Keywords: Deep learning, Graph Neural Networks, bioinformatics, Cancer Prediction Internship Duration: 30/11/-1 - 30/11/-1
Head of the hosting team: Feng Chu
Website: Click here
Address of the host laboratory: IBISC Team AROB@S 23, Boulevard de France 91034 Evry France
Supervisor: Farida ZEHRAOUIE-mail: farida.zehraoui@univ-evry.fr Phone: +33164853464
Precision medicine, also called stratified medicine, mainly uses patients' genomic characteristics for personalized care. “Omics” technologies: genomics (DNA sequencing), transcriptomics (microarrays), and proteomics, have considerably modified the scale of data and made it possible to generate massive quantities of genomic data on patients. These data can cover all the mechanisms involved in the variations that occur in the cellular networks that influence the functioning of organ systems in humans. They can be used for diagnosis, prognosis, prediction of personalized patient treatment, etc. Artificial intelligence, particularly machine learning, has become a promising tool to support precision medicine in oncology over the past decade. Deep learning, a subfield of machine learning, will play a significant role in improving the accuracy of cancer susceptibility, recurrence, and survival predictions. As cancer is a heterogeneous disease, several subtypes can be identified. Treatments and diagnostics must be tailored to each subtype. In this project, we will focus on predicting subtypes of an ordinary human cancer, bladder cancer. The aim of this internship, which is part of a collaboration between the IBISC laboratory and the Curie Institute, is to develop a multi-source and multiscale machine learning based on graph neural networks to identify cancer subtypes and potentially discover new ones, using heterogeneous data sources representing different types of omics, images and clinical data associated with patients.
Graph Neural Networks, Integration of multiple sources, precision medicine
1. V. Bourgeais, F. Zeharoui, B. Hanczar. GraphGONet: a self-explaining graph-based neural network encapsulating the Gene Ontology for phenotype prediction on gene expression. Bioinformatics, 2022. 2. V. Bourgeais, F. Zeharoui, B. Hanczar. Deep GONet: Self-explainable deep neural network based on Gene Ontology for phenotype prediction from gene expression data. BMC bioinformatics 22, 455, 2021. 3. Graw, S., Chappell, K., Washam, C.L., Gies, A.J., Bird, J.T., Robeson, M.S., & Byrum, S.D. (2020). Multi-omics data integration considerations and study design for biological systems and disease. Molecular omics.
