Modélisation grande échelle de réseaux biologiques :
vérification par contraintes booléennes de la cohérence des données

Philippe Veber 1
1 SYMBIOSE - Biological systems and models, bioinformatics and sequences
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : High-throughput techniques in molecular biology are able to measure a huge number of variables simultaneously. They are currently used as a routine investigation method and therefore produce large amounts of information. The analysis of high throughtput data is particularly difficult, due to their usual high level of noise, and the lack of independant samples for the same experiment. This thesis deals with the integration and the analysis of high throughput data : each type of data can be thought of as measuring a certain aspect of a biological process, so how to combine these heterogeneous data in order to deduce relevant conclusions ? To this end, we introduce a compatibility criterion between a graphical model of cellular regulations and equilibrium shift data. We then show how this criterion can be used to derive predictions or to propose corrections in case of an incompatibility. This tasks require the resolution of constraints on finite domain variables, and we developped two approaches for these problems. The first one is based on the notion of decision diagrams, which is classical data structure used for circuit verification~; the second approach we propose relies on recent techniques from logical programming. Finally, we applied our approach to real data on E. coli and yeast. The networks we considered have up to several thousands of genes and regulations. We show on these data that our consistency criterion can effectively be used to derive robust predictions, as well as relevant corrections to the model under study.
Complete list of metadatas

Cited literature [157 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00185895
Contributor : Philippe Veber <>
Submitted on : Friday, February 29, 2008 - 2:48:06 PM
Last modification on : Friday, November 16, 2018 - 1:23:05 AM
Long-term archiving on : Friday, November 25, 2016 - 11:03:01 PM

Identifiers

  • HAL Id : tel-00185895, version 3

Citation

Philippe Veber. Modélisation grande échelle de réseaux biologiques :
vérification par contraintes booléennes de la cohérence des données. Génie logiciel [cs.SE]. Université Rennes 1, 2007. Français. ⟨tel-00185895v3⟩

Share

Metrics

Record views

326

Files downloads

489