AI-Driven Antimicrobial Intelligence: AMR Genomics, Microbial Prediction & Next-Generation Drug Discovery

Workshop Structure

📅 Day 1: AMR Genomics and Resistance Intelligence

• Focus: Understanding antimicrobial resistance through bacterial genomics and resistance profiling.
• AMR fundamentals, multidrug resistance, and major resistance mechanisms.
• Chromosomal mutations, resistance genes, plasmids, and horizontal gene transfer.
• Bacterial whole-genome sequencing and the FASTQ → assembly → annotation workflow.
• Understanding microbial resistomes and clinically important resistant pathogens.
• AMR genomic surveillance and outbreak intelligence.
• Detection and interpretation of resistance genes, mutations, antibiotic classes, and mechanisms using CARD/RGI and AMRFinderPlus.

🛠️ Hands-on:

• Build and interpret a basic genomic AMR profile from a bacterial dataset.

🧰 Tools Covered: Google Colab, Python, Pandas, NCBI, AMRFinderPlus, CARD/RGI

📅 Day 2: AI-Based Microbial Resistance Prediction

• Focus: Using machine learning to predict antimicrobial-resistance patterns from microbial data.
• Genotype–phenotype relationships and resistant, intermediate, and susceptible phenotypes.
• Preparing genomic, metadata, and susceptibility datasets for machine learning.
• Feature engineering, data preprocessing, and train–test preparation.
• Resistance classification using Logistic Regression, Random Forest, and gradient boosting.
• Model evaluation using accuracy, precision, recall, F1 score, and ROC-AUC.
• Feature importance and explainable AI for identifying major resistance predictors.

🛠️ Hands-on:

• Build and evaluate a machine-learning model for antimicrobial-resistance prediction.

🧰 Tools Covered: Google Colab, Python, Pandas, NumPy, Scikit-learn, Matplotlib

📅 Day 3: AI-Driven Antimicrobial Drug Discovery

• Focus: Applying cheminformatics and AI for antimicrobial activity prediction and candidate prioritisation.
• Linking AMR intelligence with antimicrobial drug-discovery strategies.
• Exploring antimicrobial compounds and bioactivity data from PubChem and ChEMBL.
• Molecular representations: SMILES, descriptors, fingerprints, and RDKit.
• QSAR and machine-learning approaches for antimicrobial activity prediction.
• AI-based virtual screening and candidate ranking.
• Drug-likeness, toxicity, ADMET, selectivity, and emerging AI approaches in antibiotic discovery.

🛠️ Hands-on:

• Build an AI-assisted workflow to screen and rank promising antimicrobial candidates.

🧰 Tools Covered: Google Colab, Python, RDKit, Scikit-learn, DeepChem, PubChem, ChEMBL

Who Should Enrol?

Important Dates