Matthieu Defrance

RedRibbon is a rank-rank hypergeometric overlap method for comparing gene and transcript expression signatures, showing accuracy and efficiency on large datasets.

The paper compares the transcriptomic heat stress responses of desert and temperate ants to identify convergent and specific adaptations to thermal stress.

The paper compares differential expression methods for scRNA-seq clustering models, finding gradient-based methods outperform traditional approaches.

The paper introduces a method for subspace clustering of omic data and shows it outperforms existing techniques in identifying ground-truth subspaces.

The paper creates TIGER, a resource integrating RNA-seq and genotyping data from human pancreatic islets to identify eQTLs and their associations with type 2 diabetes.

The paper combines scRNA-seq with a molecular atlas to identify new neuron class markers in C. elegans, assigning 58 clusters to known neuron classes and validating 15 new promoters.

The paper shows that mesoderm cells in embryonic and extra-embryonic regions of the mouse embryo have distinct migration behaviors and cytoskeleton regulation.