MODELIFE Annual Meeting

MODELIFE Annual Meeting 

Dates: 5-6 novembre 2019
Place: Village Pierre et Vacances, Mandelieu la Napoule

Rue de la Pinea , 06210 Cannes Mandelieu, France



To know more about the Université Côte d'Azur Core Program for Modeling in the Life Sciences, visit our website:

Tuesday, November 5th





UCAncer "Modeling the plasticity of cancer stem cells: from fundamental mechanisms to novel bioactive molecules"

Frédéric Cazals (INRIA)

Isabelle Mus-Veteau (iBV)


Na+/H+ Exchangers perform isotopic fractionation of transported cations

Laurent Counillon (LP2M)


Pulsed perturbations in population dynamics

Ludovic Mailleret (INRA UMR ISA)


Coffee break


Formal methods for models in Biology

Gilles Bernot

Hélène Collavizza

Jean-Paul Comet (I3S)


AP-DV embryo patterning synergy in cell shape change and tissue morphogenesis

Matteo RAUZI (iBV)


From genes to perception: decoding chemical senses by numerical and molecular modeling approaches.

Sébastien Fiorucci (ICN)​


Period-control in a coupled system of two genetic oscillators for synthetic biology

Elena Firippi

Madalena Chaves (INRIA)


Computational strategies for identifying active modules in pancreatic cancer

Leandro Henrique S. Corrêa

Claude Pasquier

Oliver Soriani (I3S, iBV)


Lung volume VS air flow: a match for shear stresses distribution in the bronchial tree.

Jonathan Stéphano

Benjamin Mauroy (LJAD)


Refresh and round table




Round table and music

Wednesday, November 6th



Bacterial biofilm translocation on osmotic gradients

Agnese Seminara (INPHYNI)


Ecophysiological modeling of plankton: from laboratory data to models and from models to in-situ data.... and vice versa.

Fabien Lombard (LOV)


MODELIFE core program: achievements and future directions

Elisabeth Pécou


Coffee break


Vader center : virtual lungs and applications in UCA

Benjamin Mauroy


A novel generalized linear model framework for coevolutionary analysis

Ryan Mc Minds (UCA MSI)




Poster session


Panicein A Hydroquinone inhibits Patched drug efflux activity and increases vemurafenib efficacy against melanoma cells in vitro and in vivo

Laurie Signetti

Nelly Durand

Robert Ballotti

Stéphane Azoulay

Isabelle Mus-Veteau (iBV)


Combinatorial specificity of transcription factor binding & function

Dominic van Essen (IRCAN)


Pangenomic references: improving yeast genomics with graph-based tools

Lorenzo Tattini (UCA)


Multiple scattering-assisted fluorescence amplification: towards biological applications

Gian Luca Lippi

Frédéric Brau

Sylvain Bonnefond


On the drug export mechanism by RND efflux pumps

Méliné Simsir

Isabelle Mus-Veteau

Frédéric Cazals


A discrete cell cycle model: from phases characterization toward observable properties verification

Déborah Boyenval

Gilles Bernot

Jean-Paul Comet

Franck Delaunay



Titles and abstracts

By order of presentation:

  •  UCAncer "Modeling the plasticity of cancer stem cells: from fundamental mechanisms to novel bioactive molecules", Frédéric Cazals (INRIA), Isabelle Mus-Veteau (IPMC)

This talk will overview the activities carried out within UCAncer. In the following, we briefly comment on the consortium, and the three PhD thesis under way.

Consortium. Cancer development is a complex process combining an accumulation of mutations with dynamic changes. Tumors generally involve heterogeneous cell populations (stem-like, progenitors or differentiated tumor cells) with functionally divergent phenotypes (cancer, non cancer) and abilities (mitotic or not, migratory versus static, pro-angiogenic or not). Tumor expansion hinges upon dynamic interactions between non-tumoral and cancer cells, as the latter modify their environment to favor their expansion and trigger changes in cell states on non cancer cells. Of particular interest in this context are Cancer Stem Cells (CSCs), which form a subpopulation of tumor cells with unlimited self-renewal potential, with the ability to give rise to all tumor cell types within a tumor, and the power to resist many conventional cancer therapies affecting the more differentiated tumor bulk cells. Unfortunately, the cellular mechanisms underlying the regulation of CSCs (self-renewal, maintenance, differentiation, crosstalk with the tumor microenvironment) and their resistance to treatments are poorly understood, jeopardizing therapies and confusing prognosis. Therefore, the main challenge towards curative cancer treatments involves the dissection of molecular and cellular mechanisms that underlie CSCs self renewal, differentiation, and resistance to chemo-, targeted and immuno-therapies.

Our consortium formed of 12 teams from Nice and Sophia Antipolis ambitions to reveal fundamental molecular and cellular mechanisms underlying the aforementioned regulation mechanisms of CSCs at the atomic, cellular, tissue/tumor, and body levels. These investigations will help understanding disease propagation and developing innovative curative cancer treatments as well as enhanced prognosis tools. The work foreseen is inherently multidisciplinary, and our consortium encompasses the areas of chemistry, molecular and cellular biology, biophysics, electrophysiology, immunology, machine learning and physics/mathematical modeling of complex systems, with strong expertise at all (atomic, cellular, tissue/tumor, and whole) levels. Three PhD students have been recruited on three interdisciplinary projects. Two are funded by MODELIFE (Méliné Simsir and Leandro Correa), and one is funded by INRIA (Kevin Atsou).

  • Na+/H+ Exchangers perform isotopic fractionation of transported cations

    Mallorie Poët1,5, Yann Bouret2, Gisèle Jarretou1,5, Vincent Balter3, Nathalie Vigier4 and Laurent Counillon1,5

1 Université Côte d’Azur, CNRS LP2M, 28 Avenue de Valombrose, Nice, France

Université Côte d’Azur, CNRS INPHYNI, 28 Avenue Valrose, Nice, France

Ecole Normale Supérieure Lyon

4 Sorbonne Université, CNRS, Laboratoire Océanographique de Villefranche (LOV),

5 Ion Channels Science and Therapeutics Laboratories of Excellence

Li possesses two natural stable isotopes (6Li and 7Li) which exhibits a large range of isotopic fractionation, i.e the enrichment in 6Lio r 7Li in sedimentary rocks, marine organisms, and mammalian organs. As lithium enters cells, organs and tissues under its Li+ ionized form, we reasoned that this observed isotopic fractionation had to be due to the activity of ion channels and/or transporters.

To challenge this hypothesis, we compared the isotopic ratios of intracellular lithium, that had been specifically transported by Na+/H+ exchangers (NHEs), to that of the original lithium extracellular uptake solution. We found that Na+/H+ exchangers perform a preferential enrichment in 6Li+ during their transport cycle, coupled with a slight fractionation by leak channels, resulting in the measured data. We then carried out a series of kinetics and doses-responses of isotopic lithium separation that yielded totally novel data. Combined with the first mathematical model developed for isotopic transport this study sheds a new light on ion transport mechanisms. Finally, we will discuss how this study may provide new clues on organ function and/or geochemical processes.

  • Pulsed perturbations in population dynamicsLudovic Mailleret (UMR INRA ISA)

The dynamics of populations can be perturbed by sudden events that abruptly increase or decrease population densities. The magnitude and temporal pattern of occurrence of pulsed perturbations may interact with different components of population dynamics. In this presentation, I will explore these interactions by focusing on two kinds of perturbations, (i) pulsed removals of fractions of populations and (ii) pulsed introductions of given numbers of individuals, in different population dynamics scenarios such as resource harvesting, population migration, vaccination, and biological control. These examples highlight that the way a given perturbation intensity is spread over time affects both quantitatively and qualitatively the dynamics of pulsed perturbed populations.

  •  Formal methods for models in Biology, Gilles Bernot, Hélène Collavizza, Jean-Paul Comet (I3S)

The identification of parameters is a major problem for modeling complex systems. We propose a methodology for modeling genetic regulatory networks based on formal verification methods. We introduce the discrete formalization framework proposed by René Thomas and show how formal methods could help to identify parameters that ensure some dynamic properteis of the system. We describe a  validation technique of asymptotic properties based on  model-checking. This can be used for example to check the existence of a stable state. In the case where a trace of gene level changes could be observed, we use a well known method in program verification based on Hoare's logic

  • AP-DV embryo patterning synergy in cell shape change and tissue morphogenesis, Matteo RAUZI (iBV)

Morphogenesis is a process by which the embryo is reshaped into the final form of a developed animal. Tissue morphogenesis is under the control of genes which expression follows precise and instructive patterns that can extend from the anterior to the posterior (AP) or from the dorsal to the ventral (DV) axis of the embryo. While much work has been done in understanding how AP and DV patterning independently control morphogenesis, little is known on how cross-patterning functions.  We use the Drosophila embryo as a model system and focus on the process of tissue folding, process that is vital for the animal since folding defects can impair neurulation in vertebrates and gastrulation in all animals which are organized into the three germ layers. Passed work has shown that an acto-myosin meshwork spanning the apical-medial side of prospective mesoderm cells and under the control of the embryo DV patterning plays a key role in mesoderm invagination. Nevertheless, experimental evidence and theoretical simulations have argued that apical constriction per se is not sufficient for invagination. In our lab we have uncovered a cell junctional lateral network under the control of both AP and DV patterning. This contractile network generates tension along the apical-basal axis and within the tissue plane 10 µm inside the mesoderm epithelium initiating lateral cell intercalation. Lateral forces in mesoderm cells seem to play a multivalent role both driving mesoderm extension and invagination. Finally, by implementing 4D multi-view light sheet imaging, infra-red femtosecond ablation to perturb the cytoskeleton and optogenetics to synthetically control tissue morphology, this work shines new light on the origin and functions of a novel mechanism responsible for simultaneous tissue elongation and folding.

  • From genes to perception: decoding chemical senses by numerical and molecular modeling approaches. Sébastien Fiorucci ( Institut de Chimie de Nice )
Olfaction and taste are two strategies developed by our brain to decipher our chemical environment and give sense to odorant and sapid molecules. Our chemical senses are inherently complex neuronal processes. Human possesses ~ 400 genes encoding olfactory receptors (ROs) that are differentially activated by a virtually infinite space of molecules allowing our brain to discriminate more than a trillion of olfactory stimuli. Taste involves ~ 30 gustatory receptors and is also associated to a huge chemical space. But how is the percept encoded in the structure of a molecule? To decipher the combinatorial code of olfaction and taste, we developed numerical models based on molecular modeling [1-2], machine learning [3,4] or structural bioinformatics approaches [5,6]. To describe the underlying molecular mechanisms, we take advantage of the latest data in molecular biology and neurobiology. The presentation will focus on the performance of numerical models in predicting the relationships between a chemical structure and the activation of olfactory/taste receptors.
[1] Conserved residues control the T1R3-specific allosteric signaling pathway of the mammalian sweet taste
receptor. J.B. Chéron, A. Soohoo, Y. Wang, J. Golebiowski, S. Antonczak, P. Jiang, S. Fiorucci. Chem. Senses, 2019,
44, 303-310.
[2] Allosteric modulation mechanism of the mGluR5 transmembrane domain. X. Cong, J.B. Chéron, J. Golebiowski,
S. Antonczak, S. Fiorucci. J. Chem. Info. Model. 2019, 59(6), 2871-2878.
[3] Agonists of G Protein-Coupled Odorant Receptors are Predicted from Chemical Features. C. Bushdid, C.A. de
March, S. Fiorucci, H. Matsunami, J. Golebiowski. J. Phys. Chem. Lett. 2018, 9, 2235-2240
[4] Sweetness prediction of natural compounds. J.B. Chéron, J. Golebiowski, S. Antonczak, S. Fiorucci. Food Chem.
2017, 221, 1421-1425.
[5] The anatomy of mammalian sweet taste receptors. J.B. Chéron, J. Golebiowski, S. Antonczak, S. Fiorucci.
Proteins, 2017, 85, 332-341.
[6] Metal ions activate the human TAS2R7 receptor. Y. Wang, A.L. Soohoo, W. Lei, R.F. Margolskee, J.
Golebiowski, H. Zhao, S. Fiorucci, P. Jiang. Chem. Senses, 2019, 44, 339-347.

  • Period-control in a coupled system of two genetic oscillators for synthetic biology, Elena Firippi, Madalena Chaves

Biological complex mechanisms with oscillatory behavior are often modeled by high dimensional nonlinear ODEs systems, which makes the analysis and understanding the dynamics of the system difficult. In this work, we consider two reduced models that mimic the oscillatory dynamics of the cell cycle and the circadian clock, and study their coupling from a synthetic biology perspective. To improve the performance and robustness of the oscillatory dynamics in a living cellular environment, we consider the problem of augmenting the parameter region admitting periodic solutions. Moreover, we study the capacity for mutual period regulation and control of the coupling between the two reduced oscillators

  • Computational strategies for identifying active modules in pancreatic cancer, Leandro Henrique S. Corrêa, Claude Pasquier (i3S), Oliver Sorian (iBV)

We propose a computational strategy to help identify cancer-activated modules in a large molecular interaction network using gene expression data. A cancer-activated module is a set of genes that interact with each other and that present a high variation of differentiation expression. These regions may indicate a change in metabolic or signaling pathways involved in the same biological processes and which is more likely to be related to the phenotype being analyzed (cancer). Here, we show the computational algorithms we are developing to identify these active modules from real and simulated experiments. We have achieved better accuracy than state of the art in an independent simulation experiment used in other studies, as well as promising results considering real data obtained from cancer patients.i

  •  Lung volume VS air flow: a match for shear stresses distribution in the bronchial tree., Jonathan Stéphano, Benjamin Mauroy (LJAD)

The mucus is extracted by the mucociliary clearance and by air-mucus interaction – cough, exercise, chest physiotherapy ... – When difficulties to extract it occur, chest physiotherapy is engaged. Some techniques used aim for an increase of shear stress,the friction force between airflow and mucus, above a threshold to mobilize the mucus. However, no full scientific understanding of the underlying biophysics involved exists as of today. Our goal is to reach scientific insights on how the shear stress behaves in the bronchial tree for these two different approaches of chest physiotherapy. We developed a mathematical model to compute the quasi-static shear stress in an airway tree
accounting for air inertia and a mean lung’s tissue pressure. A model for compliant bronchi in a symmetric airway was used as lung’s model (Lambert et al., 1982, J. Appl Physiol Respir EnvironExerc Physiol 52(1):44-56). Our model is able to mimic different configuration. The shear stress distribution have a bell-like shape and depends on both tissue’s pressure and flow rate. The flow rate impacts mainly the location of the distribution and, in other hand, the pressure impacts mainly the shear stress intensity. When mucus obstructs a generation, we observe a great increase of the shear stress at the same location and a great decrease in location distal to the occlusion.
Our results allow to analyze responses of a wide range of air flow rates and tissue pressures,mimicking different CP techniques. Our model show us the mechanism origin to the shear stress distribution. Moreover, our results allow us to imagine two chest physiotherapy techniques: "The mucus waiting queue", a safer one but with contraindications, and "The mucus collection", a more adaptable one but with some risk.

  • Bacterial biofilm translocation on osmotic gradients, Agnese Seminara (INPHYNI)

The vast majority of bacteria exist in biofilms: surface-associated communities encased in an extracellular polymeric matrix, an exciting example of prokariotic multicellularity. The discovery of bacterial biofilms has revolutionized the field of microbiology and undermined the paradigm of bacteria as solitary cells. Biofilms grow on virtually every surface with some nutrients and moisture, even in extreme conditions from the bottom of the ocean, to the desert, the glaciers and the geysers. They are a threat to human health because of their increased tolerance to antibiotics, but are also used to develop technological applications for e.g. wastewater treatment and filtration. 

The extracellular polymeric matrix is the one universal feature all biofilms share. But what does the matrix do? The matrix holds the cells together and anchors them to interfaces. Physically, this turns the system from a collection of individual colloids into a complex material. Understanding the emerging properties of this material is essential to predict how biofilms respond collectively to the environment. In this talk I will introduce one such emergent property related to motility: osmotic spreading. I will describe our current experiments testing how biofilms control their own expansion on external gradients of osmotic pressure. 

  • Ecophysiological modeling of plankton: from laboratory data to models and from models to in-situ data.... and vice versa. Fabien Lombard (LOV)
Ecophysiology of organisms, ie. their capacity to adjust their metabolic rates accordingly to environment, directly determines their growth potential, and therefore their abundance in the environment. However determining a complete view of plankton ecophysiology is often complicated by the potential of cultivating organisms. Here we show how laboratory experiments allows to determine general ecophysiological models that could be extrapolated to wider range of organisms and extrapolated at global scale. On the reverse way, global collection of data may helps to uncover unknown ecophysiological properties and in turn assess ecophysiological constraints that couldn't be obtained from laboratory.
  • VADER: Modelling lung’s airflows and mechanics in realistic lung’s geometries, Benjamin Mauroy (LJAD)

Vader center aims at developing models of the respiratory system and most specifically of the mammals’ lung. Our main goal is to support interdisciplinary projects that aims at studying lung’s functions. In this talk, I will first recall the basis of the lung’s physiology and present several models that we are developing, increasing their complexity step by step. We will see how we use or how we plan to use these models in a wide range of applications: wind instruments players, athletes, patients, popularisation, art, etc.

  • Panicein A Hydroquinone inhibits Patched drug efflux activity and increases vemurafenib efficacy against melanoma cells in vitro and in vivo, Laurie Signetti, Nelly Durand, Robert Ballotti, Stéphane Azoulay and Isabelle Mus-Veteau

One of the crucial challenges in the clinical management of cancer is the resistance to chemotherapeutics. We recently demonstrated that the Hedgehog receptor Patched, which is overexpressed in many recurrent and metastatic cancers, pumps chemotherapeutic agents such as doxorubicin out of cancer cells using the proton motive force (Bidet et al. 2012). We discovered two small molecules which inhibits the doxorubicin efflux activity of Patched, enhances the cytotoxic, pro-apoptotic, antiproliferative and anticlonogenic effects of doxorubicin on adrenocortical carcinoma and melanoma cells which endogenously overexpress Patched, and thereby mitigates the resistance of these cancer cells to doxorubicin (Fiorini et al. 2015, Hasanovic et al. 2018). In the frame of the “Programme Structurant” UCAncer, we developed a collaboration with Stéphane Azoulay (ICN Nice) who performed the synthesis of one of the Patched drug efflux inhibitors discovered which was extracted from marine sponge: Panicein A hydroquinone (PAH). We showed that synthesized PAH enhances the efficacy of the targeted therapy vemurafenib against BRAFV600E melanoma cells in vitro and in vivo (in preparation, patent EP14306710.6 delivered in USA). More than 90% of patients with mutations in the serine/threonine kinase BRAF (45–50% of melanomas) have the V600E mutation. However, the clinical responses to vemurafenib are heterogeneous, with median progression-free survival of only 6–7 months. Almost all patients who experience an initial response ultimately acquire resistance that allows disease progression. Our results suggest that the use of the Patched drug efflux inhibitor PAH in combination with vemurafenib could be a promising therapeutic option to improve vemurafenib efficiency against treatment resistant BRAF mutated melanomas.

  • Poster session :

SCsim – A tool for simulation of Single Cell RNA-seq data, Deprez M, Paquet A, Lebrigand K, Nottet N, Waldmann R, Barbry P (IPMC) 

High-throughput single cell transcriptomics (scRNA-seq) has become an important component of the molecular biologist toolkit, leading to the generation of vast amounts of complex high-dimensional data. We propose SCsim, a novel R package for simulation of scRNA-seq data matching real dataset characteristics. SCsim is highly parameterizable and allows simulation of the following biological processes: (i) Transcriptional bursting, modeled using the two-state model i.e genes switch between an ‘on’ and ‘off’ state with a certain probability, and when ‘active’, they produce the number of mRNAs defined as parameter. (ii) Technical and biological mRNA expression variability due to sequencing depth, batch effects or cell type dependent mRNA levels, are simulated based on a mix of Gaussian distributions. (iii) For each cell population, differential gene expression can be simulated for multiple distribution patterns (unimodal and/or bimodal, with different means), which can mimic cells transcriptome progression along a dynamic process such as differentiation. (iv) Dropout events are modeled by a normalized logistic function, based on the hypothesis that the probability of failure to detect a gene is inversely correlated with its expression level. 

A set of summary graphs describing the simulated dataset together with a count table are provided for each simulation. To our knowledge, SCsim provides the most comprehensive simulation of scRNA-seq data. Both data complexity and descriptive QC metrics resemble real scRNA-seq datasets. Datasets generated with SCsim should help gain a better understanding of the composition of scRNA-seq data, and facilitate the development and testing of new scRNA-seq analysis methods.

Swimming gait driven by proprioceptive feedback,J. Sanchez-Rodriguez, C. Raufaste, M. Argentina (INPHYNI).
We have developed an elementary theoretical model of aquatic locomotion, based on [1] and [2]. We link the locomotion velocity to the kinematic of the foil. The amplitude and the beating frequency of the tail are still chosen by the swimmer and
we would like to propose a simple mechanism which selects them. Here,we suppose that the tail motion proportionally depends on the normal force felt by its body. We have constructed a robotic compliant fish which is attached to a force sensor.
We vary the feedback intensity and we measure the resulting thrust, amplitude and frequency. Our theoretical model accurately predicts the experimental outputs.
[1] Theodorsen, T. (1935) NACA Report 496,
[2] Garrick, Isadore E. "Propulsion of a flapping and oscillating airfoil." (1937).

Acyl chain asymmetry and polyunsaturation of brain phospholipids facilitate membrane vesiculation without leakageM.L Tiberti,  R. Gautier (IPMC)

Phospholipid membranes form cellular barriers but need to be flexible enough to divide by fission. Phospholipids generally contain a saturated acyl chain at position sn1 whereas the sn2-FA is saturated, monounsaturated or polyunsaturated. We provide a comprehensive view of the effects of the acyl chain profile of phospholipids on membrane vesiculation by dynamin and endophilin. To obtain molecular-level, we used coarse-grained and all atom molecular dynamics simulations. The development and application of several descriptors allowed to discriminate the structure and dynamics of the different monounsaturated or polyunsaturated phospholipids. Complementarity between biochemical experiments and molecular dynamics simulations reveals that the vertical movements of the chains along the z axis increase with polyunsaturation and are responsible for the increasing flexibility of the membranes which leads to : (i) phospholipids with two polyunsaturated acyl chians make membranes prone to vesiculation but highly permeable; (ii) asymmetric sn1-saturated-sn2-polyunsaturated phospholipids provide a trade-off between efficient membrane vesiculation and low membrane permeability; (iii) When incorporated into phospholipids, docosahexaenoic acid (DHA; omega-3) makes membranes more deformable than arachidonic acid (omega-6).

Acyl chain selectivity of Phosphatidylserine/ Phosphatidylinositol-4-phosphate exchangers, N.F. Lipp (IPMC)
In eukaryotes, the endoplasmic reticulum (ER) is a spread, thin, and highly convoluted membrane-bound network whereas the plasma membrane (PM) is a thicker and asymmetric membrane separating the cytosol from the outside of the cell. To acquire such features, they
notably differ in term of their lipid composition. One lipid of interest is phosphatidylserine (PS) that is an anionic glycerophospholipid synthesized in the ER but mainly enriched in the cytosolic side of the PM. Recently, Osh6p and ORP8 protein in yeast and in human respectively, have been shown to act as
PS/Phosphatidylinositol-4-phosphate (PI4P) exchangers and thereby to contribute to this build-up. PS is delivered to the PM by exchange for PI4P and this exchange is fueled by a PI4P gradient maintained by kinases and the phosphatase Sac1 in the PM and the ER, respectively. It was known that the PM contains more saturated phospholipids than the ER. We wondered whether Osh6p and ORP8 might contribute to maintain this difference by a selective sorting of PS.Using in vitro approaches, we started to examine (i) how fast Osh6p and the ORD of ORP8 transport different PS species and (ii) to what extent this transport is dependent on a PI4P gradient.
Our data suggest that Osh6p and ORP8 differentially transport PS subspecies according to the length of their acyl chains and the number of unsaturation and that thetransport of several PS species is dependent on PI4P. Our preliminary observations provide new insights on how lipid transfer proteins might contribute to the regulation of membrane homeostasis in eukaryotic cells.

Octopuses hunting localize their prey using turbulent odor, water movement and pressure. Nicola Rigolli (U. Genova)

Fluids disperse unpredictably a large number of chemicals: different animals can exploit odor plumes to acquire information about the surrounding environment. Decoding olfactory signals is fundamental to locate food, conspecifics and to avoid predation: how do animals localize the source of the odor? Which kind of algorithms drive their search? Many investigations on rodents and flies have been realized, my research project focuses on octopuses hunting in the ocean: they localize their prey using turbulent odor, water movement and pressure. I model octopuses and their environment using statistical fluid dynamics and decision theory in fact we need to (1) understand what is the signal; (2) develop algorithms able to take signal as input, and provide distance as output.

Deformation of an elastic material paired with a tree structure Michaël Brunengo(1,2) , Benjamin Mauroy(1) , Barret R. Mitchell(2) , Bernard Rousselet(1)  (1 -LJAD, 2 - RespInnovation SAS) 

In order to model effects of automated treatment of respiratory physiotherapy with focused pulses, we study deformation of an elastic material under oscillating constraints on its boundaries linked to a symmetrical and dichotomous tree built as series of cylinders, idealizing bronchial tree. To do so, under infinitesimal strain theory, we constraint airflow created by material volume change to flow in the considered tree and to go through hydrodynamic resistance. This coupling adds friction to material deformation and acts retroactively on it by flow network of the tree. Moreover, the dimensionless system of equations gives a better understanding on the influence of model parameters and could suggest physiological information on airflow inside an idealized bronchial tree and its eigen pulsations. Références [1] X. Dubois de La Sablonière, B. Mauroy, and Y. Privat. Shape minimization of the dissipated energy in dyadic trees. Discrete and Continuous Dynamical Systems - Series B, American Institute of Mathematical Sciences, 2011. 16 (3), pp.767-799 [2] F. Louf, lesson, Résolution approchée de problèmes de dynamique en régime permanent

A mathematical model for tumor-immune cell interactions, Atsou K, Anjuère F, Braud VM, Goudon T (LJAD, IPMC)

We developed a spatio-temporal and size-structured mathematical model to describe the interactions between immune cells and tumors. We model the migration of the tumor antigen-specific cytotoxic effector cells to the tumor microenvironment by a chemotactic phenomenon. Our model exhibits a controlled tumor growth in presence of an immune response and asymptotic states with residual tumors and activated immune cells. We observe tumor dormancy and relapse phenomenons and we also show that the oscillations, often observed in this dynamic, depend more intrinsically on a certain spatial heterogeneity.

Optimal ventilation in a model of the human lung, Frédérique Noel (LJAD)

Consequences of migration pulsedness for allele spread and population genetic divergence Flora Aubree (INRA)

A fundamental aspect of evolutionary theory is our understanding of population genetic divergence, whereby populations connected by gene flow may diverge at neutral loci and for adaptive traits. This forms the basis for inferring population connectivity from genetic data, predicting local adaptation in the face of gene-flow, or the development of incompatibilities between incipient species. While evolutionary theory has typically envisioned gene-flow as a steady, continuous connection among populations, it is increasingly clear that several processes (climatic events, anthropogenic transport, group dispersal) can make gene-flow fluctuating and intermittent, rather than steady and continuous. There is however little theory on how fluctuating migration impacts divergence compared to smooth migration. Some results suggest that variable migration should decrease effective migration rate, but they assumed uncorrelated fluctuations and neutral variation only. Here we analyze mathematically a stochastic model to describe in continuous time the genetic divergence of connected populations, where migration can be pulsed (intermittent). Our model covers a range of migration scenarios (unidirectional, synchronous or asynchronous) and genetic scenarios (neutral markers, locally adapted mutations, incompatible mutations). In a slow-fast limit, we derive simple analytical approximations regarding whether pulsed migration affects population divergence (probability of identity) compared to a steady flow of similar intensity, and validate them with stochastic simulations. We find that migration pulsedness can not only increase, but also decrease, the rate of population divergence. We summarize how predictions depend on the mode of migration and on the type of genetic variation, and provide graphical interpretations in terms of probabilities of fixation. We finally discuss how pulsed migration might leave a footprint in genomic data.

Keywords: pulsed migration, selection, mutation, genetic divergence, FST, Monte-Carlo simulation, Moran model, population genetics

Optimal strategies for fungal spores liberation Daniele Lagomarsino (Instituto Italiano di Tecnologia)
Fungi rely on their spores for survival. From the perspective of an individual fungus, spores represent the chance to find fresh resources and broadcast genetic heritage. Each fungus takes care of its microscopic offspring up to their release in the surrounding environment, but what happens next? Environmental flows are unpredictable and parents are bound to lose control over spore’s fate. However, statistical analysis of atmospheric transport shows that a good choice of timing for spore discharge can lead to optimized rates of survival. I will discuss what kind of information can fungi exploit to optimize survival of their progeny.

Introduction of Priority Regulation in Biological Regulatory Networks Déborah Boyenval (I3S)
The discrete modeling framework of René Thomas allows to abstract biological regulations into an interaction graph and to study its global qualitative dynamics in an exhaustive manner with using formal methods to verify biological properties. Global dynamic is represented by a transition graph where each node represents a state of the system and
each edge represents a transition between two states. In this formalism, all regulations have the same level of priority. However, among biological regulations, hierarchy may exist. This poster introduces an extension of the definition of the discrete modeling framework that takes into consideration priority regulation and proposes an illustration of the impact of this extension on the static and dynamic modeling of biological regulatory network.

Pancreas cancer modeling: a metabolic approach Laetitia Gibart (I3S)
Pancreatic ductal adenocarcinoma, (PDAC), is the most common pancreatic cancer type. PDAC remains asymptomatic during the initial stages hence when diagnosed at a much later stage, surgery is not possible in most cases. This explains why PDAC has a
very low survival rate of 5%. In the tumor, Epithelial PDAC cells are circled by cancer associated fibroblast (CAF). That confers a limited nutrient and oxygen resources. To survive to this poor intake, epithelial PDAC cells have an adapted metabolism. It is also known that some cancer cells can undergo epithelial-mesenchymal transition (EMT) and invade distant organs to form malignant metastases. It is supposed that dialogue between CAF and epithelial cancer cell promote EMT transition by messengers’ exchange. Until today none of the biological experiment manage to find the nature and the direction of those messenger, in fact biologists have only knowledges on cell supernatant. It contains many molecules that could be involved in cellular communications, whether in one sense or another between epithelial cancer cells and CAF. That make a huge number of hypothesis to test. Modelling those communications by coupling might steer in their experiences, by checking first the communications predicted by models. The first purpose of the subject is to adapt for representing the three cell types involved in PDAC by adapting a discreet model of the energetic metabolism regulatory network, and afterwards try coupling in different ways those ones. The main difficulties will be encountered during coupling because of the research space explosion, that’s why it is needed for instance to work upstream on model simplifications.


Marie Larousse, ab* Quang D. Tran,a Philippe Thomen,a Céline Cohen,a François Orange,d Fernando Peruani,c Eric Galianab, Xavier Noblina

a Institut de Physique de Nice (INPHYNI), CNRS UMR 7010, Université Côte d’Azur

b Institut Sophia Agrobiotech (ISA), UMR Inra 1355, CNRS 7254, Université Côte d’Azur

c Laboratoire J.A. Dieudonné (LJAD), CNRS UMR 7351, Université Côte d’Azur

d Centre Commun de Microscopie Appliquée (CCMA), Université Nice Sophia Antipolis

The genus Phytophthora, which belongs to the class Oomycetes, includes some of the most destructive plant eukaryotic pathogens known, responsible for diseases on crops and in natural ecosystems worldwide. In most cases, both dispersal and primary infection are mediated by airborne sporangia or waterborne zoospores. In water, zoospores are released with high-speed swimming up to 200 µm/s. The ion gradient sensing is an important process that dictates zoospore distribution in the environment. In the context of the COMOZOO project, we raise the general hypothesis that the physical and chemical distribution of ion at the soil–root interface is a parameter that may contribute to the constitution of a high-density inoculum or biofilm formation on the plant surface. We have shown that guidance of zoospores by potassium gradient sensing mediates aggregation. To get insight on the swimming mechanism of zoospores and their interactions against ion gradients we first conduct microscopy observations to achieve characteristics of their individual and coordinated swimming. High concentration of potassium makes them reduce speed, change swimming patterns and even stop moving. We also find out that the density impacts the swimming and spreading of zoospores. Beside new aspects in oomycete biology (aggregation) and physics (new microswimmer model), this work could also open the opportunity to develop biocaptors for controlling spreading of the disease by massive trapping of zoospores. This should provide to farmers new methods for early detection and prevention of epidemic risk without waiting for the onset of symptoms.

  • A novel generalized linear model framework for coevolutionary analysis Ryan Mc Minds (UCA MSI)

Neutral evolutionary processes can introduce strong correlations between unrelated traits. In any analysis of the traits of evolving organisms, these neutral correlations should be explicitly considered and controlled for. Methods to do so are common for univariate analyses, but large multivariate datasets are often analyzed in a more abstract manner that makes phylogenetic analyses more complicated. Using the Stan modeling framework, we introduce here a model for the analysis of host-microbe prevalence data, which scales to larger datasets than a previously published generative model and incorporates more flexible patterns of evolution. We apply the model to analysis of coral endosymbionts of the family Symbiodiniaceae, and find that the rate of evolution has increased in two host-specific lineages. One of these lineages is closely related to a clade of symbionts that is found only in coral skeletal samples, suggesting an interesting evolutionary transition either to or from indirect symbiosis with another eukaryotic coral symbiont.

  • Combinatorial specificity of transcription factor binding & function Dominic van Essen (IRCAN)

More than half of all classes of transcription factors are assembled as combinations of distinct subunits, yet how or whether this could affect their binding specificity or function is still unknown. One possibility is that combining smaller units simply represents the most efficient and functionally-robust assembly strategy, and that transcription factors comprising distinct but homologous subunits may perform overlapping or redundant cellular functions. On the other hand, it is also plausible that each combination of subunits may execute a unique and nonredundant role. Using a newly-developed ‘split-epitope’ system to selectively analyse pairwise combinations of tagged proteins, we have investigated the roles of specific dimer combinations of the model transcription factor family NF-kappa B. We have been able to determine the specific genome-wide binding profiles, DNA target motifs and stimulus responsiveness of each dimer independently, and to model their selective associations with distinct classes of genomic regulatory elements. Moreover, using cells lacking particular dimer combinations, we were able to examine their differential effects on target gene regulation, and to begin to computationally dissect the features that drive the genetic requirements for individual dimers. Together, these results begin to shed light on the cominatorial behaviour of the different subunits of NF-kappa B, and we suggest that this is likely to represent a widely-used strategy employed by many other families of transcription factors.

  •  Pangenomic references: improving yeast genomics with graph-based toolsLorenzo Tattini 

Linear reference genomes are the core of resequencing studies. However, mapping reads against a single consensus reference shows intrinsic limits, for example in the context of hybrid genomes. Consensus references tend to bias us away from the observation of variation in the genomes sequenced by means of high-throughput platforms and this issue get worse as the number of sequenced genomes grows. Using a pangenome is a general solution that includes both sequence and variation from many different individuals as a reference. The pangenome can be naturally modelled as an annotated graph allowing for the functionality traditionally provided by linear reference genomes. Our team is currently exploring graph-based tools to improve several fields of yeast genomics which include: phylogenetics, genome evolution, the occurrence of horizontal gene transfers and introgression as well as the study of the mycobiome.

  • Multiple scattering-assisted fluorescence amplification: towards biological applicationsGian Luca Lippi (INPHYNI), Frédéric Brau, Sylvain Bonnefond
  • On the drug export mechanism by RND efflux pumps, Méliné Simsir (IPMC), Isabelle Mus-Veteau (IPMC), Frédéric Cazals (INRIA) 

The Hedgehog receptor Patched is expressed in many cancers, and has been recently shown to have a drug efflux activity which confers resistance to chemotherapy [1, 2, 3]. However, its drug efflux mechanism is poorly understood. Patched is a member of the RND familly which extrude drugs using the proton gradient [4]. The paradigm model of those proteins is AcrB, a protein responsible for antibiotic resistance in gram negative bacteria. Despite the number of structural studies performed on this protein, its drug efflux mechanism is still under debate. AcrB forms an homotrimer, and the available structures are either symmetric ones (all subunits in thesame state), or asymmetric ones (subunits are in three different states, ABE). In its asymmetric state, the protein extrudes a drug molecule against proton uptake. Unfortunately, the size of the system (1049 amino-acid per monomer and membrane) is such that dynamic simulations failed to unveil the detailed mechanism [5]. This study [1] makes a stride towards a finer understanding of the export mechanism of AcrB, exploiting the known crystal structures (35) as well as novel modeling tools. First, we show that all asymmetric trimers occupy the ABE state. Second, we exhibit states for domains of AcrB, and ascribe these to states of whole subunits. Third, we characterize the conformational changes undergone by the domains of a given subunit, which is key to correctly classify monomers and trimers. Fourth, we delineate the evolution of contacts between subunits and the drug during the export mechanism. Finally we confront the obtained results to the other RND’s available structures. Altogether, these insights pave the way to performing dynamic simulations of AcrB, by focusing on those degree of freedom which are key during the export mechanism. In turn, these findings may foster the design of molecules blocking the efflux.


[1] M. Bidet, A. Tomico, P. Martin, H. Guizouarn, P. Mollat, and I. Mus-Veteau. The Hedgehog receptor patched functions in multidrug transport and chemotherapy resistance. Molecular Cancer Research, 2012
[2] L. Fiorini, M.A Tribalat, L. Sauvard, J. Cazareth, E. Lalli, I. Broutin, O. P Thomas and I. Mus-Veteau. Natural paniceins from mediterranean sponge inhibit the multidrug resistance activity of Patched and increase chemotherapy efficiency on melanoma cells. Oncotarget, 2015.
[3] A. Hasanovic, C. Ruggiero, S. Jung, I. Rapa, L. Signetti, M. Ben Hadj, M. Terzolo, F. Beuschlein, M. Volante, C. Hantel, E. Lalli and I. Mus‐Veteau. Targeting the multidrug transporter Patched potentiates chemotherapy efficiency on adrenocortical carcinoma in vitro and in vivo. International journal of cancer, 2018.
[4] A.V Vargiu, V.K Ramaswamy, G. Malloci, I. Malvacio, A. Atzori, and P. Ruggerone. Computer simulations of the activity of RND efflux pumps. Research in Microbiology, 2018.
[5] F. Cazals M. Simsir, I. Mus-Veteau. Studying dynamics without explicit dynamics: a structure-based study of the export mechanism by AcrB. In preparation

  • A discrete cell cycle model: from phases characterization toward observable properties verification Déborah Bozenval, Gilles Bernot, J.-P. Comet, F. Delaunay

The cell cycle is series of events that lead to correct duplication of a cell DNA (S-phase) and its equal distribution into two daughter cells (M-phase). Progression through cell cycle is driven by a regulatory network of cyclin-dependent kinases (CDKs) and phosphatases. Recent studies highlight non-canonical functions of CDKs and phosphatases notably in regulation of carbon and energy metabolism according to cell cycle phases (G1, S, G2 and M phases).
Based on an extended René Thomas' modeling framework, a discrete model of the regulation of cell cycle has been designed. Then, parameterization has been constrained using formal methods such as model checking and ad hoc discrete Hoare logic. Model checking tests if a so-called model (interaction graph associated with a parameterization) satisfies CTL formulas expressing biological behavioural properties. Hoare logic constrains parameter values so that the regulatory network dynamics is compatible with a biological trace.
In this study, the cell cycle has been considered as a biological trace, determined from experimental observations of the sequence of regulatory events across cell cycle phases. This model will be used to elucidate causal relation between the cell cycle coupled with other biological systems on the one hand (e.g. the metabolism or circadian clock) and phase-dependent phenotypes experimentally observed on the other hand. One prospect is the understanding of metabolic reprogramming in healthy and cancer cells.


People wishing to register can submit a document proposing a title and a summary, in the button below