Gabriele Gut

This paper introduces a method to predict single-cell perturbation responses using neural optimal transport, showing improved performance over existing methods on drug-treated cells.

scROSHI identifies cell types without training data by using pre-existing gene sets and hierarchical relationships, outperforming methods in low-data and high-diversity scenarios.

This paper develops a multimodal approach to study human retinal organoid development, integrating spatial, transcriptomic, and chromatin accessibility data to create a virtual atlas of retinal development.

This study uses iterative immunofluorescence imaging to analyze age-related changes in the dentate gyrus niche, identifying reorganization and microenvironmental changes affecting neurogenesis in mice.

SCIM matches single-cell measurements across technologies by learning a technology-invariant latent space and applying bipartite matching, achieving high accuracy in real-world datasets.