- BBClist - maillist.psb.ugent.be

Speed dating at the Bioinformatics Event in Maastricht
by Andra Waagmeester 04 Dec '08

04 Dec '08

On behalf of Miranda stobbe of the RSG the netherlands: Dear PhD student, Have you already registered for the special bioinformatics event on Sunday 14th of December? Registration is still possible. The RSG Netherlands is organizing a social event specially for PhD Students who have registered for this program. Please see the flyer attached for more details. See you there, Jayne Hehir Chair of RSG Netherlands, the network of PhD-students in bioinformatics

1 0

Position for a Scientific Programmer at NMC
by marijana radonjic 01 Dec '08

01 Dec '08

Vacancy ** *Scientific Programmer* * **Description * The scientific programmer will be part of a team of five programmers located at various institutes in the Netherlands, constructing a metabolomics data warehouse infrastructure and related applications, within the Netherlands Metabolomics Centre (NMC, http://www.metabolomicscentre.nl/) Within this team, the specific tasks of the scientific programmer are to adapt and expand bioinformatics tools for biological interpretation of nutritional and biomedical metabolomics data. These tools allow to link the metabolomics data, directly from the data warehouse, to intracellular and intra-organ activities, pathways and processes. This will provide a generic bioinformatics evaluation and visualization tool to be used in the area of life style associated diseases such as obesity, atherosclerosis and diabetes type II. It will help in the early diagnosis of disease and developing disease prevention strategies. All data warehouse and applications will be made exportable to controlled vocabularies and ontologies, to make easy exchange of data possible. The programmer will assist in providing web interfaces for the metabolomics database and web-enabling of applications. The data warehouse should be constructed in such a way that overview over the study design, the type of analysis used and the toolbox used for biological interpretation can be queried at once. *Background* The Netherlands Metabolomics Centre is a multicenter research program in which biologists, analytical chemists, database builders and statisticians work together. Therefore excellent networking capacities will be essential for this project. Metabolomics technologies measure a multitude of small chemical molecules in a sample, to get an accurate picture of the physiology of a cell, organ or system (such as human). The research projects within the NMC are mainly focused on human health care, nutrition and plant sciences. The scientific programmer will be appointed at the University of Leiden but work in the Physiological Genomics team of the Biosciences Busniness Unit of TNO, located in Zeist. Physiological Genomics is responsible for the biological and physiological interpretation of "omics" data in (human) health research. At the moment our team exists of 12 (young) researchers. *Requirements* - Experience with HTML, PHP, MySQL, Java and Javascript. - Master in (bio)informatics, ICT / automation or related background with a strong affinity for life sciences, - Knowledge of ontologies and pathway visualization will be of advantage. - Team worker, self-reliant, flexible, and innovative. * *Additional Information: Dr. B. van Ommen, tel : 030 6944438, e-mail: ben.vanommen(a)tno.nl

1 0

Position for a Bioinformatician at GSK
by Jacques van Helden 25 Nov '08

25 Nov '08

Begin forwarded message: From: jean-louis.ruelle(a)gskbio.com Date: 25 novembre 2008 16:16:27 GMT+01:00 Subject: Recrutement bioinfo Job Purpose ========== Within the Bioinformatics group in R&D, the bioinformaticians provide support and creative output in the DNA and protein sequence/structure aspects of vaccine design in all research projects. Bioinformatics has also taken the responsibility of statistical analysis of huge data sets coming from microarray experiments performed in various vaccine projects, in particular a novel class of oncology compounds, the therapeutic cancer vaccines. Activities in the latter field have substantially increased recently, and we are looking for an additional skilled person. You will join the vaccine headquarters located in Rixensart, Belgium, where 5 bioinformaticians are devoted to the bioinformatics support to the discovery and development of vaccines. Responsibilities ============ The job involves performing statistical data analysis of microarray experiments, in collaboration with molecular biology. This includes: - Development of prognostic and predictive classifiers in cancer using gene expression microarray data - Use of external datasets to validate and evaluate classifier performance - Collaborate with molecular biology group to facilitate the migration of diagnostic, prognostic and predictive marker-basedprojects from discovery to development and manufacturing, as regulatory approved commercial products - Interact with molecular biology group to establish and apply quality control pipeline to gene expression microarray data - Technology watch and development of new bioinformatic tools for cancer tumor classification and data processing - Share expertise and know-how - Support molecular biology and CIP with sequence analysis and annotation of biomarkers using genome databases - Maintaining and promoting contacts with external colleaguesand academic research. - Reporting on progress and results to the research project teams. Profile ===== - You have a Ph-D in Life Sciences or Computer Sciences or equivalent. - Proficiency in bioinformatics and statistics with good knowledge of sample classification methods using high dimensional data (background in machine learning is desirable) - Familiarity with pre-processing of gene expression microarray data (QC, normalization, etc ) - Knowledge in R programming is highly desirable - Knowledge of sequence analysis and annotation retrieval from genome databases will be advantageous - You are willing to learn everyday, you are open, creative, you can work autonomously. - You have excellent communication and organizational skills. - You have an good knowledge of English (oral and written). Contact and information: =================== Jean-Louis Ruelle <jean-louis.ruelle(a)gskbio.com>

1 0

Vacancy project member at Netherlands Institute for Systems Biology and Centrum Wiskunde & Informatica
by Roeland Merks 24 Nov '08

24 Nov '08

The Netherlands Institute for Systems Biology (NISB) and Centrum voor Wiskunde & Informatica (CWI) invite applications for the position of a Project member (four years) Job description The project member will join a team of motivated postdocs and PhD students who study biological systems using quantitative and predictive models and multiscale computer simulations. Topics include blood vessel growth, plant development, genetic regulatory networks and metabolic networks in a multicellular context. He/she will develop innovative research codes in collaboration with the postdocs and PhD students, and turn new and existing research codes into more widely applicable, well-documented, and user-friendly modeling tools Hours on a weekly basis: 38 Location The work will be embedded in the Biomodeling and Biosystems Analysis group of the Netherlands Institute for Systems Biology (NISB; www.sysbio.nl ) and within the Life Sciences group of the Center for Mathematics and Computer Science (CWI) in Amsterdam. As the “core modeling group” of the Netherlands Consortium for Systems Biology, the Biomodeling and Biosystems Analysis group develops quantitative and predictive models and multiscale computer simulations in collaboration with systems biology groups at participating institutes. Competences Candidates ideally fulfill the following criteria: - A strong scientific background, e.g. an M.Sc. or Ph.D in computer science, mathematics, theoretical biology or physics, or equivalent experience - Able to communicate with scientists in biology and mathematics - Excellent programming skills in C++ or equivalent; experience with object-oriented design - Experience with scientific simulation and GUI-design is a pre Offer The vacancy concerns a temporary position for four years. Salary and terms of employment are in accordance with the "CAO- onderzoekinstellingen" and based on qualifications. Besides the salary and excellent working conditions, CWI offers attractive and flexible terms of employment, like a collective health insurance, pension-fund, and initial help with housing for foreigners. The position is for 38 hrs/week. Information Direct requests for information or applications before 15 December 2008 to: Dr. Roeland Merks, Roeland.Merks(a)sysbio.nl phone +31 20 592 4117 Center for Mathematics and Computer Science (CWI) and Netherlands Institute for Systems Biology (NISB) Kruislaan 413, 1098SJ Amsterdam, The Netherlands Also see: http://www.sysbio.nl and http://www.cwi.nl.

1 0

Announcement Bioinformatics Event in Maastricht (14-12-2008)
by Andra Waagmeester 18 Nov '08

18 Nov '08

*The relevance of human genomics and bioinformatics for health sciences and the economical development of the BeNeLux and Euregio. * Genomics and bioinformatics event in the Bonnefanten museum in Maastricht on December 14. As a prelude to the annual Benelux Bioinformatics Conference which will be held in Maastricht on December 15 and 16, the Department of Bioinformatics from Maastricht University will organize an open genomics and bioinformatics event in the Bonnefanten Museum on December 14. During this evening the role of genomics and bioinformatics in health research and in the economic developments within the Benelux and the Euregio will be addressed. Bioinformatics is an emerging research field which forms the link between computer science, biomedical sciences and modern genomics research technology. We expect that this evening will not only lead to vivid discussions but will also serve as a meeting moment between conference participants, health scientist from the region and people involved in local economic developments. *Program December 14th* 17:00 Coffee/tea with cake at the aula of the Bonnefanten museum. 17:30 Opening by Dr. Chris Evelo, head of the department of Bioinformatics, Maastricht University, the Netherlands. 17:35 Welcome and introduction on behalf of Maastricht University, speaker to be announced. 17:45 Welcome by Marc Croijmans, delegate of the Maastricht municipal discussing the position of Life-Sciences in Maastricht and the Euregio. 18:00 Presentation given by Dr. Colja Laane, head of the Netherlands Genomics Initiative (NGI) and science director at DSM. 19:00 Discussion related to the theme of the presentation 19:45-22:00 Drinks and tapas at the conference location For more information about the program and venue location see http://www.nbic.nl/bbc2008/SundayEveningEvent/. To register please send an email to bbc2008(a)bigcat.unimaas.nl. I hope to welcome you at the Bonnefanten Museum in Maastricht on December 14th. Dr. Chris Evelo (Head of the department of Bioinformatics) Bioinformatics-BiGCaT The event is sponsored by:. Maastricht University the Maastricht municipality P.O. Box 616 6200 MD Maastricht Main Sponsors for the Benelux Bioinformatics the Netherlands. Conference as a whole are: T: +31(0)43-3881231 Universiteits fonds Limburg (SWOL), Netherlands E: chris.evelo(a)bigcat.unimaas.nl Bioinformatics Center (NBIC), KNAW, NWOW: www.bigcat.unimaas.nl

1 0

Postdoc position in statistical genetics at Vrije Universiteit Brussel
by Joke Reumers 17 Nov '08

17 Nov '08

Vacancy ======== One postdoctoral position in statistical genetics/genetic epidemiology is immediately available. The successful candidate will analyze high-throughput SNPs data using existing tools, and implement additional statistical/computational methods as needed. The work will be performed with a team of bioinformaticians and computer scientists from the in Silico Wet Lab project (http://insilico.vub.ac.be/) The objective is to integrate different types of high-throughput data (SNPs annotations, SNPs genotyping, gene expression and others) for a better understanding of cancers. In Silico is a consortium involving scientists and bioinformaticians from four research groups within the Brussels region. IRIDIA (ULB, iridia.ulb.ac.be) has a longstanding reputation in developing machine learning techniques and their applications, and is complemented by the machine learning expertise at COMO (VUB, como.vub.ac.be). The more biologically oriented partners, IRIBHM (ULB, iribhm.ulb.ac.be) and Swit ch (VUB, switch.vub.ac.be), have a strong background in the analysis of microarray data and functional analysis of SNPs. The successful candidate will be based at the Switch lab. REQUIREMENTS: a Ph.D. or master's degree in statistical genetics, genetic epidemiology, biostatistics, bioinformatics, statistics, computational biology or computer science, with interest in genetics and genomics. Candidates with doctoral degrees in genetics or related fields with strong computational and statistical background are also encouraged to apply. Knowledge of genetic packages is a plus. The candidate should demonstrate ability in scientific programming, ideally in R. Knowledge of Unix, SQL, Perl and C/C++ is desirable. The contract will last till the 31th of August 2010. A possible prolongation depends on the success of the project. Candidates are invited to send their CV with the detailed list of publications and a research statement in .pdf format by email to jreumers(a)vub.ac.be

1 0

OIO and postdoc positions in biomodeling at NISB and CWI
by Roeland Merks 04 Nov '08

04 Nov '08

Vacancy The Netherlands Institute for Systems Biology (NISB) and Centrum Wiskunde & Informatica (CWI) invite applications for the position of a Postdoc (two years, with possibility for extension): Computational modeling of structural genomic variation in hierarchically-regulated gene networks The opening is a research position within the field of computational biology Research background Structural variation of the genome, including translocation of DNA chunks across chromosomes, inversions, and deletions can have enormous impact on gene expression and consequently on cell function. Hierarchical regulation mechanisms, including conditional DNA folding and nuclear lamina interactions, can switch gene clusters between a state rendering the genes inaccessible to transcription and a “permissive” state in which gene transcription is possible. The consequences of hierarchical network organization for the dynamics and evolution of gene networks are poorly understood. The project will develop computational and mathematical modeling approaches to explore the effects that structural genomic variation can have on gene expression in hierarchically-structured gene networks at single-organism and evolutionary time scales. Job description The postdoc will develop and analyze theoretical, evolutionary models of hierarchical gene network regulation using discrete and/or continuous modeling approaches. Model insights and predictions based on artificial and actual regulatory networks will be interpreted in close collaboration with two functional genomics groups that participate in the NCSB: the Edwin Cuppen group at the Hubrecht laboratory, Utrecht and the Lodewyk Wessels group at the Netherlands Cancer Institute in Amsterdam. Location The work will be embedded in the Biomodeling and Biosystems Analysis group of the Netherlands Institute for Systems Biology (NISB; www.sysbio.nl) and within the Life Sciences group of the Center for Mathematics and Computer Science (CWI) in Amsterdam. As the “core modeling group” of the Netherlands Consortium for Systems Biology, the Biomodeling and Biosystems Analysis group develops quantitative and predictive models and multiscale computer simulations in collaboration with systems biology groups at participating institutes. Profile Candidates ideally fulfill the following criteria: - Ph.D. theoretical biology, computer science, mathematics or a related discipline - able to communicate with scientists in biology and mathematics - good programming skills in C++ or equivalent Conditions of employment The vacancy concerns a temporary research position for two years, with possibility of extension. The salary and terms of employment are in accordance with the "CAO-onderzoekinstellingen" and are based on qualifications. Besides the salary and excellent working conditions, CWI offers attractive and flexible terms of employment, like a collective health insurance, pension-fund, and initial help with housing for foreigners. The position is at postdoc level for 38 hrs/week. Websites http://www.sysbio.nl and http://www.cwi.nl Applications and requests for information Direct requests for information or applications before 1 December 2008 to: Dr. Roeland Merks, Roeland.Merks(a)sysbio.nl phone +31 20 592 4117, skype: roelandmerks Centrum Wiskunde & Informatica (CWI) and Netherlands Institute for Systems Biology (NISB) Kruislaan 413, 1098SJ Amsterdam, The Netherlands ----- Vacancy The Netherlands Institute for Systems Biology (NISB) and Centrum Wiskunde & Informatica (CWI) invite applications for the position of a: PhD student (OIO): Cell-based modeling of in vitro angiogenesis: role of tip cell selection, cross-talk between diffusive and contact-dependent signaling, and endothelial cell-matrix interactions. The opening is a research position within the field of computational biology Research background Angiogenesis, the outgrowth of sprouts from existing blood vessels, is essential for many biomedical phenomena, including tumor growth and tissue engineering. A detailed understanding of its mechanisms will help identify specific targets for therapeutical intervention. The key players of angiogenesis are endothelial cells (ECs), the cells lining the inner walls of mature blood vessels, and the surrounding matrix proteins. We have previously built simple models of the ECs’ self-organization into angiogenic sprouts. In this project we aim to refine our understanding of angiogenic sprouting, by including the selection of “tip cells”—the leading cells of blood vessel sprouts—and the interactions between the ECs and the protein matrix they live in. The final goal of this project is to develop a multiscale, explanatory model of angiogenesis. Job description The PhD student will further develop and analyze cell-based models of angiogenesis and, depending on interest, validate the cell-based models using mean-field approaches. Techniques include ordinary differential equations (ECs’ regulatory networks), the Cellular Potts Model (cell behavior) and PDEs (protein matrix). Model insights and predictions will be interpreted and validated in close collaboration with our experimental collaborators. Location The work will be embedded in the Biomodeling and Biosystems Analysis group of the Netherlands Institute for Systems Biology (NISB) and within the Life Sciences group of the Center for Mathematics and Computer Science (CWI) in Amsterdam. As the “core modeling group” of the Netherlands Consortium for Systems Biology, the Biomodeling and Biosystems Analysis group develops quantitative and predictive models and multiscale computer simulations in collaboration with systems biology groups at participating institutes. Profile Candidates ideally fulfill the following criteria: - M.Sc. in theoretical biology, computer science, mathematics or a related discipline - able to communicate with scientists in biology and mathematics - good programming skills in C++ and Python or equivalent Conditions of employment The vacancy concerns a temporary research position for four years. The salary and terms of employment are in accordance with the "CAO-onderzoekinstellingen". Besides the salary, CWI offers attractive and flexible terms of employment, like a collective health insurance, pension-fund, and initial help with housing for foreigners. Websites http://www.sysbio.nl and http://www.cwi.nl Applications and requests for information Direct requests for information or applications before 1 December 2008 to: Dr. Roeland Merks, Roeland.Merks(a)sysbio.nl phone +31 20 592 4117, skype: roelandmerks Centrum Wiskunde & Informatica (CWI) and Netherlands Institute for Systems Biology (NISB) Kruislaan 413, 1098SJ Amsterdam, The Netherlands Applications should include a motivation letter, a curriculum vitae, list of publications, and the names and addresses of at least two persons that can be approached to obtain further information. Closing time 1 December 2008 Literature Roeland M. H. Merks, Sergey V. Brodsky, Michael S. Goligorksy, Stuart A. Newman, James A. Glazier (2006). Cell elongation is key to in silico replication of in vitro vasculogenesis and subsequent remodeling. Developmental Biology 289, 44-54. doi:10.1016/j.ydbio.2005.10.003 Roeland M. H. Merks, Erica D Perryn, Abbas Shirinifard, James A Glazier (2008). Contact-inhibited chemotaxis in de novo and sprouting blood-vessel growth. PloS Computational Biology 4 (9), e1000163. doi: 10.1371/journal.pcbi.1000163 -- ==================================================================== Note: from September 1st 2008 I will move to the Netherlands Institute for systems biology (www.sysbio.nl) and to CWI, the National Center for Mathematics and Computer Science (www.cwi.nl) in Amsterdam. However, I will be in Ghent until September 5th 2008 for our plant modeling summer school: www.psb.ugent.be/plantmodels. My current e-mail address will still be operational for a while, but it will soon change to: Roeland.Merks(a)cwi.nl and Roeland.Merks(a)sysbio.nl. If any of these addresses does not (yet) work, please use my personal e-mail address: post(a)roelandmerks.nl. ================================================================= Roeland Merks Group Leader "Plant Systems Modeling" Tel:32 (0)9 3313823 Fax:32 (0)9 3313809 VIB DEPARTMENT OF PLANT SYSTEMS BIOLOGY, UGent Technologiepark 927, 9052 Gent, BELGIUM roeland.merks(a)psb.ugent.be http://www.psb.ugent.be ==================================================================

1 0

open tenure track position at CWI, Amsterdam
by Nada Mitrovic 04 Nov '08

04 Nov '08

In the context of its recently initiated priority research theme, `Earth and Life Sciences', the Centrum voor Wiskunde en Informatica (CWI) invites applications for the position of a scientific staff member (tenure track). CWI is the national research institute for mathematics and computer science in the Netherlands. CWI performs frontier research in mathematics and computer science and transfers new knowledge in these fields to society in general and trade and industry in particular (see http://www.cwi.nl/) The opening is a research position within the field of * bio-statistics and/or * bio-informatics/computational biology and/or * bio-mathematics CWI intends to extend its activities in these fields and to initiate new research groups that focus on computer science and/or mathematics for the life sciences. Additional support is available for the new research groups in the form of postdoc positions supplied by CWI. Profile * The applicant is an independent researcher who has a demonstrated record of high-quality research * He or she has experience in collaborating with scientists working in the life sciences. * The applicant also has the intention and ability to search for project funding, and to build up a research group. Terms of employment The vacancy concerns a tenure-track position, with an initial appointment for three years. Depending on the candidate's experience, different terms of appointment can be discussed. The salary and terms of employment are in accordance with the "CAO-onderzoekinstellingen" and are based on qualifications. Moreover, CWI offers attractive working conditions, including flexibility and initial help with housing for foreigners. Information and application Further information on this position can be obtained through Nada Mitrovic (Nada.Mitrovic(a)cwi.nl) An application should consist of a cover letter, a curriculum vitae, a list of publications, a research statement, and the names of three references. The research statement should in any case include the applicant's perspective, experience and plans regarding: the relevance of the applicant's research in the life sciences, the applicant's contacts with researchers in the life sciences, and the mathematical and computer science theory and techniques the applicant develops for and applies to the life sciences. The deadline for applications is set on 16 December 2008. Please send the application to the following address: CWI Personnel & Organisation (P&O) PD(a)cwi.nl P.O. box 94079 1090 GB Amsterdam The Netherlands

1 0

PhD research position in Bioinformatics & Machine Learning in ULB-MLG, Brussels, Belgium
by Gianluca Bontempi 23 Oct '08

23 Oct '08

***PhD research position in Bioinformatics & Machine Learning*** Functional Genomics Unit, IJB & Machine Learning Group, ULB Brussels, Belgium *Position description Applications are invited for a full-time PhD research position at the Functional Genomics Unit (FGU), Medicine Department, Institut jules Bordet (IJB), Belgium. The researcher will get his/her PhD in the Machine Learning Group (MLG), Computer Science Department, Université Libre de Bruxelles (ULB), Belgium. The vacancy available to be filled in soon offers tenure of four years. The researcher is expected to carry out fundamental research in Machine Learning (MLG) and to take an active role in the applied research projects of the FGU, which is funding the position. The Functional Genomics Unit is a research group that carries out applied research on genomics in breast cancer. Cutting-edge bio- technologies are used to bring new insights in breast cancer biology. These new discoveries are then brought to clinic (translational research) in order to improve the management of breast cancer patients. The Machine Learning Group is a research group that carries out theoretical research on statistical machine learning, predictive modeling and computational statistics, and takes part in several applied research projects in domains as diverse as data mining, bioinformatics to biomedical engineering and wireless sensor networks. This inter-disciplinary position provides unique opportunities in bioinformatics. Previous studies involving microarray and survival data in breast cancer led to high impact publications (see http://www.ulb.ac.be/di/map/bhaibeka/research.html for an overview). The machine learning methods introduced in these articles deal specifically with the high dimensionality of the microarray data and the fitting/validation of models for survival prediction of the patients. *Qualifications Candidates are expected to attach their transcripts and a letter of recommendation from their MSc supervisor or any other relevant person. Applicants are expected to have strong programming skills, the ability to work independently, strong interpersonal skills, good writing and oral communication. Any research experience on data mining, machine learning, and statistical analysis will be an asset as well as an experience in microarray data analysis. Upon offer of a position, the candidates who are non-EU citizens are required to obtain an appropriate long-term Schengen visa; further information is available from their nearest Belgian embassy. *Deadline Applicants are encouraged to apply as soon as possible as positions will remain open until filled in by a suitable candidate. *Contact Candidates are invited to send their CV with the detailed list of publications and a research statement in .pdf format by email to gbonte(a)ulb.ac.be The subject field of your email must include MLG-JOB: research position. *Equal Opportunity Employer ULB is an equal opportunity employer and is committed to employing more handicapped individuals and especially encourages them to apply. ULB wishes to increase the proportion of women in areas in which they are under-represented. Women are strongly encouraged to apply. -- Pr. Gianluca Bontempi Head of the Computer Science Department - Département d'Informatique Head of the Machine Learning Group Université Libre de Bruxelles Boulevard du Triomphe - CP212 1050 Bruxelles, Belgium email: gbonte(a)ulb.ac.be Office Phone: +32-2-650 55 91 Fax: +32 2 650.56.09 web: http://www.ulb.ac.be/di/mlg

1 0

Soutenance publique de la thèse de Sylvain Brohée
by Jacques van Helden 22 Oct '08

22 Oct '08

Soutenance publique de la thèse de Sylvain Brohée Etude bioinformatique du réseau d'interactions entre protéines de transport chez les Fungi Directeurs de thèse: Jaques van Helden & Bruno André Lundi 10 novembre 2008 à 16:30 ULB Campus Plaine, Salle Solvay Bâtiment N/O, niveau 5 Boulevard du Triomphe, Accès 2, 1050 Bruxelles Plan d'accès: http://www.ulb.ac.be/docs/campus/pla_NO.html Résumé de la thèse Les protéines associées aux membranes sont d'une importance cruciale pour la cellule. Cependant, en raison d'une plus grande difficulté de manipulation, les données biochimiques les concernant sont très lacunaires, notamment au point de vue de la formation de complexes entre ces protéines. L'objectif global de notre travail consiste à combler ces lacunes et à préciser les interactions entre protéines membranaires chez la levure Saccharomyces cerevisiae et plus précisément, entre les transporteurs . Nous avons commencé notre travail par l'étude d'un jeu de données d'interactions à grande échelle entre toutes les perméases détectées par un méthode de double hybride spécialisée pour les protéines membranaires. La qualité des données a été estimée en étudiant le comportement global des données et des témoins négatifs et positifs. Les données ont ensuite été standardisées et filtrées de façon à ne conserver que les plus significatives. Ces interactions ont été étudiées en les modélisant dans un réseau d'interactions que nous avons étudié par des techniques issues de la théorie des graphes. Après une évaluation systématique de différentes méthodes, nous avons notamment recherché au sein du réseau des groupes de protéines densément interconnectées et de fonctions similaires (clustering sur graphe) qui correspondraient éventuellement à des complexes protéiques. Les résultats révélés par l'étude du réseau expérimental se sont révélés assez décevant. En effet, même si nous avons pu retrouver certaines interactions déjà décrites, un bon nombre des interactions filtrées semblait n'avoir aucune réalité biologique et nous n'avons pu retrouver que très peu de modules hautement connectés de fonction homogène. L'approche expérimentale n'ayant eu que peu de succès, nous l'avons contournée en utilisant des méthodes de génomique comparative d'inférence d'interactions fonctionnelles. Dans un premier temps, malgré une évaluation rigoureuse, l'étude des profils phylogénétiques (la prédiction d'interactions fonctionnelles en étudiant la corélation des profils de présence – absence des gènes dans un ensemble de génomes), n'a eu que des résultats mitigés car les perméases semblent très peu conservées dès lors que l'on considère d'autres organismes que les Fungi. Par la suite, nous avons ensuite développé une nouvelle méthode qui construit un réseau de co-régulation génétique sur base de la similarité des empreintes phylogénétiques découvertes dans tous les gènes de levure. Cette méthode, qui n'utilise que les séquences génomiques pour l'inférence du réseau, a donné de très bons résultats, validés in silico et expérimentalement et son application à l'ensemble des perméases a permis de retrouver plusieurs régulations bien connues, d'en inférer de nouvelles et de proposer de nouvelles pistes de recherche. En parallèle à ce travail, nous avons développé et rendu accessible NeAT (http://rsat.bigre.ulb.ac.be/neat/) un ensemble d'outils informatiques consacrés à l'analyse de réseaux biologiques.

1 0