INRAE presentation: The French National Research Institute for Agriculture, Food, and Environment (INRAE) is a major player in research and innovation. It is a community of 12,000 people with 272 research, experimental research, and support units located in 18 regional centres throughout France. Internationally, INRAE is among the top research organisations in the agricultural and food sciences, plant and animal sciences, as well as in ecology and environmental science. It is the world’s leading research organisation specialising in agriculture, food and the environment.
INRAE’s goal is to be a key player in the transitions necessary to address major global challenges. Faced with a growing world population, climate change, resource scarcity, and declining biodiversity, the Institute has a major role to play in building solutions and supporting the necessary acceleration of agricultural, food and environmental transitions.
Context: Arboriculture is an essential part of French agriculture, contributing to a healthy, local diet, while diversifying crops and storing carbon in the soil. However, this sector is fragile and in great need of varietal innovation, particularly to adapt varieties to climate change and pesticide reduction as part of the agro-ecological transition. A key challenge in achieving this, is to propose innovative solutions for rapid and accurate phenotyping of fruit genetic resources. To this end, and as part of the “Agroecology and Digital” PEPR program, teams from INRAE’s EMMAH, GAFL and AGAP research units are developing tools and methods based on digital technologies.
In addition to methodological developments in stereovision and RGB (Red Green Blue) data processing to access these traits, the PHADA (PHénotypage HAut-débit de la Diversité génétique des Arbres fruitiers pour une meilleure adaptation au changement climatique) project focuses on fusing information, as well as the generalization and transferability of methods and traits between species. Several approaches (machine and deep learning, mathematical morphology, statistics) will be combined.
This project is based on the acquisition of datasets in core-collections of three major fruit species (peach, apricot and apple), each comprising over 150 different genotypes, thus ensuring representativeness of variability in flowering and tree structure through differences in age, species and contrasting environments. It also aims to establish temporal consistency between past measurements (visual notations) and digital phenotyping.
Activities To develop these cross-species digital phenotyping methods and enable an initial genetic screening of the GAFL and AGAP core collections, the candidate’s activities will focus on 4 tasks:
- A literature review: this will allow identifying the traits of interest and the methods already developed in the literature. The traits targeted will concern flowering (e.g. date, floribundity, distribution of flowers or flower clusters in the tree and on twigs) and tree architecture in the broadest sense (shape, convex hull, wood volume and growth, trunk diameter, type and distribution of twigs, density of associated organs, foliar and wood angles, leaf area index).
- The organization of experimental campaign(s) and management of acquired data: the study will be based on data already acquired in the past. Two additional campaigns will be conducted in the three core-collections located in the south-east quarter of France (Hérault, Gard and Vaucluse). Acquisitions will be made mainly with the LITERAL device developed at CAPTE, and equipped with two stereovision RGB cameras. The post-doctoral fellow will supervise the two acquisition campaigns, and may be supported by available technical staff. The multi-year monitoring of these core-collections will enable to study the kinetics of certain traits (wood volume, for example).
- The development of image processing tools for estimating generic and specific traits: the obejctive is to find effective approaches for identifying and quantifying different organs (wood, flowers, leaves, fruit) or traits (tree volume, leaf surface, etc.). These approaches will combine image analysis methods in mathematical morphology (e.g. determination of convex envelopes, characterization of the spatial distribution of flowers or twigs) and deep learning (semantic and instance segmentation). The information provided by stereovision (3D) will be used both for separability criteria (extraction of trees in the image) and to quantify distances (spatial distribution of elements), surfaces (flower or wood surface), or form factors (trunk diameter, tree volume and habit, branch orientation, etc.). A significant part of the work will be dedicated to the generalization and transfer of the methods developed, in particular to domain adaptation techniques to limit image annotation requirements for the constitution of training and test datasets for deep methods.
- Genetic analysis to identify resilience typologies: once the traits have been estimated at the level of each orchard, multivariate statistical analysis will be used to identify particular genotype behaviours across the diversity of each species, and to identify which of the new variables should be routinely measured by genetic research teams. Quantifying the heritability of the estimated traits will provide information on the ability to use more advanced genetic analyses to select these traits and create innovative varieties. The comparative approach will enable us to compare the adaptive potential of these species.
The work carried out by the candidate will be valorized by at least one publication in a peer-reviewed journal presenting the results of the study and the publication of an open access dataset. The post-doc will also be able to present his/her results at an international conference.
The selected candidate will be based mainly in the CAPTE team in Avignon. Bi-monthly videoconferences between the four teams will be organized to monitor and steer the project. He/she will travel to the apple, apricot and peach core-collections of the GAFL and AGAP units for measurement campaigns, and if necessary for collaborative methodological developments.
Funding for this post-doctorate is provided by the Convergences Institute #DIGITAG. He/she will be member of the DIGITAG #community (access to training courses, research schools, seminars, transfer support from SATT AxLR, and the possibility of applying for calls reserved for Convergences Institute PhD students and post-docs (IT support, international mobility, etc.). #DIGITAG post-docs also benefit from the Digital Showcase service (development of a small demonstrator/web service presenting a contribution/result of their work, with the assistance of an IT service provider).
Training and skills: PhD PhD in Engineering Sciences with skills in image processing and deep learning, who has worked on problems related to Biology. Good Python skills are required, as well as a good English (reading, writing, speaking). Experience or knowledge of agronomy will be appreciated. Experience and an appetite for teamwork and fieldwork are essential.
INRAE’s life quality By joining our teams, you benefit from (depending on the type of contract and its duration): – up to 30 days of annual leave + 15 days “Reduction of Working Time” (for a full time);
– parenting support: CESU childcare, leisure services;
– skills development systems: training, career advise;
– social support: advice and listening, social assistance and loans;
– holiday and leisure services: holiday vouchers, accommodation at preferential rates;
– sports and cultural activities
– collective catering.
- Contract: Postdoctoral position
- Duration: 18 mois
- Beginning: 01/01/2024
- Remuneration: INRAE salary scale, based on experience (~2000€ )
- Reference: OT-18378
- Deadline: 15/09/2023
Centre Provence-Alpes-Côte d’Azur More information on centre UMR EMMAH- 1114 84000 Avignon Contact Weiss Marie +33 (0)4 32 72 23 79 firstname.lastname@example.org Boudon Frédéric email@example.com Living in France and working at INRAE Our guide for international scientists