Susanna Zucca

PhenoXtract extracts phenotypes from clinical descriptions using a hybrid approach of Large Language Models and Knowledge Graph embeddings, achieving high recall and precision.

diVas interprets digenic variants using hypothesis-driven machine learning and XAI, achieving 73% sensitivity in prioritizing causative gene pairs across 11 cases.

The paper evaluates variant prioritization methods for rare disease diagnosis using data from the Rare Genomes Project, finding that top models identify causal variants in most solved families and uncover potential candidates in unsolved ones.

The paper compares RNA profiles in plasma-derived large and small extracellular vesicles from neurodegenerative disease patients and healthy controls, identifying disease-specific miRNA signatures.