Ilias Tagkopoulos

The paper develops a pipeline using large language models to extract food-chemical relationships from literature, identifying many novel interactions.

The paper develops machine learning models to predict nutrient profiles of cooked foods from raw composition and cooking method, achieving 31% better performance than baselines.

The paper builds a normalized synovial transcriptomic compendium of RA patients to identify gene expression signatures predictive of treatment response to infliximab and other drugs.

This paper develops a network-based model to identify drug repositioning targets for RA, finding that anti-diabetic agents have similar network signatures to DMARDs and predicting synergistic drug combinations.

The paper investigates cross-stress responses in Escherichia coli by analyzing genetic and transcriptional changes under multiple abiotic stresses, identifying cases of protection and vulnerability.