En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)
Portail > Offres > Offre UMR7093-JEAIRI-002 - Post-doc (H/F) : Modélisation de la distribution d'espèces et de groupes de gènes à l'échelle globale via apprentissage supervisé sur les données Tara Océans - Projet Blue Cloud

Post-doc on global scale species and gene clusters distribution modelling from Tara Oceans data using machine learning - Blue Cloud project (H/F)

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : lundi 15 mars 2021

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

Faites connaître cette offre !

General information

Reference : UMR7093-JEAIRI-002
Date of publication : Monday, February 22, 2021
Type of Contract : FTC Scientist
Contract Period : 12 months
Expected date of employment : 1 June 2021
Proportion of work : Full time
Remuneration : Between 2675.28 and 3805.92€ / month before taxes, depending on experience, for an experience < 7 years.
Desired level of education : PhD
Experience required : Indifferent


The scientific goal of this post-doctoral position is to predict the potential distribution of planktonic entities at the scale of the world's ocean based on a limited number of points where we know their concentrations. Knowing that plankton is strongly constrained by the conditions of the water mass it lives in, we will quantify the relationships between plankton and its environment at the points where we have data and then project those relationships at global scale (a.k.a. species distribution modelling).
This approach is classic but the main originality of this project is that the entities to predict will come from a deep analysis of the meta-omics data of Tara Oceans by a current post-doc. They may therefore be “species” but also clusters of genes associated with a function or coherent entities whose identity is unknown (i.e. genomic “dark matter”). The modelling of their distribution will therefore help in the interpretation of their nature and their role in the ecosystems.
While we will start with established methods for a few entities, to provide a baseline, further possible developments include: (i) predicting multiple entities simultaneously to exploit the information of their relative concentrations, (ii) using deep learning methods for species distribution modelling to capture environmental context, (iii) using ~3000 points from a worldwide, quantitative, imaging dataset and fine resolution satellite data to compare with the Tara Oceans one. These developments, and others, will be discussed and decided with the selected candidate.
The goal of the Blue Cloud project that funds this position is to provide a scientific infrastructure for others to use. Therefore, the work will be embedded into this EU-scale infrastructure and should be made accessible through well documented code or a web-based application.
Finally, it is expected from the post-doc to disseminate his/her work within the Blue Cloud community (especially during dedicated workshops) and beyond (international conference, scientific publication, code and product sharing, etc.).


- Communicate with the current post-doc to understand the output of the meta-omics analysis
- Assemble the database of occurences/concentrations and environmental data
- Model the relationship between presence/concentration and environmental data
- Project it at global scale
- Prepare scientific papers and presentations


- Hold or be close to completion of a Ph.D. in numerical ecology or applied maths.
- Experience with machine learning methods, ideally with deep learning (in a context of regression, not classification)
- Basic knowledge of ecology and ideally experience with species distribution models (a.k.a. niche models)
- Excellent writing and communication skills to write papers and present findings, but also write basic code documentation.
- Good programming skills and ability to handle large datasets, with R and/or Python
- Experience with querying netCDF files would be a plus
- Scientific and technical English (B2 level for written and oral). French would be a plus, but it is not mandatory.

Work Context

The Laboratory of Oceanography of Villefranche-sur-mer (LOV; http://lov.obs-vlfr.fr/) belongs to one of the three marine stations of Sorbonne Université; it is located close to Nice, on the French Riviera. With about 40 permanent staff and 90 total staff, the LOV is a leading research laboratory in oceanography, tackling questions ranging from marine ecology to biogeochemistry, producing ~100 scientific papers per year, and receiving prestigious grants from ANR, the European Commission and various international funding agencies.

The COMPLEx (COMPutational PLankton Ecology) team gathers ~25 members studying marine plankton. Plankton encompasses all organisms drifting with marine currents. Those organisms are responsible for producing half the oxygen we breathe, storing the carbon we emit, feeding the fish we eat; plankton is therefore a major building block of Earth's ecosystem. In COMPLEx, we use numerical methods (modeling, statistics, machine learning), quantitative imaging instruments and, increasingly, meta-omics data to further our understanding of plankton. We strongly interact with the Quantitative Imaging Platform of Villefranche (PIQv; https://sites.google.com/view/piqv), which oversees the daily operation of tools that the team develops. Those tools include imaging sensors, such as the Underwater Vision Profiler or the ZooScan, and software packages, such as ZooProcess or the EcoTaxa web application (https://ecotaxa.obs-vlfr.fr/) that uses machine learning to assist taxonomists into sorting plankton images. COMPLEx has a large network of collaborators, in oceanography, imaging, genomics, and computer sciences, in particular as part of the Tara Oceans consortium.

Constraints and risks


Additional Information

For more information, please contact Lionel GUIDI (guidi@obs-vlfr.fr), Sakina-Dorothée AYATA (sakina@obs-vlfr.fr), and Jean-Olivier IRISSON (irisson@obs-vlfr.fr)

We talk about it on Twitter!