Mots-Clés
Multimodal biomarkers
epigenetic
machine learning
high-throughput sequencing
psychiatry
psychosocial factors
addiction
Description
Missions
The recruited fellow will join the BEBOP project on epigenetic and psychosocial biomarkers of addiction, led by Dr. Pierre-Eric Lutz at INCI. Their work will be part of a collaborative project involving psychiatrists, sociologists, and neurobiologists, aiming to identify biomarkers for opioid use disorder (or addiction). These biomarkers are meant to be “multimodal”, meaning that they combine epigenetic measurements from blood samples with the assessment of social and psychiatric factors that are known to contribute to the severity of the disorder.
The project will rely on a multicenter cohort (n=275 individuals recruited in Lyon, Paris, and Strasbourg) and a large, already existing dataset. The primary objective will be to simultaneously analyze all these factors to predict the severity of opioid use disorder, using bioinformatics and machine learning approaches. Ultimately, such an identification of multimodal biomarkers could enable the development of personalized interventions, aligning with the goals of precision medicine.
The initial 20-month contract may be extended.
Activities
The recruited individual will be responsible for conducting bioinformatics analyses of sequencing data, focusing on gene expression (3’RNA-Sequencing) and DNA methylation (Enzymatic Methylation Sequencing). They will then be responsible for applying machine learning methods to integrate our multimodal data (epigenetic, psychiatric, and social), adopting an integration approach that includes the following steps and procedures:
- Dimensionality reduction and feature selection (e.g., using penalized regression approaches such as Sparse Generalized Canonical Correlation Analysis, SGCCA);
- Integration through fusion of similarity networks between individuals (e.g., Similarity Network Fusion, SNF) to identify clusters related to the severity of opioid use disorder;
- Predictive modeling of addiction severity (e.g., using classifiers such as Random Forests or Neural Networks), including cross-validation procedures with bootstrapping;
- Co-expression network analyses to enhance the biological interpretability of predictive models.
Skills
1) Previous experience in bioinformatics research is required (PhD level).
2) Previous experience in using R (preferably) or Python is required, as well as knowledge in biology.
3) Experience in the following areas is also desirable: Processing and analysis of transcriptomic or epigenomic data; Use of a high-performance computing center and a task manager (e.g., Slurm); Use of virtual environments (e.g., Conda); Use of standardized analysis pipelines (e.g., Snakemake).
4) The recruited individual shall demonstrate autonomy and organizational skills.
5) The recruited individual will work as part of a team with our laboratory staff involved in this project, as well as with our collaborators in Paris, Lyon and Strasbourg.
Work environment
The recruited individual will be employed by the French National Centre for Scientific Research (CNRS) and will work at the Institute of Cellular and Integrative Neuroscience (INCI, UMR 3212), located on the Esplanade Campus of the University of Strasbourg, in the heart of Strasbourg, France. The institute benefits from an outstanding scientific environment, bringing together numerous research units in biology, neuroscience, and health sciences, and is easily accessible by public transportation (tram and bus).
At INCI, the recruited individual will join the “Pain and Psychopathology” team, led by Dr. Ipek Yalcin. The team currently consists of about 20 to 30 members, including around ten PhD students, one post-doctoral researcher, two research engineers, three study engineers, and five permanent researchers. It offers a stimulating, friendly, and international work environment, with many nationalities represented.
Additionally, to carry out the project, the recruited individual will have access to the resources of the University of Strasbourg’s High-Performance Computing Center (CCUS, see: https://hpc.pages.unistra.fr/). This project is also conducted in collaboration with a mathematician specializing in machine learning approaches, Dr. Vincent Vigon (Institute for Advanced Mathematical Research in Strasbourg, IRMA UMR7501).
Other accomodations include:
- 46 days of annual leave and RTT days per year,
- Partial reimbursement of commuting expenses in accordance with CNRS regulations,
- Access to subsidized catering facilities located near the institute,
- Access to social, cultural, and welfare benefits provided by CNRS,
- Opportunities for continuing education and professional development,
- Possibility of occasional remote work, in accordance with CNRS regulations and subject to compatibility with the experimental requirements of the project.