Translational Genomics of pancreatic Neoplasia Heterogeneity
The GeNeHetX lab
The GeNeHetX lab aims to decipher the phenotypic heterogeneity of pancreatic diseases to derive clinically-applicable phenotyping tools. The team gathers expertise from the fields of computational biology, pathology and clinical oncology to develop novel diagnostic tools and therapeutic options in pancreatic cancer.
The lab was an ATIP-Avenir program of excellence laureate in 2022 fouding the GeNeHetX lab in the INSERM (Institut national de la santé et de la recherche médicale) as part of the Inflammation Research Center (cri1149.fr).
As part of our work on molecular tissue imaging, we are affiliated to the FHU MOSAIC (Fédération Hospitalo-Universitaire, Multiscale Optimised Strategy for Artificial intelligence-based Imaging biomarkers in digestive Cancer fhu-mosaic.com).
Our work focuses on two main axes to characterize the heterogeneity of pancreatic tumors, a highly fibrotic, chemoresistant and aggressive cancer.
First to unravel the complex tissue architecture of pancreatic adenocarcinomas combining tools from the field of artificial intelligence with spatialized molecular profiling.
Then to leverage on the morpho-molecular heterogeneity models to develop clinical tools, based on molecular signatures and predictive models applied to standard histology.
Our group endeavors to transfer phenotypic description tools, in particular RNA theranostic signatures, for personalized medicine.
PACpAInt: A computer vision approach based on deep learning methods to phenotype Pancreatic Adenocarcinoma both at the tumor and local level (112μm-wide square tiles). BiorXiv Prepint
Nicolle, R., Gayet, O., Duconseil, P., Vanbrugghe, C., Roques, J., Bigonnet, M., ... & Dusetti, N. J. (2021). A transcriptomic signature to predict adjuvant gemcitabine sensitivity in pancreatic adenocarcinoma. Annals of Oncology, 32(2), 250-260.
We are in the campus of the Beaujon Hospital (Clichy) in the outskirts of Paris, France.