AI-Driven Phage Therapy: Designing Precision Antimicrobial Solutions Against Drug Resistance

To provide researchers and professionals with practical skills to apply bioinformatics and machine learning in phage therapy, host prediction, cocktail optimisation, and antimicrobial candidate discovery.

• Understand the role of bacteriophages in combating antimicrobial resistance.
• Analyse phage genomic data for host prediction.
• Apply machine learning for phage–host classification.
• Design rational and resistance-aware phage cocktails.
• Screen phages for therapeutic suitability and genomic safety.
• Identify and prioritise phage-derived antimicrobial proteins.
• Develop reproducible computational workflows using Google Colab.

📅 Day 1: AI-Based Phage–Host Prediction

• Introduction to bacteriophage therapy and antimicrobial resistance
• Lytic vs temperate phages and therapeutic suitability
• Understanding phage–host specificity and host range
• Genomic signals for host prediction:
  • k-mer patterns
  • GC content
  • CRISPR spacer matches
  • sequence similarity
  • codon usage
• Introduction to AI/ML-based phage–host prediction
• Overview of tools: iPHoP, CHERRY, PhaBOX and RaFAH
• Prediction confidence and interpretation

🛠️ Hands-On: Phage–Host Prediction

• Extract genomic features from phage FASTA sequences
• Train an ML model to predict bacterial hosts
• Generate host probability and confidence scores

🧰 Tools: Python, Biopython, pandas, scikit-learn

🗓️ Day 2: Rational Phage Cocktail Design

• Why single-phage therapy may fail
• Host-range and bacterial susceptibility matrices
• Phage resistance and receptor diversity
• Genomic safety screening of candidate phages
• Removing temperate, virulence and AMR-associated candidates
• Host coverage and redundancy analysis
• Resistance-aware phage selection
• Set-cover and optimisation approaches for cocktail design
• Phage–antibiotic synergy and sequential therapy concepts

🛠️ Hands-On: Optimise a Phage Cocktail

• Analyse a phage–bacteria host-range matrix
• Apply safety and coverage filters
• Select an optimised multi-phage combination

🧰 Tools: Python, pandas, NumPy, SciPy / NetworkX

🗓️ Day 3: AI-Guided Phage-Derived Antimicrobial Discovery

• Phages as sources of antimicrobial molecules
• Endolysins and phage-derived lysins
• Depolymerases and biofilm-targeting enzymes
• Protein sequence-based antimicrobial discovery
• Amino-acid and physicochemical feature extraction
• Introduction to protein embeddings and protein language models
• AI-based classification of phage proteins
• Candidate scoring and prioritisation
• Translational applications in AMR and precision antimicrobial therapy

🛠️ Hands-On: Antimicrobial Protein Prioritisation

• Analyse phage protein sequences
• Apply ML-based protein classification
• Rank endolysin/depolymerase candidates

🧰 Tools: Python, Biopython, pandas, scikit-learn, ESM/Hugging Face

This workshop is suitable for:

• Researchers and Research Scholars
• PhD and Postgraduate Students
• Academicians and Faculty Members
• Microbiologists and Molecular Biologists
• Bioinformatics and Computational Biology Researchers
• Antimicrobial Resistance Researchers
• Infectious Disease Researchers
• Biotechnology and Pharmaceutical Professionals
• Researchers working in Phage Biology, Microbial Genomics, Viromics and Alternative Antimicrobial Strategies

By the end of the workshop, participants will be able to:

• Predict probable bacterial hosts from phage genomic features.
• Interpret host-range and prediction confidence data.
• Construct phage–bacteria interaction matrices.
• Optimise multi-phage cocktails based on host coverage and resistance considerations.
• Apply genomic safety filters to candidate phages.
• Analyse phage protein sequences for antimicrobial potential.
• Prioritise endolysin and depolymerase candidates using AI/ML approaches.
• Apply computational phage therapy workflows in research projects.