DeepMRG: a multi-label deep learning classifier for predicting bacterial metal resistance genes
Bibliographic record
Abstract
Abstract The widespread misuse of antibiotics has escalated antibiotic resistance into a critical global public health concern. Beyond antibiotics, metals function as antibacterial agents. Metal resistance genes (MRGs) enable bacteria to tolerate metal-based antibacterials and may also foster antibiotic resistance within bacterial communities through co-selection. Thus, predicting bacterial MRGs is vital for elucidating their involvement in antibiotic resistance and metal tolerance mechanisms. The “best hit” approach is mainly utilized to identify and annotate MRGs. This method is sensitive to cutoff values and produces a high false negative rate. Other than the best hit approach, only a few antimicrobial resistance (AMR) detection tools exist for predicting MRGs. However, these tools lack comprehensive annotation for MRGs conferring resistance to multiple metals. To address such limitations, we introduce DeepMRG, a deep learning-based multi-label classifier, to predict bacterial MRGs. Because a bacterial MRG can confer resistance to multiple metals, DeepMRG is designed as a multi-label classifier capable of predicting multiple metal labels associated with an MRG. It leverages bit score-based similarity distribution of sequences with experimentally verified MRGs. To ensure unbiased model evaluation, we employed a clustering method to partition our dataset into six subsets, five for cross-validation and one for testing, with non-homologous sequences, mitigating the impact of sequence homology. DeepMRG consistently achieved high overall F1-scores and significantly reduced false negative rates across a wide range of datasets. It can be used to predict bacterial MRGs in metagenomic or isolate assemblies. The web server of DeepMRG can be accessed at https://deepmrg.cs.vt.edu/deepmrg and the source code is available at https://github.com/muhit-emon/DeepMRG under the MIT license.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".