AI Predicts E. coli Antibiotic Resistance in Agriculture
Scientists at the University of the Philippines–Diliman College of Science have developed an innovative approach to predict E. coli's resistance to antibiotics using artificial intelligence (AI) in agricultural settings. E. coli, a bacterium commonly found in the intestines of animals and humans, is often used to detect fecal contamination. It is prone to developing antibiotic resistance, making it crucial for monitoring in environments like farms where manure and wastewater are used. Traditional methods for testing antibiotic resistance are slow and labor-intensive, prompting researchers to explore faster alternatives like whole-genome sequencing (WGS) combined with AI.
AI Models2 Tested for Resistance Prediction
Dr. Pierangeli Vital and Marco Christopher Lopez, alongside Dr. Joseph Ryan Lansangan, tested several AI models to predict antibiotic resistance using genetic data and results from the National Center for Biotechnology Information database. The models included Random Forest (RF), Support Vector Machine (SVM), and two ensemble methods—Adaptive Boosting (AB) and Extreme Gradient Boosting (XGB). These models were chosen for their strengths in handling biological and imbalanced data. The AI models were particularly effective at predicting resistance to streptomycin and tetracycline, with AB and XGB proving especially reliable.
Also Read: SandboxAQ Releases AI-Generated Data to Accelerate Drug Discovery
Potential for Real-Time Monitoring in Agriculture
The study demonstrates AI's potential for real-time monitoring of antimicrobial resistance in agriculture. As DNA sequencing becomes faster and more affordable, AI models could detect resistant bacteria early, preventing outbreaks and enhancing food safety, agriculture, and public health initiatives. Dr. Vital emphasized the importance of collaboration between microbiologists and statisticians, and the integration of metagenomic data for more accurate predictions. This innovative approach opens new doors for faster, more effective monitoring of antibiotic resistance in agricultural systems.
