Related Experiment Video
Updated: Jul 5, 2025
Isolation and Identification of Waterborne Antibiotic-Resistant Bacteria and Molecular Characterization of their Antibiotic Resistance Genes
Published on: March 3, 2023
Predicting Salmonella MIC and Deciphering Genomic Determinants of Antibiotic Resistance and Susceptibility
Moses B Ayoola1, Athish Ram Das1, B Santhana Krishnan1
1Department of Comparative Biomedical Sciences, College of Veterinary Medicine, Mississippi State University, Starkville, MS 39762, USA.
Abstract:
Salmonella spp., a leading cause of foodborne illness, is a formidable global menace due to escalating antimicrobial resistance (AMR). The evaluation of minimum inhibitory concentration (MIC) for antimicrobials is critical for characterizing AMR. The current whole genome sequencing (WGS)-based approaches for predicting MIC are hindered by both computational and feature identification constraints. We propose an innovative methodology called the "Genome Feature Extractor Pipeline" that integrates traditional machine learning (random forest, RF) with deep learning models (multilayer perceptron (MLP) and DeepLift) for WGS-based MIC prediction. We used a dataset from the National Antimicrobial Resistance Monitoring System (NARMS), comprising 4500 assembled genomes of nontyphoidal Salmonella, each annotated with MIC metadata for 15 antibiotics. Our pipeline involves the batch downloading of annotated genomes, the determination of feature importance using RF, Gini-index-based selection of crucial 10-mers, and their expansion to 20-mers. This is followed by an MLP network, with four hidden layers of 1024 neurons each, to predict MIC values. Using DeepLift, key 20-mers and associated genes influencing MIC are identified. The 10 most significant 20-mers for each antibiotic are listed, showcasing our ability to discern genomic features affecting Salmonella MIC prediction with enhanced precision. The methodology replaces binary indicators with k-mer counts, offering a more nuanced analysis. The combination of RF and MLP addresses the limitations of the existing WGS approach, providing a robust and efficient method for predicting MIC values in Salmonella that could potentially be applied to other pathogens.
Insights
This study introduces a new pipeline using machine learning and deep learning to predict antimicrobial resistance in Salmonella. The method accurately identifies genomic features influencing minimum inhibitory concentration (MIC) values.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Salmonella spp. is a major cause of foodborne illness, with antimicrobial resistance (AMR) posing a significant global threat.
- Accurate Minimum Inhibitory Concentration (MIC) evaluation is crucial for understanding and combating AMR.
- Existing whole genome sequencing (WGS) methods for MIC prediction face computational and feature identification challenges.
Purpose of the Study:
- To develop an innovative WGS-based methodology for predicting MIC values in Salmonella.
- To integrate traditional and deep learning models for enhanced AMR characterization.
- To identify specific genomic features that influence Salmonella's resistance to antibiotics.
Main Methods:
- Developed the "Genome Feature Extractor Pipeline" combining random forest (RF) and deep learning (MLP, DeepLift).
- Utilized a dataset of 4500 Salmonella genomes from the National Antimicrobial Resistance Monitoring System (NARMS) with MIC metadata.
- Employed RF for feature importance, Gini-index for selecting 10-mers, expanding to 20-mers, and MLP for MIC prediction, with DeepLift for feature identification.
Main Results:
- The pipeline successfully predicted MIC values for 15 antibiotics in Salmonella.
- Identified the 10 most significant 20-mers influencing MIC for each antibiotic.
- The methodology demonstrated enhanced precision in discerning genomic features affecting Salmonella MIC prediction.
- Replaced binary indicators with k-mer counts for a more nuanced analysis.
Conclusions:
- The proposed pipeline offers a robust and efficient WGS-based approach for predicting Salmonella MIC values.
- This integrated machine learning and deep learning strategy overcomes limitations of current WGS methods.
- The methodology holds potential for application to other bacterial pathogens for AMR surveillance and prediction.
