MODELIFE WORKSHOP

Qualitative modelling techniques play an important role in biology and are seen as crucial for developing scalable methods for modelling and synthesizing biological systems. While a range of interesting work has been done in this area there still exists challenging issues that need to be addressed for the practical application of these modelling techniques. The aim of this workshop is to update a broad audience with current examples of research works applying “qualitative modelling in biology” concepts and methods to following questions: how to learn the model from the input data, and how complementary data can help to the customization of qualitative models.

Date : November 12th, 2019

Adress : Room 101 – Forum level 1 (Sophi@Tech Campus, Sophia-Antipolis)

Schedule : 14:00-17:15

Keynote Speakers

Keynote Speakers

- Nathalie Théret

Title : From integrative approach to analysis of pathway dynamics: application to TGF-b signaling in chronic liver disease

Abstract : Most cases of hepatocellular carcinoma (HCC) develop in cirrhosis resulting from chronic liver diseases and the Transforming Growth Factor β (TGF-β) is widely regarded as both the major pro-fibrogenic agent and a critical inducer of tumor progression and invasion. Targeting the deleterious effects of TGF-β without affecting its physiological role is the common goal of therapeutic strategies. However, identification of specific targets remains challenging because of the pleiotropic effects of TGF-β linked to the complex nature of its extracellular activation and signaling networks. Developing a systemic approach and predictive models is a necessary condition for understanding the contextual regulation of TGF-β activity and for identifying effective therapeutic strategies.

To address the complexity of TGF-β networks, we developped a discrete formalism for building large-scale discrete models. The CADBIOM language is a state-transition formalism, based on a simplified version of guarded transition and describes the dynamic behavior of the system by introducing temporal parameters to manage competition and cooperation between parts of the models. The application is now deployed in four modules allowing to build large dynamic models based on knowledge formalized in BioPAX language, and to analyze these networks in order to identify, for example, regulators (genes, proteins) of targets of interest.

After a brief introduction on the biological complexity of the TGF-b signal in chronic liver diseases, we will explain the Cadbiom language and show its interest in the causality analysis of signalling networks

- Morgan Magnin

Title : Learning dynamic models from time series data using logic programming

Abstract : The scientific motivation of our recent researches lies in the fact that a large amount of time series data can now be obtained quite easily, with both (i) the spread of numerical tools in every part of daily life and (ii) the development of New Generation Sequencing methods (NGS) in biology. A critical question here is to attach a meaning to these data, i.e., build relevant models (a task that cannot be designed anymore by hands only) that are both meaningful (for the researcher to have a better understanding of the processes at stake) and predictive enough. It then becomes crucial to be able to connect the time series data with models to improve one's understanding of a targeted system. This means that we need to be able to learn the model from the input data, but also to analyze some key properties on these models. In other words, this implies either to formally prove that some properties are satisfied or to guarantee that these properties are not satisfied. And the designer then obviously needs some automotive help to control the system in such a way that the property may be satisfied. In recent years, we have investigated two complementary learning approaches to infer models from time series data: one is based on the use of Answer Set Programming, the other on inductive logic programming. In this talk, we will show how, thanks to these methods, we have been able to address systems with hundreds of interacting components. We will also expose the current limits of such approaches, drawing some challenges for our future works.

- Laurence Calzone

Title : Personalisation of logical models using omics data

Abstract : Logical models of cancer pathways are typically built by mining the literature for relevant experimental observations or by inquiring pathway databases. They are usually generic as they apply for large cohorts of individuals. As a consequence, they generally do not capture the heterogeneity of patient tumours and their therapeutic responses.

After introducing our approach for constructing logical models and simulating them stochastically, I will present the methodology for personalising logical models to data and show how these models can be used for testing the effect of drugs.

13:45   Welcome

14:00   N. THERET, INSERM-IRISA,Rennes

            From integrative approach to analysis of pathway dynamics:        

            application to TGF-b signaling in chronic liver disease

15:00   M. MAGNIN, LS2N, Nantes

            Learning dynamic models from time series data using logic programming

16:00   Coffee break

16:15   L. CALZONE, Institut Curie, Paris

            Personalisation of logical models using omics data

17:15   Closing

Adresse

SophiaTech Campus

Batiment Forum

930 Route des Colles, 06410 Biot Sophia Antipolis


Inscription

Nous vous rappelons qu’en raison du plan vigipirate attentat et des normes en vigueur dans notre établissement l’inscription à cet événement est obligatoire.Toute personne non identifiée sur un listing d’inscription se verra refuser l’accès à la conférence/colloque/séminaire/événement. Vous devrez être en mesure de justifier votre identité.

Vous venez :