1 June 2026
Glasgow, UK — Researchers from REPROCELL Europe have co-authored a new study published in Scientific Reports demonstrating how biomedical foundation models and multi-omics feature engineering can improve the prediction of drug responses in patients with inflammatory bowel disease (IBD).
The paper, titled “Multi-omics feature engineering driven by biomedical foundation models improves drug response prediction for inflammatory bowel disease patients,” was co-authored by REPROCELL scientists alongside collaborators from IBM Research Europe and the STFC Hartree Centre, who supported the technical validation of the AI application used in the study.
The study explores how artificial intelligence (AI) and machine learning approaches can be applied to complex biological datasets to better predict patient responses to therapeutic compounds. By integrating transcriptomic and multi-omics data with biomedical foundation models, the study demonstrates improved predictive performance for drug response modelling in IBD patients.
Key findings from the research include:
- Improved drug response prediction: The AI-driven workflow enhanced the ability to predict therapeutic responses in inflammatory bowel disease patients compared to conventional modelling approaches.
- Multi-omics integration: Combining diverse biological datasets enabled the development of richer and more informative predictive models.
- Foundation model applications in pharmacology: The study highlights the growing potential of biomedical foundation models to support translational research and precision medicine initiatives.
- Potential for patient stratification: The approach may help identify which patients are most likely to benefit from specific therapies, supporting more targeted drug development strategies.
Advancing AI-enabled pharmacology research
The publication reflects REPROCELL’s continued investment in innovative approaches that combine human tissue-based research with advanced computational and AI technologies. As the pharmaceutical industry increasingly seeks more predictive preclinical models, AI-driven analysis of complex biological data is becoming an important tool for improving translational success.
By leveraging multi-omics datasets and machine learning approaches, studies such as this may help researchers better understand disease mechanisms, identify novel biomarkers, and improve the prediction of therapeutic efficacy earlier in the drug discovery process.
The findings are particularly relevant for inflammatory bowel disease, where patient responses to therapies can vary significantly and where improved predictive approaches could help accelerate the development of more personalized treatment strategies. REPROCELL Europe has conducted more than 60 IBD projects for global pharmaceutical sponsors and continues to broaden its capabilities to improve the prediction of human clinical outcomes throughout the drug development process.
The full paper is available in Scientific Reports here: https://www.nature.com/articles/s41598-026-44366-y
About REPROCELL
REPROCELL Inc. is based in Yokohama, Japan, with subsidiaries in Europe (REPROCELL Europe Ltd), USA and India. REPROCELL supports the advancement of drug discovery and regenerative medicine through human tissue research services, preclinical pharmacology solutions, stem cell technologies, and translational research platforms. The company works with pharmaceutical, biotechnology, and academic partners worldwide to deliver disease-relevant human biology solutions that improve translational research outcomes.