Résumé in brief
|
Samuel Drulhe obtained a M.Sc. in electrical engineering in 2000 as a student of the Université Paris XI - Orsay, the scientific university of south Paris, and of Ecole Normale Supérieure,
a selective school dedicated to high level research and teaching in
Cachan. He also passed an "agrégation" in applied physics in 2002,
which is a highly competitive degree entitling the holder to teach in
the French educational state system.
In the meantime, he was welcomed at the Robotics Institute of the Carnegie Mellon University by Martial Hebert (Pittsburgh, PA, 2001) and the Centre for Imaging Science of the Johns Hopkins University by Donald Geman and Laurent Younès (Baltimore, MD, 2004).
After he had passed a "Diplôme d'Etudes Approfondies"
dealing with mathematics, computer vision and machine learning at the CMLA (CNRS, ENS Cachan), he started in 2004 a PhD in computer science at the Institut National de Recherche en Informatique et en Automatique, in the Helix
research group of bioinformatics, headed by Alain Viari. His work is
supervised by Hidde de Jong and Giancarlo Ferrari-Trecate. He is a
student of the Université Joseph Fourier where he also teaches as a "moniteur" CIES.
| 
Printable CV
|
Top
Systems biology (PhD subject)
Context :
It is now commonly accepted that most interesting properties of an
organism emerge from the interactions between its genes, proteins,
metabolites, and other constituents. This implies that, in order to
understand the functioning of an organism, we need to elucidate the
networks of interactions involved in gene regulation, metabolism,
signal transduction, and other cellular and intercellular processes.
Genetic regulatory networks
control the spatiotemporal expression of genes in an organism, and thus
underlie complex processes like cell differentiation and development.
Genetic regulatory networks consist of genes, proteins, metabolites,
and other small molecules, as well as their mutual interactions. Their
study has taken a qualitative leap through the use of modern genomic
techniques that allow simultaneous measurement of the expression of all
genes of an organism.
In addition to experimental tools,
mathematical methods supported by computer tools are indispensable for
the analysis of genetic regulatory networks. As most networks of
interest involve many genes connected through interlocking positive and
negative feedback loops, an intuitive understanding of their dynamics
is difficult to obtain and may lead to erroneous conclusions. Modeling
and simulation tools allow the behaviour of large and complex systems
to be predicted in a systematic way.
A variety of methods
for the modeling and simulation of genetic regulatory networks have
been proposed, such as approaches based on differential equation models
and stochastic models. These models provide detailed descriptions of
genetic regulatory networks, down to the molecular level. In addition,
they can be used to make precise, numerical predictions of the
behaviour of regulatory systems. Many excellent examples of the
application of these methods to prokaryote and eukaryote networks can
be found in the literature.
In many situations of biological
interest, however, the application of differential equations and
stochastic models is seriously hampered. In the first place, the
biochemical reaction mechanisms underlying regulatory interactions are
usually not or incompletely known. In the second place, quantitative
information on kinetic parameters and molecular concentrations is only
seldom available, even in the case of well-studied model systems.
Our research follows the development of methods for the modeling and
simulation of genetic regulatory networks that are capable of dealing
with the current lack of detailed, quantitative data. In particular,
our team have developed a method for the qualitative simulation of
genetic regulatory networks that has been implemented in the computer
tool Genetic Network Analyzer (GNA).
The method and the tool have been applied to the analysis of prokaryote
regulatory networks in collaboration with experimental biologists at
the Université Joseph Fourier (Grenoble) and the Ecole Normale
Supérieure (Paris).
The choice of well-studied
piecewise-linear models are proved to be relevant considering the
trade-off between complexity of the model and preservation of the
dynamical properties (attractors): graphs of reachable states are
builded that correspond to a certain interval of tolerance in the
choice of the parameters. The qualitative behavior of the network is
then simulated.
PhD subject :
Hidde de Jong (INRIA Rhônes-Alpes - HELIX) and Giancarlo Ferrari-Trecate (INRIA Rocquencourt - SOSSO and Universita` degli Studi di Pavia) are supervising this research.
An other major aspect of the exploration of genetic regulatory networks
is the inference of the parameters from a chosen model based on the
collection of biological observations (measures of expression levels
and protein concentrations). Precise time dependant acquisitions are presently available.
In piecewise linear systems, different domains of a
closed convex space are defined to partition it, in which the dynamic
of the network can be described through well-known linear differential
equations. The "frontiers" as well as the dynamical parameters in each
domain are to be estimated from the data.
Our research aims at adapting powerful system
identification tools to piecewise affine models in order to infer model
parameters from measures of protein concentrations. Piecewise affine
systems are a special case of hybrid systems; the major difficulty
comes from the context where few information is available (a small
amount of trajectories are measured), and this information is very
degraded (noisy). More than an inference problem, the hyperplane
frontiers are to be estimated simultaneously with the local parameters.
Furthermore, piecewise affine models of genetic regulatory networks
confine the frontier geometry and constrain the parameters related to
"neighbor regions" to be strongly dependent.
Learning management systems (information technologies for education)
The potential applications number of Information and Communication Technologies
(ICT) in education is presently booming: many innovative tools have been developed
for many years and are now in a process of being fully integrated in national
educational systems. If it first allowed elaborating new ways of communicating
and data exchanging, we are now exploring Learning Management Systems (LMS)
that aim at helping students in their learning practice. Such an approach might
modify the present education strategies. It then becomes critical to examine
the integration of these technologies in the current teaching practices.
We set up a teaching platform based on Moodle software,
and we evaluated its use in the context of our courses.
We tried different pedagogical experiments and we are now evaluating (through
a quantitative inquiry) the effect of using such sysems in a mixed context
(as an add-on to the usual academic manners of teaching in presence of the
students).
Top
Courses
| Year 2006/2007: |
|
|
Year 2005/2006: |
Level: Bac+3 , filière TIS Polytech'Grenoble, 1ère année
Title: Information retrieval
Head teacher: Catherine Berrut
Description: use of search engines, IR evaluation |
Level: Bac+4, filière RICM Polytech'Grenoble, 2ème année
Title: Data bases and
information systems
Head teacher: Marie-Christine
Fauvet
Description: relational algebra, SQL, functional dependancies, UML |
|
Level: Bac+4 info (ecole info IUP2 Miage)
Title: Data bases and
information systems
Head teacher: Marie-Christine
Fauvet
Description: relational algebra, SQL, functional dependancies, UML
|
|
| Year 2004/2005: |
|
Top
Selected publications and talks in - Systems Biology, Hybrid Systems -
Paper
S. Drulhe, G. Ferrari-Trecate, H. de Jong, The switching threshold reconstruction problem for piecewise affine models of genetic regulatory networks, IEEE Transactions on Circuits and Systems I, special issue on systems biology (in press)
Conference
S. Drulhe, G. Ferrari-Trecate, H. de Jong, A. Viari (2006), Reconstruction of switching thresholds in piecewise-affine models of genetic regulatory networks
, J.P. Hespanha, A. Tiwari (eds), Hybrid Systems: Computation and
Control, HSCC 2006, Lecture Notes in Computer Science, Springer-Verlag,
Berlin. 
Conference posters
S. Drulhe, R. Porreca, G. Ferrari-Trecate, and H. de Jong. An identification procedure for piecewise-affine models of genetic regulatory networks. The Eight International Conference on Systems Biology (ICSB2007), Long Beach, CA, USA, 2007.
S. Drulhe, R. Porreca, G. Ferrari-Trecate, and H. de Jong. An identification procedure for piecewise-affine models of genetic regulatory networks. Toward systems biology, Grenoble, France, 2007.
Technical Report
S. Drulhe, G. Ferrari-Trecate, H. de Jong, A. Viari (2006), Reconstruction of switching thresholds in piecewise-affine models of genetic regulatory networks, Technical report INRIA RT-0322, INRIA, March 2006.
Talks
Identification Procedure
for PWA Models of
Genetic Regulatory Networks,
Journée satellite modélisation dynamique et simulation des réseaux biologiques, JOBIM 2007
(*), Marseille, Juillet 2007.
Reconstruction des seuils de transition dans le cadre de modèles affines par morceaux de réseaux de régulation génique, projects "GDyn - Gene networks DYNamics"(*) and "BacAttract - Theoretical and Experimental Analysis of Attractors of Genetic Regulatory Networks"(*), Ecole Normale Supérieure, Paris, December 2005.
Selected publications and talks in - Learning Management Systems -
Conference
S. Drulhe, Moodle permet-il de faire un enseignement vraiment différent
?, in Conférence Francophone Moodle Moot (*),
Castres, 2007.
Research Report
M. Coquerelle, S. Drulhe, S. Laborie, Y. Lambert, Étude et
expérimentation
d'un nouveau support d'échange et de suivi pédagogique,
in Annales des ateliers du CIES de l'Académie de Grenoble, 2007.
Talks
S. Abry, S. Drulhe, S. Laborie, Moodle en cinq minutes, Séminaire
LMS (dédié aux cellules TICE des universités et IUT de Grenoble), mars 2007.
S. Abry, M. Coquerelle, S. Drulhe, S. Laborie, Y. Lambert, Étude
et expérimentation d'un nouveau support d'échange et de suivi
pédagogique, Journée de présentation des ateliers du CIES
de l'Académie
de Grenoble, mars 2007.
S. Drulhe, S. Laborie, Université virtuelle : des innovations pour l'enseignement,
Journée d'accueil des moniteurs du CIES de l'Académie de Grenoble, octobre
2006.
Top
Organizational and other activities
Samuel Drulhe has been elected in March 2005 to represent people without INRIA open-ended contract at the advisory board called Comité d'Unité de Recherche.
Standing by FOURMI with others PhD students of the
Doctorate School in Computer Science, Applied and Pure Mathematics of Joseph Fourier University, Samuel has been elected in June 2006 to take part of the school body.
By the way, I recommend the "Guide
du Doctorant MSTII" (in French) written by Alexandre Donze which is an excellent
travel book concerning the administrative labyrinth that any PhD student has to face : it is focused on the Graduate
school of Mathematics and Computer Science of Joseph
Fourier University, and being precise, it is precious.
Top
|
