Microswimmers learning chemotaxis with genetic algorithms

Benedikt Hartl,Maximilian Hübl,Gerhard Kahl,Andreas Zöttl
DOI: https://doi.org/10.1073/pnas.2019683118
2021-03-05
Abstract:Various microorganisms and some mammalian cells are able to swim in viscous fluids by performing nonreciprocal body deformations, such as rotating attached flagella or by distorting their entire body. In order to perform chemotaxis, i.e. to move towards and to stay at high concentrations of nutrients, they adapt their swimming gaits in a nontrivial manner. We propose a model how microswimmers are able to autonomously adapt their shape in order to swim in one dimension towards high field concentrations using an internal decision making machinery modeled by an artificial neural network. We present two methods to measure chemical gradients, spatial and temporal sensing, as known for swimming mammalian cells and bacteria, respectively. Using the NEAT genetic algorithm surprisingly simple neural networks evolve which control the shape deformations of the microswimmer and allow them to navigate in static and complex time-dependent chemical environments. By introducing noisy signal transmission in the neural network the well-known biased run-and-tumble motion emerges. Our work demonstrates that the evolution of a simple internal decision-making machinery, which we can fully interpret and is coupled to the environment, allows navigation in diverse chemical landscapes. These findings are of relevance for intracellular biochemical sensing mechanisms of single cells, or for the simple nervous system of small multicellular organisms such as C. elegans.
Biological Physics,Disordered Systems and Neural Networks,Soft Condensed Matter,Fluid Dynamics
What problem does this paper attempt to address?