AI + Probiotic Genomics
Use genomic features to rank strains for safety, stress tolerance, functional traits and probiotic potential.
Microbial AI for safer biological decisions.
The programme combines genomic and microbiome features with laboratory evidence to support probiotic strain selection, safety screening, functional prediction and product assessment.
01 / MICROBIOLOGY + AI
Data should strengthen biological reasoning.
Microbial AI at the Probiotic Genomics & AI Laboratory is positioned as a support system for research questions. It does not replace experimental validation. It helps organize biological evidence, identify patterns, prioritize candidates and make complex data easier to interpret.
Published work includes machine-learning models for therapeutic peptide prediction and computational analyses of probiotic-derived peptides. Current project work extends AI/ML toward animal probiotic performance and decision support.
See computational publicationsPositioning
AI-enabled microbiology and probiotic-product research, with AMR reduction and One Health relevance.
Evidence chain
Genome features → biological hypotheses → laboratory validation → animal/product outcomes.
02 / FOUR WORKSTREAMS
From sequences to decisions.
The research direction develops through four connected workstreams that link genomic evidence, biological function, animal performance and One Health outcomes.
AI + Therapeutic Peptides
Machine-learning prediction, docking and molecular-dynamics analysis of biologically active peptide candidates.
AI + Animal Performance
AS-0236 connects microbial strain data, feed supplementation and poultry performance with ML-supported assessment.
AI + AMR / One Health
Integrate microbial genomes, gut health, animal production and safer food systems into one risk-aware narrative.
03 / AMR & ONE HEALTH
Probiotic alternatives to antibiotics
Select and validate beneficial strains that can support gut health and reduce reliance on routine antimicrobial use.
Genomic AMR / safety interpretation
Use WGS/NGS to screen resistance, virulence and safety-associated markers before product development.
AI/ML for microbial risk
Integrate genomic, microbiome, phenotype, feed and farm data for prediction and decision support.
One Health translation
Connect animal gut health, food production, microbial ecology and wider AMR risk into one research direction.
Future / current research direction
Current work is exploring AI-guided development of animal probiotic feed additives for poultry gut health, performance assessment and responsible AMR reduction.
04 / EMERGING DIRECTIONS
Questions for the next stage.
The direction remains connected to ongoing research, collaboration and future validation.
- AI-guided probiotic strain ranking from WGS and phenotype data.
- AI-supported AMR and virulence/safety screening for candidate strains.
- Virtual-gut and in-silico host-microbe modelling for probiotic-response hypotheses.
- Multi-omics integration: microbial genome → microbiome shift → host response → animal performance.
- Precision probiotic feed: predict the best strain combination, dose and response profile.