Physics of Biological Systems - Institut Pasteur - CNRS URA 2171


Our group works on quantitative modeling of biological systems by a combination of analytical and computational tools, coupled with small-scale experiments that we realize ourselves or in close collaboration with experimental groups.

Activity for the past four years has focused on computational biology of cell regulatory processes and motility of macroscopic organisms.
For the former, we have predicted and validated about 50 small non-coding RNAs in the pathogen Listeria monocytogenes and found some of their mRNA targets. A pair of ncRNAs were demonstrated, in collaboration with P. Cossart’s group, to regulate the virulence of the bacterium by mechanisms that remain yet to be discovered.
As for the motility of macroscopic organisms, such as insects and birds, we have introduced a novel search strategy, infotaxis, which permits to locate source of odors, pheromones and other diffusible substances even from very long distances, when the rate of detection is low. This lack of cues makes it useless to employ gradient-climbing strategies, as in bacterial chemotaxis. Decision-making in infotaxis is based on the basic principle of moving so as to maximize the rate of acquisition of information on the location of the source. Other activities of the group have included inference methods for biological problems. This has led to a novel method for the inference of maps of forces within membrane microdomains from the time trajectories of nanoparticles and other biomolecules that can tracked with sufficient resolution.

Current and future activities of the group mostly feature projects on biological strategies of motility, namely infotaxis, bacterial and eukaryotic chemotaxis.
As for infotaxis, we shall pursue its applications to collective searches performed by a swarm of agents (see below for a Hollywood version from the numb3rs tv series) collaborating in the build-up of the infotactic space-time map.
Concerning the chemotactic response of E. coli to pulses of chemoattractants, we model the functional reasons for the observed strategic behavior of the bacterium and we test predictions by a new experimental method based on inferences of the bacterial response from the images of swimming bacteria. The major advantage is that the method is not invasive and assays the behavior of the whole flagellar bundle (rather than a single flagellum as in standard tethering assays).
As for eukaryotic chemotaxis, we are working on experimentally tracking D3-phosphoinositide PIP3 patches in Dictyostelium discoideum in order to quantify their space-time dynamics on the cell membrane. The motivation stems from models for directional sensing that we have developed, which predict very similar responses to static gradients yet strongly differ in their kinetics of polarization.
Finally, computational activities of the group will consider evolutionary population genetics data, namely those that pertain to the contribution of recombination events to genomic diversity. Statistical methods properly reconstructing the population sample evolutionary history will be developed to refine current recombination maps.
Capture d’écran 2010-01-22 à 23.37.06
Numb3rs: Click to see the video