Antoine Allard

The paper evaluates using indicator species from co-occurrence networks to predict biodiversity, finding they can reconstruct species assemblages but are often outperformed by climate-based models.

The paper develops a framework using deep learning and graph theory to analyze neuronal network maturation from DHM images, identifying key topological changes during in vitro development.

The paper shows functional connectivity gradients in zebrafish brains align with geometric eigenmodes when connectivity is local, but this alignment breaks down with more long-range connections.

This paper introduces NBS-SNI, an extension of NBS that integrates node properties with network analysis to detect connectivity abnormalities in case-control studies, showing improved prediction power.

This paper develops a network-based method to reconstruct biodiversity using indicator species, finding that latitudinal gradients affect species composition and interaction patterns, with higher complexity in tropical regions.