Mots-Clés

imaging-genetics data integration machine learning data science

Description

NeuroSpin is an outstanding research center on the Human brain. Part of the CEA (Atomic Energy Commission) and Paris-Saclay University, the NeuroSpin teams are leaders in very high field MRI and carry out fundamental and clinical neurosciences studies. The BrainOmics team (http://www.brainomics.org) works in imaging-genetics, at the crossroad where neuroinformatics, bioinformatics and machine learning meet and in collaboration with Gustave Roussy, ICM-La Pitié-Salpétrière and CentraleSupelec.

Machine learning in imaging-genetics on EU-AIMS

The post-doc researcher will design new machine learning models incorporating multi-modal MRI and multi-omic data for prediction and stratification purposes, especially in ill-posed problems (very large p, small n). Then, he/she will perform the imaging-genetics analyses of the EU-AIMS cohort (aims-2-trials.eu), dedicated to Autism Spectrum Disorder (ASD).

Post-doc Activities

• Genotype microarray, brain anatomical and functional MRI data integration.
• Develop innovative and integrative machine learning prediction and stratification models.
• Applications in clinical neurosciences, ASD.

Searched profile

PhD in one of the following fields : Data Science, Machine Learning, Applied Statistics, Data integration, Neuro-Imaging, Genomics. Fluent in english.

Job-related skills

• Excellent skills in statistics and applied mathematics.
• Programming : Python, R, Matlab.
• Curiosity, taste for a multi-disciplinary environment and innovation.
• Knowledge in biomedical image analysis and/or genetics and/or neuroscience is an asset.

Behavioral skills

Good team player, strong motivation, rigor, autonomy and resourcefulness.

Duration : 14 months, starting in January 2022.

Location : NeuroSpin-CEA, Plateau de Saclay, Gif-sur-Yvette.

1. Mihailov et al. (2020). Cortical signatures in behaviorally clustered autistic traits subgroups: a population-based study. Translational Psychiatry.
2. Guigui et al. (2019) Network Regularization in Imaging Genetics Improves Prediction Performances and Model Interpretability on Alzheimer's Disease. ISBI 2019 Venise, Italie.
3. Marquand et al. (2016). Beyond lumping and splitting: a review of computational approaches for stratifying psychiatric disorders. Biol Psy: CNNI.