JOBIM 2009

Author notifications

2009-04-30T08:08:53Z

Posters guidelines

The dimensions of poster layouts are 150 x 95 cm.

The A0 format is recommended for printing posters.

Fasteners (Velcro) will be provided.

Instructions aux auteurs

2009-04-30T07:51:34Z

Posters

Les dimensions des panneaux pour l'affichage des posters sont les suivantes : 150 x 95 cm

Le format A0 est conseillé pour l'impression des posters.

Les attaches (velcro) seront fournies.

RULBI

2009-04-15T10:01:05Z

Date : 12 juin 2009

Lieu : Cité des congrès

Description : L'exploitation par les biologistes, de l'ensemble des données issues d'approches transcriptomiques, protéomiques, métabolomiques, dans l'optique d'une biologie intégrative, nécessite l'utilisation de logiciels bioinformatiques perfectionnés comme GENOMATIX, INGENUITY PATHWAY STUDIO ou GENESPRING… Ces outils permettent d'intégrer les données issues de la bibliographie aux données expérimentales, dans le but de reconstruire des réseaux biologiques (fouille de données et de textes). Leur utilisation est déjà largement répandue dans de grands instituts de recherche internationaux et dans des laboratoires pharmaceutiques. Ils comportent des bases de données de qualités associées aux outils d'analyse couplés à une interface aboutie qui se prête parfaitement à leur maitrise par des biologistes non formés en informatique. Ces logiciels ne sont actuellement disponibles que sous licence à des prix incompatibles avec une large diffusion dans les laboratoires de recherches académiques. Comme tout moyen de recherche, leur utilisation est souvent discontinue voire ponctuelle ce qui rend leur acquisition problématique au regard des budgets des équipes.

Nous avons œuvré depuis plusieurs mois auprès de ces sociétés pour faire baisser le prix des licences. Un recensement des besoins, couplé à une négociation des prix au niveau national était indispensable pour permettre à une majorité de laboratoires de recherche d'avoir accès à ces logiciels.

Le RULBI (Réseau d'Utilisateurs de Logiciels Bio-Informatiques) vient de recevoir le soutien du GIS IBIsA pour sa création. A l'interface entre les éditeurs de logiciels et les biologistes ce réseau vise à faciliter l'accès aux logiciels payants pour les biologistes des laboratoires de recherche publique en négociant leur prix et en mutualisant ceux-ci dans l'optique d'une utilisation plus rationnelle et donc plus économique.

Lors de cette journée satellite, après la présentation de l'action actuelle du RULBI les scientifiques utilisateurs pourront présenter les travaux réalisés avec ces outils. Par ailleurs, les sociétés éditeurs de ces logiciels présenteront les potentialités actuelles et à venir de ces outils. Un groupe de travail sera constitué pour identifier les nouveaux besoins et évaluer d'autres logiciels qui pourraient également faire l'objet de négociation de prix et de mutualisation. Les éditeurs de logiciels ont également proposé d'animer des micro-ateliers où des données expérimentales personnelles (de taille raisonnable) pourraient être traitées avec les différents outils.

Programme complet :

Programme

La participation à cette journée est gratuite mais pour son organisation la préinscription des participants est indispensable avant le 20 mai 2009

Contact : joelle Henry-Berger, UMR 6247-GReD – U931, Campus Universitaire des Cézeaux 63177 AUBIERE Cedex Tel : 04.73.40.77.57 : johenry@univ-bpclermont.fr

Partenaires : IBISA, GENOMATIX, PATHWAY STUDIO-ARIADNE, GENESPRING

Coté calcul

2009-04-01T13:02:58Z

Date : Lundi 8 juin 14h-18h

Lieu : A préciser

Description : Les acteurs en BIM (Biologie, Informatique et Mathématiques) font face à une augmentation considérable de la taille des objets qu'ils manipulent et modélisent ; on pense notamment aux nouvelles données de séquençage (nombre de séquences et d'espèces), à la taille des réseaux biologiques ou à la complexité des modèles de simulation. Par ailleurs, on assiste à une modification de l'évolution du matériel de calcul vers du “plus nombreux” aux dépends du “plus rapide”.

Ce contexte invite tout naturellement à se préoccuper des aspects de calcul, de l'algorithmique jusqu'à l'architecture matérielle, suivant plusieurs facettes :

la mise au point d'algorithmes de plus faible complexité
le développement d'algorithmes adaptés au calcul parallèle et distribué
l'utilisation ou le développement de bibliothèques logicielles optimisées
le couplage entre développement et architecture cible (thread SMP, MPI, distribution)
le recours aux nouvelles technologies HPC
le “divide and conquere” des architectures de type grille

Plus de détails

Organisateurs :

Christophe Blanchet, IBCP Lyon, christophe.blanchet/AT/ibcp.fr
Christophe Caron, Station Biologique de Roscoff, caron/AT/sb-roscoff.fr
Vincent Miele, LBBE Lyon, miele/AT/biomserv.univ-lyon1.fr
Bruno Spataro, LBBE Lyon, bspataro/AT/biomserv.univ-lyon1.fr

Modélisation dynamique

2009-04-01T12:46:50Z

Date : Vendredi 12 juin 2009

Lieu : Ecole Centrale de Nantes

Description :

Dans le cadre des Journées Ouvertes Biologie, Informatique et Mathématiques (JOBIM 2009) et dans la foulée des rencontres organisées depuis 2003, nous proposons une nouvelle réunion satellite sur la modélisation dynamique et la simulation des réseaux biologiques. Cette rencontre est organisée avec le soutien du GDR Bio-Informatique et de la Région Pays de la Loire via le projet BIL (Bio-Informatique Ligérienne). Nous lançons un appel à communication aux jeunes doctorants ou post-doctorants. Dans une démarche d'animer la communauté de la thématique, nous souhaitons réunir les jeunes doctorants ou post-doctorants, leur permettre de présenter leurs travaux et dialoguer avec un panel de chercheurs du domaine.

Pour plus de détails

Organisateurs :

Grégory Batt,, INRIA, Rocquencourt.
Jérémie Bourdon, LINA, Nantes et IRISA, Rennes.
Claudine Chaouiya,, TAGC, INSERM U928.
Hidde de Jong,, INRIA, Grenoble.
Damien Eveillard, LINA, Nantes.
Adrien Richard, I3S, Nice.
Delphine Ropers, INRIA, Grenoble
Olivier Roux, IRCCyN, Nantes
Anne Siegel, IRISA, Rennes.
Denis Thieffry,, TAGC, INSERM U928 et INRIA Rocquencourt.

Final version

2009-03-27T13:43:28Z

The final version of articles must be uploaded to EasyChair submission server before ** Tuesday April 14 **. We have very tight deadlines for printing documents, please respect this date.

Your contribution text may be in English or French, the abstract should be in English. Submissions must use presentation templates RTF and LaTeX :

Modèle Latex

Modèle Latex (cls)

Modèle RTF

Contributions not compliant with the templates will not be taken into account. Of course, authors are requested not to change the settings of presentation (margins, font size, font, ...).

Format : long paper : 6 pages maximum.

Format : short paper : 2 pages maximum.

Submission

The final version must be sent in editable format (RTF or LaTeX + figures), and as a PDF.

To upload your documents, use the submission server and the menu "Submit a new version". Your final submission will contain:

your final PDF "jobimXXX.pdf". The PDF file should be sent via "File" field.
all your sources: an archive "jobimXXX.zip" in a directory "jobimXXX" containing :
- A "jobimXXX.rtf"
- Or a master file "jobimXXX.tex" and any other files.

The ZIP file should be sent via the "Attachment" field.

("XXX" is replaced by the number given by easychair to your submission)

Submission long paper (6 pages maximum)

Submission short paper (2 pages maximum)

In addition to abstract book distributed during the conference, we plan to publish all final articles on the website. If you are opposed to this electronic publication, thank you to let us know until April 14 by sending an email to jobim09@univ-nantes.fr

Long talks

2009-03-27T09:59:18Z

Abstract of selected articles for a long talk (25 min)

Emmanuelle Becker, Alain Guénoche and Christine Brun. Système de Classes Chevauchantes pour la Recherche de Protéines Multifonctionnelles.

Abstract: This work aims at developing a method to detect multifunctional proteins, i.e. proteins performing several apparently unrelated functions. To detect these proteins, we consider a network of binary direct interactions between proteins that we decompose in an overlapping class system using a criteria based on graph topology and extending Newman's modularity. As a result, some proteins are found in several final classes meaning that they are interacting with several groups of proteins apparently functionally unrelated. These multiply classified proteins are thus good candidates for multifunctionality. In this paper, we will first introduce the concept of multifunctionality, then fully explain the method, and finally present the preliminary results obtained by applying the method to a large human protein interaction network.

Etienne Birmele. Detecting Network Motifs by Local Concentration

Abstract: Biological networks exhibit small over-represented subgraphs, called motifs, some of which are known to have a biological function. Several algorithms exist to detect motifs, most of them being based on time-consuming simulations or leading to many false positives. We propose an efficient and conservative procedure to detect network motifs and apply it on the Yeast gene regulation network.

Chun-Long Chen, Aurélien Rappailles, Lauranne Duquenne, Maxime Huvet, Guillaume Guilbaud, Benjamin Audit, Yves d'Aubenton-Carafa, Alain Arneodo, Olivier Hyrien and Claude Thermes. Single-nucleotide substitution rates increase during the replication S phase of the human genome

Abstract: Naturally occurring mutations in mammalian genomes play a key role in evolution and genetic disease but their causes are still poorly understood. In particular, nucleotide substitutions occur at strongly variable rates along genomes and it is essential to unravel the mechanisms responsible of these fluctuations. A number of evolutionary studies have exhibited complex correlations between substitution rates and parameters like regional or local nucleotide composition, crossover rate or distance to telomeres (1-6). Here, we study the role of replication on neutral substitution rates in the human genome. Using replication timing data determined by massive sequencing of replicating strands, we show that all non-CpG substitution rates correlate with timing: they are minimum in early replicating regions and increase to maximum values in late regions. These correlations are still observed after controlling for nucleotide composition, cross-over rate and distance to telomeres. These data demonstrate for the first time that replication timing plays a key role in shaping the profile of mutations along the genome.

Anne Crumiere. Cellular automata modeling of intercellular genetic regulatory networks

Abstract: Biologists often represent genetic interactions by directed graphs, named genetic regulatory graphs. Vertices represent genes, whereas edges represent regulatory effects from one gene on another. Edges are labelled with a positive sign in the case of an activation and negative for an inhibition. This article deals with relationships between the structure of such graphs and their dynamical properties.

The biologist R.Thomas enounced, thirty years ago, the following two general rules: a necessary condition for multistability is the presence of a positive circuit in the regulatory graph (the sign of a circuit being the product of the signs of its edges) and the existence of a negative circuit is a necessary condition for the existence of an attractive cycle. These rules are about the dynamic of a single cell, and it has given rise to mathematical statements and proofs. This article aims at extending these rules to regulatory interactions spanning within cells and between cells in the discrete formalism.

Hugo Devillers, Hélène Chiapello, Meriem El Karoui and Sophie Schbath. How to measure the robustness of bacterial genome comparisons?

Lionel Dupuy, Matthieu Vignes, Blair McKenzie and Philip White. Meristematic Waves, a new approach to model root architecture dynamics

Abstract: During their development, plants must develop efficient root architectures to secure access to nutrients and water in soil. A series of expansion and branching mechanisms fulfils this aim in the proximity of root apical meristems where the plant senses the environment and explores immediate regions of soil. We have developed a new approach to study the dynamics of root meristems in soil, using the relationship between the increase in root length density and the root meristem density. Initiated at the seed, the location of root meristems was shown to propagate, wave-like, through the soil, leaving behind a permanent network of roots for the plant to acquire water and nutrients. Models higlighted that the morphologies of the waves of meristems are inherent to individual root developmental processes, namely expansion, lateral root initiation and gravitropic responses. The "meristematic wave" observed on data collected on barley might be a more general and fundamental aspect of plant rooting strategies to access underground resources.

Paul Garcin and Yves Boulard. Construction et analyse d'un modèle tridimensionnel du complexe [(SLR1738-Zn-Fe)2-ADN]

Abstract: Slr1738 is the Peroxide regulon Repressor protein (PerR) of Synechocystis. Active as a dimer, this protein must contain an iron atom to be able to bind DNA molecule and regulates targeted genes. The binding mechanism involves a classic recognition helix inserted in the DNA major groove. But to date there is no three-dimensional structure available for this kind of transcription factor complexed to DNA. As a consequence, both global and specific interactions that lead this protein to bind DNA and to recognize specific ‘Per Box' sequence are still misunderstood. In order to better define and analyse these interactions, we built in silico the first three-dimensional structure of a [PerR-DNA] complex. This article describes the method used to build the complex and presents an analysis of the contacts between the two partners.

Sandrine Grossetete, Bernard Labedan and Olivier Lespinet. FUNGIpath: a new tool for analysing the evolution of fungal metabolic pathways

Abstract: FUNGIpath is a new tool dedicated to perform in-depth analysis of fungal metabolic pathways. It is freely accessible at http://www.fungipath.u-psud.fr. FUNGIpath consists in a collection of orthologous groups of proteins that have been predicted using complementary methods of detection and further mapped on KEGG and MetaCyc pathways. It allows an easy comparison of the primary and secondary metabolisms afforded by the different fungal species present in the database with the possibility to assess the level of specificity of various pathways at different taxonomic distances. As more and more fungal genomes are expected to be decrypted in the next years, this tool is expected to help to progressively reconstruct what were the primary and secondary metabolisms of the ancestors of the main branches of the fungi tree and to understand how these ancestral fungal metabolisms evolved to various specific derived metabolisms.

Nicolas Lebreton, Christophe Blanchet, Julie Chabalier and Olivier Dameron. Utilisation d'ontologies de tâches et de domaine pour la composition semi-automatique de services Web bioinformatiques

Abstract: Nowadays, bioinformatics tasks typically involve large scale data analysis that require the integration of web services from heterogeneous platforms. In spite of the efforts for improving Web services interoperability, integration remains difficult and still has to be performed manually by users. Improving the composition of Web services requires to analyze what Web services do as well as the nature and the type of their input and output parameters. This work shows that existing technologies support automating the selection, composition and execution of Web services, and that the current limiting factor to a wider use is the lack of precise enough task and domain ontologies.

Aurelie Leduc, Pierre Nicolas, Philippe Bessieres and Stephane Robin. Probabilistic modeling of tiling array expression data

Abstract: For organisms with small genomes such as bacteria, the current microarray technology allows adopting a tiling design where the whole genome is covered by overlapping probes. These arrays permit to measure the transcriptional activity of the whole genome with unprecedented resolution. Model-based approaches currently used to analyze these data remain however very simple, the most popular model being the piecewise constant Gaussian model with a fixed number of breakpoints. Here we present a new approach based on hidden Markov modelling designed for the probabilistic reconstruction the trajectory of a continuous-valued signal. The use of this model does not require the choice of a fixed number of breakpoints and permits to account for subtle effects such as drift in the signal. The model also includes direct correction for the variations of probe affinities via the use of covariates.

Celine Lefebvre, Mariano Alvarez, Presha Rajbhandari, Wei Keat Lim and Andrea Califano. Master regulator analysis reveals key transcription factors for Germinal Center formation

Abstract: We describe a new method for the identification of master regulators of a phenotype of interest. The master regulator analysis identifies transcription factors that are candidate master regulators of a phenotype of interest based on its transcriptional targets. We applied this method for deciphering the regulation of Germinal Center B cell programs, revealing the two transcription factors MYB and FOXM1 as synergistic master regulators

Sébastien Loriot, Frédéric Cazals, Michael Levitt and Julie Bernauer. A geometric knowledge-based coarse-grained scoring potential for structure prediction evaluation

Abstract: Knowledge-based protein folding potentials have proven successful in the recent years. Based on statistics of observed interatomic distances, they generally encode pairwise contact information. In this study we present a method that derives multibody contact potentials from measurements of surface areas using coarse-grained protein models. We show that this construction is able to distinguish native structures from decoys. We tested different potentials from a reference set of 66 protein structures. These functions, encoding up to 5-body contacts are evaluated on the reference set and its 45000 decoys and also on the often used lattice_ssfit set from the decoys'R us database. We show that the most relevant information for discrimination resides in 2- and 3-body contacts. The potentials we have obtained can be used for evaluation of putative structural models; they could also lead to different type of protein structure refinement that uses multi-body interactions

Christine Martin and Antoine Cornuéjols. Using Frequent and Surprising Item Sets for the Characterization of Protein-Protein Interfaces

Abstract: Numerous research effort have aimed to characterize and predict protein-protein interfaces. This paper introduces a method that rely only on known protein-protein interfaces (positive instances only). It combines frequent item set mining techniques with statistical tests to ensure the selection of interesting features. Starting from a database of known interfaces described with geometrical elements, the method produces the elements and combinations thereof that are \emphcharacteristic of the interfaces. This approach allows one to easily interpret the results, as compared to techniques that operate as ``black-boxes'' and ensures a satisfactory proportion of reliable item sets. The results obtained on a set of 459 protein-protein interfaces from the DOCKGROUND database confirm that the findings are consistent with current knowledge about protein-protein interfaces.

Marie-José Mhawej and Claude H. Moog. Drug dosage control of the HIV infection dynamics

Abstract: An increasing number of sophisticated control algorithms become available in the current literature to optimize the HIV therapy. Unfortunately, the pharmacokinetics and pharmacodynamics of antiretroviral drugs are ignored and these algorithms remain purely theoretic. This issue is investigated explicitly in this paper. An elementary pharmacodynamics model is combined with a non linear feedback control computed from standard engineering methods. It is shown that it results in the design of a realistic dosage regimen which drives the immunological system close to the healthy equilibrium state. Although the problem is dealt as a single input system, it is argued that the procedure can be extended to a multitherapy design or to any available control law.

Gregory Nuel. Counting patterns in degenerated sequences

Abstract: In this paper, we propose a rigourous method to take into account the uncertainty of sequencing for biological sequences (DNA, Proteins). For example, this method allows to study the distribution of a pattern of interest in a degenerated sequence defined on the standard IUPAC DNA alphabet. We first introduce a Forward-Backward approach to compute the marginal distribution of the constrained sequence and use it both to perform a Expectation-Maximization estimation of parameters, as well as deriving a heterogeneous Markov distribution for the constrained sequence. This distribution is hence used along with known DFA-based pattern approaches to obtain the exact distribution of the pattern count under the constraints. As an illustration, we consider a EST dataset from the EMBL database. Despite the fact that only 1% of the position in this dataset are degenerated, we show that not taking into account these positions might lead to erroneous observations, further proving the interest of our approach.

Nicolas Terrapon, Olivier Gascuel and Laurent Brehelin. Détection de nouveaux domaines protéiques par co-occurrence : Application à P. falciparum

Abstract: Hidden Markov Models (HMMs) have proved to be powerful for protein domain identification. However, numerous domains may be missed in highly divergent proteins. This is the case for the proteins of Plasmodium falciparum, the main causal agent of human malaria. Here, we propose a method that uses domain co-occurrence to increase the sensitivity of the approach while controlling its false discovery rate. Applied to P. falciparum, our method identify (with an error rate below 20%) 482 new domains (versus 3482 in PlasmoDB), which involve 158 new GO annotations.

Raluca Uricaru, Celia Michotey, Laurent Noe, Hélène Chiapello and Eric Rivals. Improved sensitivity and reliability of anchor based genome alignment

Abstract: Whole genome alignment is a challenging problem in computational comparative genomics. It is essential for the functional annotation of genomes, the understanding of their evolution, and for phylogenomics. Many global alignment programs are heuristic variations on the anchor based strategy, which relies on the initial detection of similarities and their selection in an ordered chain. Considering that alignment tools fail to align some pairs of bacterial strains, we investigate whether this is intrinsically due to the strategy or to a lack of sensitivity of the similarity detection method. For this, we implement and compare $6$ programs based on three different detection methods (from exact matches to local alignments) on a large benchmark set. Our results suggest that the sensitivity of well known methods, like MGA or Mauve, can be greatly improved in the case of divergent genomes if one exploits spaced seeds at the detection phase. In other cases, such methods yield alignments that covers nearly the whole genome. Then, we focus on global reliability of alignments: should an aligned pair of segments be included in the global genome alignment? We investigate this reliability according to both the segment "alignability" and to inclusion of orthologs. Again, we provide evidence that for both close and divergent genomes, one of our program, YH, achieves alignments with sometimes a lower coverage, but a higher inclusion of orthologs. It opens the way to the first reliable alignments for some highly divergent species like Buchnera aphidicola or Prochlorococcus marinus.

Version finale

2009-03-27T09:28:36Z

La version finale doit être déposée sur le serveur de soumission EasyChair au plus tard ** le mardi 14 avril **. Nous avons des délais très serrés pour l'impression des actes, merci de respecter cette date.

Le texte des contributions peut être rédigé en anglais ou en français, le résumé devant être en anglais. Les soumissions doivent impérativement utiliser les modèles de présentation RTF et LaTeX suivants :

Modèle Latex

Modèle Latex (cls)

Modèle RTF

Les contributions ne respectant pas le formatage ne seront pas prises en considération. Les auteurs sont bien sûr priés de ne pas changer les réglages de présentation (marges, taille des caractères, polices, ...).

Format : Communication longue : 6 pages maximum.

Format : Communication courte : 2 pages maximum.

Soumission

La version finale devra être transmise en format éditable (RTF ou LaTeX + figures), ainsi que sous forme de PDF.

Pour déposer vos documents, utilisez le serveur de soumission et le menu "Submit a new version". Votre soumission finale contiendra :

votre PDF final "jobimXXX.pdf". Le fichier PDF doit être envoyé via le champ "File".
l'ensemble de vos sources : une archive "jobimXXX.zip" contenant, dans un répertoire "jobimXXX"
- soit un fichier "jobimXXX.rtf"
- soit un fichier principal "jobimXXX.tex" ainsi que tout autre fichier utile.

Le fichier ZIP doit être envoyé via le champ "Attachment".

("XXX" est à remplacer par le numéro donné par easychair à votre soumission)

Soumission papier long (6 pages maximum)

Soumission papier court (2 pages maximum)

En plus des actes "papier" distribués durant la conférence, nous prévoyons de publier l'ensemble des articles finaux sur le site web. Si vous êtes opposé à cette publication électronique, merci de nous le faire savoir d'ici le 14 avril en envoyant un courriel à jobim09@univ-nantes.fr

Présentations longues

Audrey Bihouée — 2009-03-26T15:49:00Z

Résumés des articles sélectionnés pour une présentation longue

Emmanuelle Becker, Alain Guénoche and Christine Brun. Système de Classes Chevauchantes pour la Recherche de Protéines Multifonctionnelles.

Abstract : This work aims at developing a method to detect multifunctional proteins, i.e. proteins performing several apparently unrelated functions. To detect these proteins, we consider a network of binary direct interactions between proteins that we decompose in an overlapping class system using a criteria based on graph topology and extending Newman's modularity. As a result, some proteins are found in several final classes meaning that they are interacting with several groups of proteins apparently functionally unrelated. These multiply classified proteins are thus good candidates for multifunctionality. In this paper, we will first introduce the concept of multifunctionality, then fully explain the method, and finally present the preliminary results obtained by applying the method to a large human protein interaction network.

Etienne Birmele. Detecting Network Motifs by Local Concentration

Abstract : Biological networks exhibit small over-represented subgraphs, called motifs, some of which are known to have a biological function. Several algorithms exist to detect motifs, most of them being based on time-consuming simulations or leading to many false positives. We propose an efficient and conservative procedure to detect network motifs and apply it on the Yeast gene regulation network.

Abstract : Naturally occurring mutations in mammalian genomes play a key role in evolution and genetic disease but their causes are still poorly understood. In particular, nucleotide substitutions occur at strongly variable rates along genomes and it is essential to unravel the mechanisms responsible of these fluctuations. A number of evolutionary studies have exhibited complex correlations between substitution rates and parameters like regional or local nucleotide composition, crossover rate or distance to telomeres (1-6). Here, we study the role of replication on neutral substitution rates in the human genome. Using replication timing data determined by massive sequencing of replicating strands, we show that all non-CpG substitution rates correlate with timing : they are minimum in early replicating regions and increase to maximum values in late regions. These correlations are still observed after controlling for nucleotide composition, cross-over rate and distance to telomeres. These data demonstrate for the first time that replication timing plays a key role in shaping the profile of mutations along the genome.

Anne Crumiere. Cellular automata modeling of intercellular genetic regulatory networks

Abstract : Biologists often represent genetic interactions by directed graphs, named genetic regulatory graphs. Vertices represent genes, whereas edges represent regulatory effects from one gene on another. Edges are labelled with a positive sign in the case of an activation and negative for an inhibition. This article deals with relationships between the structure of such graphs and their dynamical properties.

The biologist R.Thomas enounced, thirty years ago, the following two general rules : a necessary condition for multistability is the presence of a positive circuit in the regulatory graph (the sign of a circuit being the product of the signs of its edges) and the existence of a negative circuit is a necessary condition for the existence of an attractive cycle. These rules are about the dynamic of a single cell, and it has given rise to mathematical statements and proofs. This article aims at extending these rules to regulatory interactions spanning within cells and between cells in the discrete formalism.

Hugo Devillers, Hélène Chiapello, Meriem El Karoui and Sophie Schbath. How to measure the robustness of bacterial genome comparisons ?

Lionel Dupuy, Matthieu Vignes, Blair McKenzie and Philip White. Meristematic Waves, a new approach to model root architecture dynamics

Abstract : During their development, plants must develop efficient root architectures to secure access to nutrients and water in soil. A series of expansion and branching mechanisms fulfils this aim in the proximity of root apical meristems where the plant senses the environment and explores immediate regions of soil. We have developed a new approach to study the dynamics of root meristems in soil, using the relationship between the increase in root length density and the root meristem density. Initiated at the seed, the location of root meristems was shown to propagate, wave-like, through the soil, leaving behind a permanent network of roots for the plant to acquire water and nutrients. Models higlighted that the morphologies of the waves of meristems are inherent to individual root developmental processes, namely expansion, lateral root initiation and gravitropic responses. The "meristematic wave" observed on data collected on barley might be a more general and fundamental aspect of plant rooting strategies to access underground resources.

Paul Garcin and Yves Boulard. Construction et analyse d'un modèle tridimensionnel du complexe [(SLR1738-Zn-Fe)2-ADN]

Abstract : Slr1738 is the Peroxide regulon Repressor protein (PerR) of Synechocystis. Active as a dimer, this protein must contain an iron atom to be able to bind DNA molecule and regulates targeted genes. The binding mechanism involves a classic recognition helix inserted in the DNA major groove. But to date there is no three-dimensional structure available for this kind of transcription factor complexed to DNA. As a consequence, both global and specific interactions that lead this protein to bind DNA and to recognize specific ‘Per Box' sequence are still misunderstood. In order to better define and analyse these interactions, we built in silico the first three-dimensional structure of a [PerR-DNA] complex. This article describes the method used to build the complex and presents an analysis of the contacts between the two partners.

Sandrine Grossetete, Bernard Labedan and Olivier Lespinet. FUNGIpath : a new tool for analysing the evolution of fungal metabolic pathways

Abstract : FUNGIpath is a new tool dedicated to perform in-depth analysis of fungal metabolic pathways. It is freely accessible at http://www.fungipath.u-psud.fr. FUNGIpath consists in a collection of orthologous groups of proteins that have been predicted using complementary methods of detection and further mapped on KEGG and MetaCyc pathways. It allows an easy comparison of the primary and secondary metabolisms afforded by the different fungal species present in the database with the possibility to assess the level of specificity of various pathways at different taxonomic distances. As more and more fungal genomes are expected to be decrypted in the next years, this tool is expected to help to progressively reconstruct what were the primary and secondary metabolisms of the ancestors of the main branches of the fungi tree and to understand how these ancestral fungal metabolisms evolved to various specific derived metabolisms.

Abstract : Nowadays, bioinformatics tasks typically involve large scale data analysis that require the integration of web services from heterogeneous platforms. In spite of the efforts for improving Web services interoperability, integration remains difficult and still has to be performed manually by users. Improving the composition of Web services requires to analyze what Web services do as well as the nature and the type of their input and output parameters. This work shows that existing technologies support automating the selection, composition and execution of Web services, and that the current limiting factor to a wider use is the lack of precise enough task and domain ontologies.

Aurelie Leduc, Stephane Robin, Philippe Bessieres and Pierre Nicolas. Probabilistic modeling of tiling array expression data

Abstract : For organisms with small genomes such as bacteria, the current microarray technology allows adopting a tiling design where the whole genome is covered by overlapping probes. These arrays permit to measure the transcriptional activity of the whole genome with unprecedented resolution. Model-based approaches currently used to analyze these data remain however very simple, the most popular model being the piecewise constant Gaussian model with a fixed number of breakpoints. Here we present a new approach based on hidden Markov modelling designed for the probabilistic reconstruction the trajectory of a continuous-valued signal. The use of this model does not require the choice of a fixed number of breakpoints and permits to account for subtle effects such as drift in the signal. The model also includes direct correction for the variations of probe affinities via the use of covariates.

Celine Lefebvre, Mariano Alvarez, Presha Rajbhandari, Wei Keat Lim and Andrea Califano. Master regulator analysis reveals key transcription factors for Germinal Center formation

Abstract : We describe a new method for the identification of master regulators of a phenotype of interest. The master regulator analysis identifies transcription factors that are candidate master regulators of a phenotype of interest based on its transcriptional targets. We applied this method for deciphering the regulation of Germinal Center B cell programs, revealing the two transcription factors MYB and FOXM1 as synergistic master regulators

Sébastien Loriot, Frédéric Cazals, Michael Levitt and Julie Bernauer. A geometric knowledge-based coarse-grained scoring potential for structure prediction evaluation

Abstract : Knowledge-based protein folding potentials have proven successful in the recent years. Based on statistics of observed interatomic distances, they generally encode pairwise contact information. In this study we present a method that derives multibody contact potentials from measurements of surface areas using coarse-grained protein models. We show that this construction is able to distinguish native structures from decoys. We tested different potentials from a reference set of 66 protein structures. These functions, encoding up to 5-body contacts are evaluated on the reference set and its 45000 decoys and also on the often used lattice_ssfit set from the decoys'R us database. We show that the most relevant information for discrimination resides in 2- and 3-body contacts. The potentials we have obtained can be used for evaluation of putative structural models ; they could also lead to different type of protein structure refinement that uses multi-body interactions

Christine Martin and Antoine Cornuéjols. Using Frequent and Surprising Item Sets for the Characterization of Protein-Protein Interfaces

Abstract : Numerous research effort have aimed to characterize and predict protein-protein interfaces. This paper introduces a method that rely only on known protein-protein interfaces (positive instances only). It combines frequent item set mining techniques with statistical tests to ensure the selection of interesting features. Starting from a database of known interfaces described with geometrical elements, the method produces the elements and combinations thereof that are \emphcharacteristic of the interfaces. This approach allows one to easily interpret the results, as compared to techniques that operate as ``black-boxes'' and ensures a satisfactory proportion of reliable item sets. The results obtained on a set of 459 protein-protein interfaces from the DOCKGROUND database confirm that the findings are consistent with current knowledge about protein-protein interfaces.

Marie-José Mhawej and Claude H. Moog. Drug dosage control of the HIV infection dynamics

Abstract : An increasing number of sophisticated control algorithms become available in the current literature to optimize the HIV therapy. Unfortunately, the pharmacokinetics and pharmacodynamics of antiretroviral drugs are ignored and these algorithms remain purely theoretic. This issue is investigated explicitly in this paper. An elementary pharmacodynamics model is combined with a non linear feedback control computed from standard engineering methods. It is shown that it results in the design of a realistic dosage regimen which drives the immunological system close to the healthy equilibrium state. Although the problem is dealt as a single input system, it is argued that the procedure can be extended to a multitherapy design or to any available control law.

Gregory Nuel. Counting patterns in degenerated sequences

Abstract : In this paper, we propose a rigourous method to take into account the uncertainty of sequencing for biological sequences (DNA, Proteins). For example, this method allows to study the distribution of a pattern of interest in a degenerated sequence defined on the standard IUPAC DNA alphabet. We first introduce a Forward-Backward approach to compute the marginal distribution of the constrained sequence and use it both to perform a Expectation-Maximization estimation of parameters, as well as deriving a heterogeneous Markov distribution for the constrained sequence. This distribution is hence used along with known DFA-based pattern approaches to obtain the exact distribution of the pattern count under the constraints. As an illustration, we consider a EST dataset from the EMBL database. Despite the fact that only 1% of the position in this dataset are degenerated, we show that not taking into account these positions might lead to erroneous observations, further proving the interest of our approach.

Nicolas Terrapon, Olivier Gascuel and Laurent Brehelin. Détection de nouveaux domaines protéiques par co-occurrence : Application à P. falciparum

Abstract : Hidden Markov Models (HMMs) have proved to be powerful for protein domain identification. However, numerous domains may be missed in highly divergent proteins. This is the case for the proteins of Plasmodium falciparum, the main causal agent of human malaria. Here, we propose a method that uses domain co-occurrence to increase the sensitivity of the approach while controlling its false discovery rate. Applied to P. falciparum, our method identify (with an error rate below 20%) 482 new domains (versus 3482 in PlasmoDB), which involve 158 new GO annotations.

Raluca Uricaru, Celia Michotey, Laurent Noe, Hélène Chiapello and Eric Rivals. Improved sensitivity and reliability of anchor based genome alignment

Abstract : Whole genome alignment is a challenging problem in computational comparative genomics. It is essential for the functional annotation of genomes, the understanding of their evolution, and for phylogenomics. Many global alignment programs are heuristic variations on the anchor based strategy, which relies on the initial detection of similarities and their selection in an ordered chain. Considering that alignment tools fail to align some pairs of bacterial strains, we investigate whether this is intrinsically due to the strategy or to a lack of sensitivity of the similarity detection method. For this, we implement and compare $6$ programs based on three different detection methods (from exact matches to local alignments) on a large benchmark set. Our results suggest that the sensitivity of well known methods, like MGA or Mauve, can be greatly improved in the case of divergent genomes if one exploits spaced seeds at the detection phase. In other cases, such methods yield alignments that covers nearly the whole genome. Then, we focus on global reliability of alignments : should an aligned pair of segments be included in the global genome alignment ? We investigate this reliability according to both the segment "alignability" and to inclusion of orthologs. Again, we provide evidence that for both close and divergent genomes, one of our program, YH, achieves alignments with sometimes a lower coverage, but a higher inclusion of orthologs. It opens the way to the first reliable alignments for some highly divergent species like Buchnera aphidicola or Prochlorococcus marinus.

Detailed program

2009-03-25T14:09:25Z

**Tuesday june 9 2009**
9h00-9h45	Registration opens, Welcome coffee
9h45-10h15	Welcoming speech
10h15-11h30	Selection and randomness in evolution or selection of randomness in evolution ? Christian Gautier
11h30-12h00	Single-nucleotide substitution rates increase during the replication S phase of the human genome. Chun-Long Chen, Aurélien Rappailles, Lauranne Duquenne, Maxime Huvet, Guillaume Guilbaud, Benjamin Audit, Yves d'Aubenton-Carafa, Alain Arneodo, Olivier Hyrien and Claude Thermes.
12h-12h30	Flash presentations (5 min) *
12h30-14h30	Lunch
14h30-15h30	Computational Short Read Metagenomics Jens Stoye
15h30-16h00	Counting patterns in degenerated sequences. Gregory Nuel.
16h00-17h00	Posters and coffee break
17h00-17h30	How to measure the robustness of bacterial genome comparisons ? Hugo Devillers, Hélène Chiapello, Meriem El Karoui and Sophie Schbath.
17h30-18h00	Improved sensitivity and reliability of anchor based genome alignment. Raluca Uricaru, Celia Michotey, Laurent Noe, Hélène Chiapello and Eric Rivals.
18h00-18h30	Drug dosage control of the HIV infection dynamic. Marie-José MHAWEJ and Claude H. Moog. S
19h00	Welcoming cocktail at the city hall of Nantes

* Flashs :

EuGène Maize : A gene prediction web tools for maize. Pierre Montalent and Johann Joets.
Intégration automatique d'une ontologie de domaine dans un annuaire Biomoby. Julien Wollbrett, Pierre Larmande and Manuel Ruiz.
Estimation of sequence errors and capacity of genomic annotation in transcriptomic and DNA-protein interaction assays based on next generation sequencers. Nicolas Philippe, Jorma Tarhio, Anthony Boureux, Laurent Bréhélin, Thérèse Commes and Eric Rivals.
Oenococcus oeni genome plasticity associated with adaptation to wine, an extreme ecological niche. Elisabeth Bon, Arnaud Delaherche, Eric Bilhere, Antoine de Daruvar, Aline Lonvaud-Funel and Claire Le Marrec.
Databases of homologous gene families for comparative genomics. Simon Penel, Anne-Muriel Arigon, Vincent Daubin, Pascal Calvat, Stephane Delmotte, Jean-Francois Dufayard, Manolo Gouy, Guy Perriere, Anne-Sophie Sertier and Laurent Duret.

**Wednesday june 10, 2009**
9h00-10h00	What directs a transcription factor to its target sites ? New insights from ChIP-Seq data analysis. Philipp Bucher
10h00-10h30	Detection de nouveaux domaines proteiques par co-occurrence : application a p. falciparum Nicolas Terrapon, Olivier Gascuel and Laurent Brehelin.
10h30-11h00	Posters and coffee break
11h00-11h30	Système de Classes Chevauchantes pour la Recherche de Protéines Multifonctionnelles. Emmanuelle Becker, Alain Guénoche and Christine Brun.
11h30-12h00	Utilisation d'ontologies de tâches et de domaine pour la composition semi-automatique de services Web bioinformatiques. Nicolas Lebreton, Christophe Blanchet, Julie Chabalier and Olivier Dameron.
12h-12h30	Flash presentations (5 min) *
12h30-14h30	Lunch
14h30-15h30	Le recodage : Qu'est-ce que c'est ? A quoi ça sert ? Comment ça marche ? Et la bioinformatique dans tout ça ? Jean-Pierre Rousset
15h30-16h00	Probabilistic modeling of tiling array expression data. Aurelie LEDUC, Pierre Nicolas, Philippe Bessieres and Stephane Robin.
16h00-17h00	Posters and coffee break
17h00-17h30	Master regulator analysis reveals key transcription factors for Germinal Center formation. Celine Lefebvre, Mariano Alvarez, Presha Rajbhandari, Wei Keat Lim and Andrea Califano.
17h30-18h00	Cellular automata modeling of intercellular genetic regulatory networks. Anne Crumiere.
18h00-18h30	Using Frequent and Surprising Item Sets for the Characterization of Protein-Protein Interfaces. Christine Martin and Antoine Cornuéjols.
18h30-19h30	General Assembly of the French Society of Bioinformatics (SFBI)
19h30	Guided tour of « l'île de Nantes »
20h30	Gala diner on « l'île de Nantes »

* Flashs :

ace.map – a comprehensive tool for advanced microarray analysis. Guillaume Brysbaert, Brice Targat, Nicolas Tchitchek, Jose Felipe Golib Dzib, Christophe Bécavin, Sebastian Noth and Arndt Benecke.
Crossing genome and transcriptome: deciphering links between structure and function in Arabidopsis thaliana genes. Véronique Brunaud, Virginie Bernard, David Armisén, Jean-Philippe Tamby, Séverine Gagnot, Sandra Derozier, Franck Samson, Cécile Guichard, Marie-Laure Martin-Magniette, Alain Lecharny and Sebastien Aubourg.
Generalized Peptide Mass Fingerprinting on whole-cell HPLC-MS proteomics experiments. Pascal F. Bochet, Frank Rügheimher, Guina Tina, David R. Goodlett, Peter Clote and Benno Schwikowski.
Multiple perturbation mapping of biological systems. Magali Michaut and Gary Bader.
Dynamic modelisation of transcriptional regulatory networks involved in yeast antifongal resistance. Jennifer Becq, Sophie Lèbre, Frédéric Devaux and Gaëlle Lelandais.

Thursday june 11, 2009
9h00-10h00	Tackling regulatory networks : from biological models to theorems, and vice-versa. Denis Thieffry
10h00-10h30	FUNGIpath: a new tool for analysing the evolution of fungal metabolic pathways. Sandrine Grossetete, Bernard Labedan and Olivier Lespinet.
10h30-11h00	Posters and coffee break
11h00-11h30	Detecting Network Motifs by Local Concentration. Etienne Birmele.
11h30-12h00	Meristematic Waves, a new approach to model root architecture dynamics. Lionel Dupuy, Matthieu Vignes, Blair McKenzie and Philip White.
12h-14h00	Lunch
14h00-15h00	Méthodes d'analyse de séquences et de structures 3D de protéines et leur intégration au sein de Webiciels. Gilbert Deleage
15h00-15h30	Construction et analyse d'un modèle tridimensionnel du complexe (SLR1738-Zn-Fe)2-ADN. Paul Garcin and Yves Boulard.
15h30-16h00	A geometric knowledge-based coarse-grained scoring potential for structure prediction evaluation. Sébastien Loriot, Frédéric Cazals, Michael Levitt and Julie Bernauer.
16h00-16h30	Annual Review of GDR BIM
16h30-16h45	Closing remarks