSIs) and sepsis are major health problems, annually claiming millions of lives. Traditional blood culture techniques, employed to identify sepsis-causing pathogens assess antibiotic susceptibility, usually take 2–4 days. Early accurate prescription is vital in mitigate mortality resistance. This study aimed reduce the wait time for diagnosis by employing shorter incubation times BD BACTEC™ bottles using standard laboratory incubators, followed real-time nanopore sequencing data analysis. The method was tested on nine samples spiked with clinical isolates from six most prevalent pathogens. results showed that pathogen identification possible at as low 10 2 –10 4 CFU/mL, achieved after just h within 40 min sequencing. Moreover, all antimicrobial resistance genes were identified 3 7 5 only Therefore, total turnaround sample collection information required an informed decision right treatment between 9 h. These hold significant promise better management compared current culture-based methods.

Language: Английский

Citations

9

A Practical Approach for Predicting Antimicrobial Phenotype Resistance in Staphylococcus aureus Through Machine Learning Analysis of Genome Data DOI Creative Commons
Shuyi Wang, Chunjiang Zhao,

Yuyao Yin

et al.

Frontiers in Microbiology, Journal Year: 2022, Volume and Issue: 13

Published: March 2, 2022

With the reduction in sequencing price and acceleration of speed, it is particularly important to directly link genotype phenotype bacteria. Here, we firstly predicted minimum inhibitory concentrations ten antimicrobial agents for Staphylococcus aureus using 466 isolates by extracting k-mer from whole genome data combined with three machine learning algorithms: random forest, support vector machine, XGBoost. Considering one two-fold dilution, essential agreement category could reach >85% >90% most agents. For clindamycin, cefoxitin trimethoprim-sulfamethoxazole, >91% >93%, providing information clinical treatment. The successful prediction resistance showed that model identify methicillin-resistant S. aureus. results suggest small datasets available large hospitals bypass existing basic research known genes accurately predict bacterial phenotype.

Language: Английский

Citations

30

Genomic Features Associated with the Degree of Phenotypic Resistance to Carbapenems in Carbapenem-Resistant Klebsiella pneumoniae DOI
Zackery P. Bulman, Fiorella Krapp, Nathan B. Pincus

et al.

mSystems, Journal Year: 2021, Volume and Issue: 6(5)

Published: Sept. 14, 2021

Carbapenem-resistant Klebsiella pneumoniae strains cause severe infections that are difficult to treat. The production of carbapenemases such as the K. carbapenemase (KPC) is a common mechanism by which these resist killing carbapenems. However, degree phenotypic carbapenem resistance (MIC) may differ markedly between isolates with similar genes, suggesting our understanding underlying mechanisms remains incomplete. To address this problem, we determined whole-genome sequences 166 clinical resistant meropenem, imipenem, or ertapenem. Multiple linear regression analysis collection largely blaKPC-3-containing sequence type 258 (ST258) indicated blaKPC copy number and some outer membrane porin gene mutations were associated higher MICs A trend toward was also observed those genes carried d isoform Tn4401. In contrast, ompK37 lower MICs, extended spectrum β-lactamase not in carbapenem-resistant pneumoniae. machine learning approach based on did result substantial improvement prediction high low MICs. These results build upon previous findings multiple factors influence overall levels isolates. IMPORTANCE can blood, urinary tract, lungs. Resistance carbapenems an urgent public health threat, since it make While individual contributors have been studied, few reports explore their combined effects We sequenced evaluate contribution known try identify novel one specific porin, ompK37, Machine nor resistance. enhance many

Language: Английский

Citations

38

Hybrid Assembly Provides Improved Resolution of Plasmids, Antimicrobial Resistance Genes, and Virulence Factors in Escherichia coli and Klebsiella pneumoniae Clinical Isolates DOI Creative Commons
Abdolrahman Khezri, Ekaterina Avershina, Rafi Ahmad

et al.

Microorganisms, Journal Year: 2021, Volume and Issue: 9(12), P. 2560 - 2560

Published: Dec. 10, 2021

Emerging new sequencing technologies have provided researchers with a unique opportunity to study factors related microbial pathogenicity, such as antimicrobial resistance (AMR) genes and virulence factors. However, the use of whole-genome sequence (WGS) data requires good knowledge bioinformatics involved, well necessary techniques. In this study, total nine Escherichia coli Klebsiella pneumoniae isolates from Norwegian clinical samples were sequenced using both MinION Illumina platforms. Three out directly blood culture, one sample was mixed-blood culture. For genome assembly, several long-read, (Canu, Flye, Unicycler, Miniasm), short-read (ABySS, Unicycler SPAdes) hybrid assemblers (Unicycler, hybridSPAdes, MaSurCa) tested. Assembled genomes best-performing (according quality checks QUAST BUSCO) subjected downstream analyses. Flye performed best for assembly long short reads, respectively. top-performing assembler produced more circularized complete assemblies. Hybrid assembled substantially better in analyses predict putative plasmids, AMR β-lactamase gene variants, compared Thus, has potential reveal pathogenicity mixed samples.

Language: Английский

Citations

37

A culture-, amplification-independent, and rapid method for identification of pathogens and antibiotic resistance profile in bovine mastitis milk DOI Creative Commons

Asal Ahmadi,

Abdolrahman Khezri, Håvard Nørstebø

et al.

Frontiers in Microbiology, Journal Year: 2023, Volume and Issue: 13

Published: Jan. 6, 2023

Rapid and accurate diagnosis of causative pathogens in mastitis would minimize the imprudent use antibiotics and, therefore, reduce spread antimicrobial resistance. Whole genome sequencing offers a unique opportunity to study microbial community resistance (AMR) mastitis. However, complexity milk samples presence high amount host DNA from infected udders often make this very challenging.Here, we tested 24 bovine (18 six non-mastitis) using four different commercial kits (Qiagens' DNeasy® PowerFood® Microbial, Norgens' Milk Bacterial Isolation, Molzyms' MolYsis™ Plus Complete5) combination with filtration, low-speed centrifugation, nuclease, 10% bile extract male (Ox bile). Isolated was quantified, checked for presence/absence pathogen PCR sequenced MinION nanopore sequencing. Bioinformatics analysis performed taxonomic classification gene detection.The results showed that designed explicitly bacterial isolation food dairy matrices could not deplete/minimize DNA. Following Complete 5 + Ox micrococcal nuclease combination, on average, 17% 66.5% reads were classified as Staphylococcus aureus reads, respectively. This also effectively enriched other pathogens, including Escherichia coli Streptococcus dysgalactiae. Furthermore, approach, identified important AMR genes such Tet (A), (38), fosB-Saur, blaZ. We even 40 min run enough identification detecting first gene.We implemented an effective method (sensitivity 100% specificity 92.3%) removal enrichment (both gram-negative positive) directly milk. To best our knowledge, is culture- amplification-independent nanopore-based metagenomic real-time detection (within hours) profile 5-9 hours), samples. These provide promising potential future on-farm adaptable approach better clinical management

Language: Английский

Citations

14

Machine learning and feature extraction for rapid antimicrobial resistance prediction of Acinetobacter baumannii from whole-genome sequencing data DOI Creative Commons
Yue Gao, Henan Li, Chunjiang Zhao

et al.

Frontiers in Microbiology, Journal Year: 2024, Volume and Issue: 14

Published: Jan. 11, 2024

Whole-genome sequencing (WGS) has contributed significantly to advancements in machine learning methods for predicting antimicrobial resistance (AMR). However, the comparisons of different AMR prediction without requiring prior knowledge remains be conducted.

Language: Английский

Citations

5

Systematic Analysis of Mobile Genetic Elements Mediating β-Lactamase Gene Amplification in Noncarbapenemase-Producing Carbapenem-Resistant Enterobacterales Bloodstream Infections DOI Creative Commons
William C. Shropshire, Anna Konovalova, Patrick M. McDaneld

et al.

mSystems, Journal Year: 2022, Volume and Issue: 7(5)

Published: Aug. 29, 2022

Noncarbapenemase-producing carbapenem-resistant

Language: Английский

Citations

22