Offres
Post-doctoral position - Mathematical modeling of Bacteria colonization
Contract type : Fixed-term contract
Duration : 1 year and a half
Renewable contract : No
Level of qualifications required : PhD or equivalent
Fonction : Post-Doctoral Research Position
Location : LJLL, Sorbonne Université
Context and Location
This position is in the frame of the project 'Towards a realistic model for biofilm formation' funded by the Institut des Mathématiques pour la Planète Terre. The project involves an interdisciplinary collaboration between Sophie Hecht, Diane Peurichard (LJLL, Sorbonne Université) and Nicolas Desprat (LPENS, Ecole Normale Supérieure), and aims to develop realistic mathematical models to understand the main mechanisms of biofilm formation. The recruited person will be hosted at LJLL, Sorbonne Université, with strong interactions and exchanges with the mentioned collaborators. The expected starting date can be as soon as possible.
Short description
Bacteria colonize surfaces to settle sessile colonies called biofilms. Biofilm formation is crucial in industrial processes, in animal health and in biotechnological applications. Biofilm formation has been extensively studied and modeled in-silico using rigid spherocylinders [1]. However, experimental data have shown that bacteria can significantly bend in order to maximize cell-cell interactions in the colony. In this project, we propose to model the local deformations of bacterial skeletons and determine how these effects impact biofilm density. We will perform mathematical modeling, and run in-vitro experiments to calibrate and validate the model. On the mathematical viewpoint, we will develop an individual-based model composed of a minimal set of heuristic (mechanical) rules to identify the key factors required to recover realistic structured patterns observed in bacterial populations. The project will be highly interdisciplinary, as the mathematical model will be systematically confronted to experimental data generated in N. Desprat's lab in order to validate the model assumptions, and in turn the model might lead to testable experimental predictions.
The postdoc will participate in the development of the model as well as its calibration and validation on experimental data. Extension of the project to either 3D modeling on curved surface or micro-macro limit toward a continuous model will be considered.
Main Activities
Specifically, the following aspects will be considered:
Building and analysis of a complete agent-based model for deformable bacteria
Development and implementation of systematic sensitivity analysis tools for the numerical models, relying on the use of SHAP values [2].
Calibration and validation of the ABM on experimental data.
Writing reports/papers
Throughout the post-doctoral period, frequent visits between the various members of the projects will be encouraged. Dedicated fundings will be allocated to these visits and to participation to workshops/conferences to disseminate the results of the project.
Skills
The candidate should hold a PhD in mathematics, physics or other relevant fields. We are looking for a highly motivated candidate with a strong background in mathematical modelling, computational physics / numerics, in particular if in the fields of classical mechanics, bio-physics, soft matter or granular matter, and a deep interest in biological applications. Proven experience in a programming language (Fortran90, C++, matlab and / or python) is required. The ideal candidates are able to work effectively as part of a team, but also to develop and pursue independent ideas. Experience in modelling and applications in biology is highly recommended and collaboration with other people, best cross-disciplinary will be highly appreciated. Previous experience with academic environment and software is an advantage.
Technical skills and level required :
Knowledge in mechanics, stochastic processes and / or in statistical physics
Knowledge in mathematical model building and implementation on the computer
Basic knowledge in numerics, ready to read and understand the necessary algorithms
Programming language (Fortran90, C++, matlab and / or python) advanced level
Relational skills :
Good interpersonal skills
Good communications skills
Reasonable presentations skills
Instruction to apply
To apply, candidates should send the following documents by email to sophie.hecht@sorbonne-universite.fr and diane.a.peurichard@inria.fr
Documents:
CV
Cover letter
Recommendation letters (or references)
(optional) Reports of the PhD
List of publication
References
[1] M. Doumic, S. Hecht, D. Peurichard, A purely mechanical model with asymmetric features for early morphogenesis of rod-shaped bacteria micro-colony, Math. Biosci. and Eng., (2020), 17(6): 6873-6908
[2] S.M. Lundberg, S-I Lee, "A Unified Approach to Interpreting Model Predictions". Advances in Neural Information Processing Systems. (2017) 30: 4765–4774