Anne Poupon (INRA BIOS Tours), on methodology and biological knowledge formalization validation).
Katsumi Inoue (National Institute of Informatics, Tokyo) for use and extension of SOLAR, a system developed for consequence-finding.
Abstract
The aim of the work is to semi automatically construct signaling pathways (represented as networks) from a Knowledge Base. Two directions are investigated. First, in a deductive process, pieces of the network are inferred, and various properties of the network are studied. Then in a diagnosis perspective, new experiments will be automatically proposed to recover elements of the network considered as missing by scientists. Moreover quality of the components of the network will be studied.
Signaling pathways based on two receptors, FSH and EGF, will serve as the basis of this work.
The work will be done in close collaboration with scientists from INRA Tours (BIOS group) and of K. Inoue’s group at NII (Tokyo).
Context
The work will be done in the context of the ASAM project (INRIA/AMIB - INRA/BIOS) which aims to help the understanding of signaling pathways involving G protein-coupled receptors (excellent targets in pharmacogenomics). In this work we are only interested in the influence graph (static model) and do not take into account the dynamics of the network. Large amounts of experiments are available in this context while globally interpreting all the results remains a very challenging task for biologists. As a solution, ASAM proposes to design a Knowledge Base (KB) containing expert rules to interpret the various experimental results and semi automatically construct signaling pathways (represented as networks).
The initial knowledge base is built by the scientists from experimental data and information contained in scientific literature. It is expressed in a logical setting (first-order logic, mainly logic programming).
Objectives
The goal to achieve during the PHD is many folds:
1- Construct and analyzing signaling pathways by reasoning on the expert knowledge provided by the scientists.
2- Guide scientists in the choice of new experiments to be conducted by diagnosing missing interactions in the network.
3- Study the quality of the data obtained based on its provenance.
4 - Build signaling pathways based on two receptors (for which knowledge has been collected and formalized), FSH and EGF, as a basis for this work.
Work program
Step1: Systems for automated deduction for consequence finding or answer set programming, as well as constraint-based settings will be investigated. The formalization will enable us to (i) establish the consistency or the inconsistency of the model (ii) suggest updates of the knowledge base (model revision) in order to ensure consistency (iii) query properties of the network.
Step2: The completeness of the network will be evaluated in collaboration with the scientists. Discovering missing experiments (initial facts) should be performed automatically.
Step3: The completeness of the network will be evaluated in relation with the scientists. Discovering missing experiments (initial facts) should be performed automatically by building hypotheses (abductive reasoning).
Signaling pathways based on two receptors (for which knowledge has been collected and formalized), FSH and EGF, will serve as the basis of this work.
No prerequisite in biology is required.
Interest in genomics together with motivation for Artificial Intelligence approaches to Bioinformatics are expected.
Basic skills in logic (logic programming) are needed. Depending on the applicant’s profile emphasis can be put on Machine Learning methods to acquire signaling networks.
Détails
Expected funding
Institutional funding
Status of funding
Expected
Candidates
Utilisateur
Créé
Dimanche 10 avril 2011 19:45:11 CEST
dernière modif.
Lundi 02 mai 2011 21:41:56 CEST
Fichiers joints
filename
créé
hits
filesize
Aucun fichier joint à cette fiche
Connexion
Ecole Doctorale Informatique Paris-Sud
Directrice
Nicole Bidoit Assistante
Stéphanie Druetta Conseiller aux thèses
Dominique Gouyou-Beauchamps
ED 427 - Université Paris-Sud
UFR Sciences Orsay
Bat 650 - aile nord - 417
Tel : 01 69 15 63 19
Fax : 01 69 15 63 87
courriel: ed-info à lri.fr