Postdoc: Modelling bacterial transport strategies in real-life environments (LAI, Marseille-Luminy)

About this position

Understanding bacterial transport strategies in real-life environments: an optimal-control perspective on well-controlled experiments with model bacteria & environment

Supervisor: Jean-François Rupprecht, Laboratoire Adhésion et Inflammation (LAI), Marseille-Luminy (France), and TBMTM consortium ( https://www.cite-des-energies.fr/biam/projets/bacteries-magnetotactique… )

Abstract Magnetotactic bacteria (MTB) navigate along Earth’s magnetic field by combining magnetic alignment with tactic responses to chemical and oxygen cues. In nature, however, these bacteria do not swim in homogeneous fluids: they live in sediments, porous gels and confined microstructures where chemical landscapes are strongly shaped by local heterogeneities. This post-doctoral project aims to understand how MTB exploit such heterogeneous environments to achieve long-distance navigation.

The central hypothesis is that a weak large-scale chemical gradient, too shallow to be directly detected at the scale of an individual bacterium, can be reshaped by the porous structure of sediments into local chemical signals strong enough to guide bacterial motion. The project will investigate two complementary mechanisms:

• Objective 1 — Random sequence of strong chemotactic steps. The sediment converts a smooth macroscopic gradient into a random sequence of large local steps in chemotactic signal. Although the global gradient remains weak, bacteria may detect and follow these intermittent local cues one after another, so that long-range taxis emerges from a succession of stochastic short-range decisions. This is conceptually related to infotaxis; here, however, intermittency arises not from turbulence (MTB operate at low Reynolds number) but from the static heterogeneity of the sediment itself.
• Objective 2 — Periodic sequence of asymmetric chemotactic steps. The sediment converts a smooth gradient into a periodic sequence of asymmetric local steps, each motif consisting of a steep increase toward the source and a weaker decrease in the opposite direction. Such profiles could act as a chemotactic ratchet, rectifying stochastic trajectories and generating net directional transport even when each local trajectory remains noisy.

The project will combine magnetotactic navigation, chemotaxis, porous-media physics, first-passage statistics and optimal control to understand how bacteria reach distant chemical targets.

Keywords: active matter, magnetotactic bacteria, chemotaxis, magnetotaxis, porotaxis, porous media, natural sediments, infotaxis, ratchet transport, first-passage time, stochastic modelling, optimal control.

Context Magnetotactic bacteria orient along magnetic field lines thanks to chains of magnetic nanocrystals. This internal compass biases their swimming direction while the cells sense chemical and oxygen gradients and regulate their flagellar motors accordingly. For decades MTB have mostly been studied in simple, homogeneous media, whereas in nature they live in the first millimetres of aquatic sediments or in stratified water columns, where physical structure and chemical fields are highly heterogeneous.

How can a microscopic swimmer detect and exploit environmental information over distances much larger than its own size? In a heterogeneous sediment, pores, bottlenecks, dead ends and preferential channels may locally amplify, fragment or distort a weak gradient, so that the bacterium follows a sequence of local cues generated by the coupling between chemistry and geometry.

The project is embedded in the AMIDEX-funded TBMTM excellence consortium. Recent experiments on MTB in model porous media have revealed rich nonlinear transport phenomena, including vortical trapping, heavy-tailed passage-time distributions and bacterial-diode regimes where flow can be strongly rectified or blocked depending on the applied magnetic field.

Project ambition The goal is to formulate a quantitative theory of bacterial navigation in heterogeneous sediments, viewed as an optimal-control problem under noisy and spatially fragmented information. Key questions:

• How does a weak macroscopic chemical gradient become decomposed into strong local gradients by the structure of a porous sediment?
• Can MTB exploit such local gradients to perform effective long-distance chemotaxis, despite the absence of turbulence?
• How do magnetic alignment and chemotactic motor control combine to bias transport in disordered microstructures?
• Can asymmetric local chemical landscapes generate ratchet-like rectification of bacterial motion?
• Which control strategy for magnetic sensitivity, chemotactic response and reorientation dynamics optimizes first arrival to a target?
• How do rare events (first arrivals, trapping, escape from dead ends) encode the underlying tactic mechanisms?

Methods

• Stochastic modelling of magneto-chemotactic navigation: Langevin dynamics, run-and-tumble models, continuous-time random walks and first-passage theory, with parameters encoding magnetic alignment, chemotactic sensitivity, reorientation rules, motor control and noise.
• Porous-media-induced signal decomposition: idealized geometries, microfluidic porous networks and numerical reconstructions of chemical fields in disordered media.
• Ratchet-like transport and rectification: whether asymmetric chemical motifs can rectify bacterial motion (“ratchet-taxis”).
• Extreme statistics and first-passage observables: comparison with experiments via first-arrival distributions, trapping times, escape probabilities and heavy-tailed passage-time statistics.
• Optimal control: how an MTB should modulate its magnetic and chemotactic sensitivity to maximize target reaching, minimize trapping or optimize exploration.

Skills to be developed

• Active and soft matter physics: bacterial transport, motility, rectification, nonlinear response to external fields
• Stochastic modelling: Langevin dynamics, run-and-tumble, CTRW, first-passage theory, extreme-value statistics
• Porous-media physics: transport in disordered microstructures, trapping, bottlenecks, preferential pathways
• Microfluidics and microscopy: analysis of experiments in artificial sediments and controlled chemical landscapes
• Image and trajectory analysis: bacterial tracking, segmentation, statistical analysis of trajectories
• Biophysical modelling: linking motor regulation, environmental sensing and emergent transport

Work environment The post-doc will be directly supervised by Jean-François Rupprecht, CNRS researcher at the Laboratoire Adhésion et Inflammation (LAI, Luminy campus, Marseille), in close interaction with the TBMTM consortium, which provides experimental expertise on magnetotactic bacteria, microfluidic porous media and artificial sediments.