Rapid prediction of antibiotic resistance in <i>Enterobacter cloacae</i> complex using whole-genome and metagenomic sequencing
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
ABSTRACT Clinical management and surveillance of the Enterobacter cloacae complex (ECC) face significant challenges due to inaccurate species identification and prolonged turnaround time for culture-based antimicrobial susceptibility testing (AST). To date, no studies have leveraged whole-genome sequencing (WGS) and metagenomic next-generation sequencing (mNGS) to develop a rapid AST prediction model for ECC. Here, a total of 1,054 ECC strain genomes with AST data were collected from a public database and a local hospital. The results of species identification between the average nucleotide identity (ANI)-based method on culture were compared, and machine learning was employed to identify resistance features for imipenem (IPM), meropenem (MEM), ciprofloxacin (CIP), levofloxacin (LEV), and trimethoprim-sulfamethoxazole (SXT). By referring to ANI-based species classification, culture-based methods showed a 74% misidentification rate for 1,054 ECC isolates. The antimicrobial resistance prediction model demonstrated good performance, with the area under the curve values of 91.25% (IPM), 89.69%, 88.17% (CIP), 91.01% (LEV), and 90.93% (SXT) respectively. Moreover, a combined WGS and mNGS approach was utilized and validated using 104 pediatric sputum specimens. Compared to culture-based AST, the overall accuracy of models exceeded 95%, especially achieving 100% for IPM and 98.80% for MEM, and the detection turnaround time was shortened by 69.64 h. Furthermore, it would enable early escalated therapy in 20.83% of cases, significantly improving patient management. This established WGS and mNGS-based AST prediction model addresses the limitations of traditional methods, offering a rapid, accurate, and clinically applicable tool for managing multidrug-resistant ECC infections. IMPORTANCE The Enterobacter cloacae complex (ECC) poses a major challenge to clinical management due to difficulties in accurate species identification and the slow turnaround times of conventional culture-based antimicrobial susceptibility testing (AST). Current methods are often inefficient and prone to misidentification, leading to delayed or inappropriate treatment. This study introduces a novel approach that combines whole-genome sequencing (WGS) and metagenomic next-generation sequencing (mNGS) to develop a rapid and accurate AST prediction model for ECC. By leveraging machine learning to analyze WGS data from over 1,000 ECC isolates and validating the model with pediatric clinical specimens. The model achieved over 88% area under the curve accuracy for all antibiotics, demonstrated >95% accuracy in clinical validation, and reduced detection turnaround time by 69.64 h compared to traditional methods. The model has the potential to revolutionize ECC management by facilitating timely, targeted therapies and enhancing patient outcomes, especially in the context of multidrug-resistant infections.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it