Born: June 14, 1949, France
Former Associate Professor, INSERM
Head, Biomathematics and Bioinformatics
Department of Biological Hematology
CHRU Montpellier, Hopital Saint-Eloi
80, av. Augustin Fliche
34295 Montpellier - France
Tel. +33 685 900 512
2001 Master in Bioinformatics (Montpellier II)
1984 Doctor of Sciences (Montpellier II)
1982 Pharmacy (Montpellier I)
1979 Certified in Oncology (Paris XI)
1979 Medicine (Montpellier I, Paris VI, VII, XI)
1972 Engineer (ENSCM Montpellier)
Assistant/Associate Professor, INSERM Units 236, 291, 475, 847 and 1040, Montpellier
Postdoctoral research, Ludwig Institute, Lausanne, Switzerland, JC Cerottini.
Research project :
The aim of the department is to combine high throughput biological data (DNA chips, Q-PCR) and clinical data from documented patients to allow :
- building classification tools for diagnosis, prognostic and treatment response, starting from either recognized or personally developed biomathematical methods
- using such tools to predict classification and further therapeutic indication of every new patient as well as extracting new therapeutic targets from the data.
Medical diagnosis (30% of activity)
A INSERM-CHU interface contract allowed the development of a web tool http://rage.montp.inserm.fr with a friendly user interface for clinicians, allowing to filter, sort and analyze data with supervised or unsupervised methods, to provide dedicated files and links to other bioinformatics websites, to visualize and graphically compare results and survival curves (relapse, death)in selected cohorts, in the field of cancer (multiple myeloma, gliomas) as well as for medically assisted reproduction. This platform also provides quality control for the DNA chips processed in the Transcriptome platform of IRB.
Basic research (70% of activity) :
The present web interface has become an undispensable day-to-day tool for students and research groups using the IRB transcriptome platform alltogether. By combining biological and clinical data, new methods are currently developed to:
- refine diagnosis process: more precise determination of clinical stages for a easier classification of new patients
- select the best prognosis genes from survival data (event-free or overall) for an optimally adapted treatment
- find new potential therapeutic targets from such prognosis genes
We have published the first predictive tool based on simple presence or absence of expression of a gene list (BMC Bioinformatics - 2008).
A prognosis stratification of multiple myeloma, breast cancer and glioma using our risk signature for overall survival (Bioinformatics - 2013) is available through the "Prognoweb" website: https://gliserv.montp.inserm.fr/
We recently published a molecular predictor able to classify gade II/III human gliomas with a much stronger prediction of outcome compared to the WHO grade II/III classification (Plos ONE - 2013)
Our tools are actualized with the last published developments. They allowed to establish numerous collaborative studies either locally (CHU, INM, IGF), nation-wide (Toulouse, Paris, Besançon) or internationally (Heidelberg, Germany, Rotterdam, Netherlands, Little Rock, USA).