International Symposium on Integrative Bioinformatics 2010: Poster Abstracts

International Symposium on
Integrative Bioinformatics
6th annual meeting
22^nd to 24^th March 2010
Cambridge, United Kingdom

Abstracts of Accepted Posters

	Poster abstracts
	Abstracts are posted after acceptance.

	Abstract No. 258
	Offord, VA; Coffey, TJ; Werling, D
	Identification of Leucine-rich repeats using LRR-Finder
	Leucine-rich repeats (LRRs) characterise a diverse family of regulatory proteins. The expanding gap between available LRR crystal structures and number of sequences of unknown structure has increased the use of homology modelling in resolving 3D structures. However, the accuracy of each prediction relies heavily upon template sequences and the identification of their structural features. Known LRR structures reveal a highly conserved region and assemble into the curved parallel β-sheet, lining the inner circumference of their distinctive solenoid structure. Thus, prediction of these structurally important regions is essential for modelling LRR proteins and their interactions. Over 360 Toll-like receptor (TLR) sequences obtained from 106 species, spanning 51 families, form the present TLR database (tLRRdb). Each individual sequence has been manually separated into its constituent regions to form a catalogue of over 4000 distinct, naturally occurring LRRs. This comprehensive collection of TLR-LRRs records the classified 11 amino acid highly conserved regions from which the Position-Specific Scoring Matrix is derived. The inclusion of standalone tools (SignalP 3.0, TMHMM 2.0, BLAST, ClustalW) provides the ability to not only predict LRRs but also to identify common domains, perform similarity searches and align sequences. Derived from this is LRRfinder (www.lrrfinder.com), a user-friendly web application for the identification of LRRs within user defined sequences and prediction of signal peptides, N- and C-terminus, TIR-domains and transmembrane helices. Our tool facilitates the identification of structurally important regions within LRR proteins to provide accurate information for homology modelling templates, particularly relevant for protein-protein interaction studies and classification of novel sequences.

	Abstract No. 259
	Edgar Antonio Reyes-Monta?o, Reyes-M, EA
	HOMOLOGY MODELLING OF Salvia miltiorrhiza LECTIN
	Salvia miltiorrhiza is a well-known traditional Chinese herb and broadly planted in China. Its roots (called Danshen in Chinese) contain two groups of biologically active compounds, caffeicacid-derived phenolic acids and various tanshinones belonging to the group of diterpene quinines. It is one of the most popular traditional herbal medicines in some Asian countries and has been used clinically for the treatment of various ailments such as cardiovascular, cerebrovascular, hyperlipidemia, and acute ischemic stroke diseases. Pharmacologic studies revealed its secondary metabolites with various biological activities and protective effects against cerebral and heart ischemia-reperfusion, inhibitory activity against hepatic fibrosis and hepatoprotection, and antioxidant activity, antithrombosis activity, antihypertension activity, antivirus activity, antitumor activity and anti-ulcer activity. S. miltiorrhiza lectin (SML) has been not strongly studied before. Some authors have shown that SML inhibits Neutrophil-Endothelial adhesion (Adamo, P., Verisimilitude and Malignancy, available in: http://www.northamericanpharmacal.com/Verisimilitude_Cancer.pdf). SML is actually available in NCBI with the accession number ABU87404, but there are not published papers about that. Previous reports indicate that SML has a recognition pocket similar to legume lectins. The tertiary model of SML was constructed using some bioinformatic tools as Clustal, Swiss-Model, HHPred, Modeller, Anolea, Hex 5.1 and others. The model was evaluated (Procheck) and a N-glycosylation was added. The final model shows a high interaction with mannosides and Tn antigen mainly.

	Abstract No. 275
	Shilov, BV; Goremykin, CV; Ivlev, IV; Ryzhov, SV; Koroleva, YuA; Shabanova, EP; Serebrov, VYu; Sazonov, AE
	ADENOSINE A2B RECEPTOR INTERACTION NVESTIGATION IN SILICO
	This research was devoted to study of adenosine receptors A2B subtype, which is not full studied. The research?s aim was to create computer models of A2B adenosine receptor, and also its 9 activators and 15 inhibitors; using these models to analyze special features of interaction between A2B adenosine receptor molecule and molecules of different activators and inhibitors in silico. The amino acid sequence of A2B receptors was taken from open source in the NCBI data base, ID: NP_000667.1, for modeling A2B adenosine receptor and its mutant forms the MODELLER software was used. The modeling of ligand-receptor interaction was made on the supercomputer SKIF Cyberia (Tomsk State University, Russia) using AutoDock 4.0 as the part of MGL Tools 1.4.5. In this research work, realized on the unique A2B adenosine receptor computer model, it was shown, that both activators and inhibitors produce the hydrogen bindings with different amino acids as a result of interaction between them and A2B adenosine receptor and its mutant forms. The current investigation detected the amino acids, playing the main part in activation A2B adenosine receptor: Ile67, Phe173 and Glu174. Also, it was shown, that applying amino acid substitutions to the third intracellular loop potentially may cause the alteration of Ile67 and receptor adhesive site binding. To make final decision about the role of aforesaid acids in receptor activation it is needed to implement the test with A2B receptors, mutant by mentioned amino acids, in vitro.

	Abstract No. 301 View the poster
	Chuckravanen, D; Rajbhandari, S; Angelova, M; St Clair Gibson, A; Ansley, L; Thompson, KG
	Continuous Wavelet Analysis of Physiological Data for various Pacing Strategies of a 20-km Cycling Time Trial
	Abstract: Recently, in exercise physiology, a novel model to regulate the central neural effort and fatigue has been proposed and named the Central Governor Model. This model theorizes that physical activity is controlled by a central regulator in the brain, and the human body works as a complex integral system, unlike the Cardiovascular/Anaerobic/Catastrophe model of Sir A.V. Hill of exercise physiology. In this study, physiological data were collected from club level cyclists for different pacing strategies that were self paced, even paced, and variable paced for a 20km cycling time trial in order to assess the underlying system control mechanisms that show how the brain paces the human body during exercise. Continuous Wavelet Transform (CWT) was used to analyse the non-stationary and nonlinear physiological signals that were heart rate (HR/bpm) and volume of oxygen consumption (VO2/ Lmin-1). Normalised mean wavelet powers were used to compare the powers at different frequency bands of the continuous wavelet spectrum. These frequency bands were classified as High Frequency (HF), Low Frequency (LF) and Ultra Low Frequency (ULF) bands. There was a significant difference in the ULF band for the volume of oxygen consumption (p<0.01) that decreased with increasing performance times of cyclists for all pacing strategies. As for the heart rate activities, both ULF and LF band powers were practically constant for all cyclists, and there was a significant difference in the HF band power compared to the other frequency bands. It was shown that the brain paces the human body by acting as an external drive to that particular peripheral system and it uses specific frequency bands to control and communicate with a particular peripheral system in the aims to reach the end of that physical task without homeostasis failure.

	Abstract No. 308
	Spooner, W; Stein, JC; Wei, S; Ware, D
	Gramene GeneTrees: A comprehensive phylogenomics database in plants and other model Eukaryotes
	Since the completion of the Arabidopsis thaliana genome in 2000, more than 20 plant genomes have been sequenced, with the number set to increase rapidly in the coming years. It is now becoming possible to apply phylogenomics-based techniques from metazoan biology to the plant arena. We have applied an automated methodology, EnsemblCompara GeneTrees, to proteins predicted from a wide range of genomes to develop the first comprehensive plant phylogenomics resource. This consists of protein-level phylogenetic trees between twelve whole genomes; four dicotyledon plants (Arabidopsis lyrata, Arabidopsis thaliana, Populus trichocarpa, Vitis vinifera), three monocotyledon plants (Oyrza sativa Japonica Group, Oryza sativa Indica Group, Sorghum bicolor), and four model metazoa/fungi (Caenorhabditis elegans, Drosophila melanogaster, Homo sapiens, Saccharomyces cerevisiae). Validation of our data through comparison data with similar resources and well-studied gene families suggests accurate and consistent results both for summaries of the database as a whole and for individual example trees. The GeneTrees form a component of the Gramene database (http://www.gramene.org), an established resource for comparative plant genomics, and form a useful platform for the study of plant molecular evolution as well as functional annotation of newly sequenced genomes.

	Abstract No. 311 View the poster
	Anwar, N; Bader, G; Demir, Emek; Rodchenkov, I; Sander, C
	Using BioPAX Level 3 for Exchange and Integration of Pathway Data.
	The BioPAX ontology (www.biopax.org) is a standard language for formally representing biological pathways and is available to the biological community to enable exchange and integration of pathway data. Data exchange and integration continues to be a challenge given the complex nature of both pathway data and data sources. Biological pathways are constructs that biologists use to represent relationships between and within chains of cellular events. Metabolic pathways typically represent flow of chemical reactions, while signal transduction pathways represent the chain of molecules that are used to transmit an external signal received by a cell to deliver the response within the cell. The data is as heterogeneous as its numerous sources (pathguide.org). BioPAX was developed to address these issues and to ease the access, use, exchange and aggregation of pathway data. The BioPAX pathway ontology is defined using the Web Ontology Language, OWL, with the view of facilitating automatic processing and the integration of information held in biological pathways. BioPAX has been a community effort spanning 7 years, culminating in the recent release of BioPAX level 3. Level 3 supports the representation of metabolic pathways, signal transduction pathways, protein-protein interaction networks, gene regulatory networks and genetic interactions. We will outline data representation in BioPAX and the use of the BioPAX ontology in integration of pathway data, thus enabling more efficient use and reuse of these data. We wish to highlight the successes of this community project, the core entities within the BioPAX ontology that have changed from Level 2, demonstrate example SPARQL queries across heterogeneous pathway related data and finally, outline a successful use case of data integration using OWL within the PathwayCommons knowledge base.

	Abstract No. 312 View the poster
	Ondex Consortium, The
	Integrating life sciences data sources using Ondex
	The aim of this project, funded under the UK BBSRC Systems Biology initiative, is to develop an extensible data integration system for supporting systems biology research. The project is extending the Ondex data integration platform (http://ondex.org) and supporting four demonstrator projects at three systems biology research centres. The demonstrator projects represent a range of general data integration problems in systems biology and validate new features through interaction with end-users. The three core demonstrator projects identify new genetic and molecular targets to improve bioenergy crops (Rothamsted ); integrate, and validate different yeast metabolome models (Manchester, MCISB ) and support studies of telomere function relating to ageing research in yeast (Newcastle, CISBAN ). A fourth demonstrator project at the CSBE (Edinburgh ) studies signal transduction processes controlling circadian rhythms in Arabidopsis thaliana. This project increases the range of system biology projects that can be supported by: ? Extending the core data structures, interfaces and data integration framework to support probabilistic relationships ? Adapting NaCTeM text mining tools to use semantic parsing to extract more complex relationships and semantics from bio-text sources for the core demonstrator projects ? Upgrading the workflow management to incorporate new developments from myGrid and support long-running and asynchronous workflows ? Developing the user interfaces and other components to support comparative analyses (e.g. for comparing pathways, gene orders, etc from different species) ? Exploiting the new data structures with statistical data analysis methods and associated visualisation methods

	Abstract No. 314
	Dupont, P-Y; Loiseau, D; Morvan, D; Demidem, A; Stepien, G
	Study of a bioenergetic regulatory network involved in cancer cell response to chemotherapy
	Our study links the constitutive glycolytic activity and the gene regulation in transformed cells. One of these genes, ANT2, should allow cells to keep their mitochondrial integrity through maintenance of internal membrane potential gradient (ΔΨm). In this study, we correlated biological results to a specific gene regulatory network adjusting cell requirements. We compared the response of two transformed cell lines to an anticancer agent: HepG2 (hepatocarcinoma) with a partially differentiated phenotype and 143B (osteosarcoma) with an undifferentiated one. Treatment effects were tested on global metabolite profiling by NMR spectroscopy. Over and under-expressed enzymes involved in disrupted metabolic pathways were deduced from NMR metabolite profiles. We developed an informatics pipeline to analyze the mechanisms of transcriptional regulation of genes encoding for selected enzymes. Our biological results showed an increased energy request to regenerate ΔΨm in both models. This could not be met by undifferentiated cells (143B), which ATP content decreased leading them to death, while partially differentiated cells (HepG2) could activate their oxidative metabolism and escape chemotherapy. Our bioinformatics study allowed us to: 1 ? construct specific sets of regulatory sequences (modules) in gene promoters; 2 - scan the whole human genome for these regulatory modules; 3 - identify human genes including such modules in their promoter sequence. We propose that mitochondrial OXPHOS background confers a survival advantage to more differentiated cells in response to chemotherapy. The set of genes either selected from the NMR analysis or deduced from bioinformatics analysis was used to construct a regulatory network involved during cancer treatment.

	Abstract No. 315
	Walther, Dirk; Strassburg, Katrin; Kopka, Joachim
	Granger causality testing identifies potential cause-effect relationships between metabolites and transcripts.
	The integrated analysis of omics datasets covering different levels of molecular organization has become a central task of systems biology. We investigated the transcriptional and metabolic response of yeast exposed to increased (37?C) and lowered (10?C) temperatures relative to optimal reference conditions (28?C) within the framework of known metabolic pathways. Pairwise metabolite correlation levels were found to carry more pathway-related information and to extend to farther distances within the metabolic pathway network than associated transcript level correlations. Metabolites were detected to correlate stronger to their cognate transcripts (metabolite is reactant of the enzyme encoded by the transcript) than to more remote or randomly chosen transcripts reflecting their close metabolic relationship. We observed a pronounced temporal hierarchy between metabolic and transcriptional molecular responses under heat and cold stress. Changes of metabolites were most significantly correlated to transcripts encoding metabolic enzymes acting on them, when metabolites were considered leading in time-lagged correlation analyses. By applying the concept of Granger causality, we detected directed relationships between metabolites and their cognate transcripts. When interpreted as substrate-to-product directions, most of these directed Granger causality pairs were found to be in agreement with the KEGG-annotated preferred reaction direction. Thus, the introduced Granger causality approach may prove useful for detecting the preferred direction of metabolic reactions in living systems. The metabolites glutamic acid and serine were identified as central causative metabolites influencing transcript levels at later time points. Selected examples are presented illustrating the intertwined relationships between metabolites and transcripts in the yeast temperature stress adaptation process.

	Abstract No. 316 View the poster
	Schoenherr, SS; Wei?ensteiner, HW; Brandstaetter, AB; Coassin, SC; Specht, GS; Kronenberg, FK
	eCOMPAGT ? efficient Combination and Management of Phenotypes and Genotypes for Genetic Epidemiology
	High-throughput genotyping and phenotyping projects of large epidemiological study populations require sophisticated laboratory information management systems (LIMS). Especially an easy combination of genotypes and phenotypes can be achieved, if the complete set of project-relevant information is stored in a centralized database. Since most genetic studies include also subject-related personal information, this information needs to be handled with care by following data privacy protection guidelines. Additionally, the recent past has uncovered severe problems with mtDNA genotyping, not only due to the genotyping method itself, but mainly due to the post-lab transcription, storage and reporting of mtDNA genotypes. We developed a freely available, database based Java application (eCOMPAGT) for an efficient combination and management of phenotypes and genotypes deriving from genetic epidemiological studies. eCOMPAGT is a system to store, administer and connect phenotype data to all kinds of genotype data (SNPs, STRs, and mtDNA profiles). A versioning component for traceability, a widespread project & customer management and an export interface to statistical software packages are further important features of our software. eCOMPAGT is suited for small to medium-sized human genetic, forensic and clinical genetic laboratories. The direct support of Oracle and the Open-Source database MySQL with a developed two layer security system renders eCOMPAGT to a powerful tool for building automated workflow architectures for project management, laboratory and evaluation processes all-in-one. eCOMPAGT, a user manual and a demonstration video are freely available at http://dbis-informatik.uibk.ac.at/ecompagt.

	Abstract No. 320 View the poster
	Al-Haj Ibrahim, M; Jassim, S; Cawthorne, MA; Langlands, K
	A Topology-Based Pathway Regulation Score
	Increasingly, microarray-based transcriptional profiling techniques offer biologists insight into complex pathophysiological processes. However, reducing this complexity to yield testable hypotheses remains difficult. Simple fold change analysis and ontology enrichment may yield potential targets for further study, but pathway enrichment analysis, such as that facilitated by mining the KEGG database, comes closer to providing true biologically-meaningful insights. However, once a differentially-expressed gene set is identified, existing approaches tend not to exploit fold-change information. Moreover, pathway organisation does not inform the scoring systems used. We have implemented a novel approach that addresses these limitations by calculating a pathway regulation score (PRS). The calculation of PRS takes into consideration the number of genes whose expression exceeds the fold change threshold, the magnitude of these fold changes and positions of genes within a given KEGG pathway. Scores are evaluated using a phenotype permutation test and adjusted for false discovery rate. For a given microarray dataset, pathways are ranked by PRS providing enrichment for biologically relevant processes (assessed by literature searching) compared to methods that exploit hypergeometric distributions. This new approach illustrates the important role of topologies in measuring the relative contribution of a pathway to a biological process.

	Abstract No. 322 View the poster
	Czauderna, T; Junker, A; Schreiber, F
	Editing, translating and validating SBGN maps: RIMAS - a first application
	The Systems Biology Graphical Notation (SBGN) was developed representing an agreed-upon convention for the display and handling of biological networks. It aims to support more efficient and accurate communication of biological knowledge between different research communities in the life sciences. SBGN-ED was developed as a VANTED Add-on to provide a tool for creating and editing all three types of SBGN maps such as Process Description Maps, Entity Relationship Maps and Activity Flow Maps. Furthermore the syntactical and semantical correctness of created or edited maps can be validated by SBGN-ED. Already existing non-SBGN diagrams from the KEGG and MetaCrop databases can be translated automatically into SBGN. As a first application of SBGN, the RIMAS (Regulatory Interaction Maps of Arabidopsis Seed Development) web-based information portal provides a comprehensive regularly updated overview of regulatory pathways and genetic interactions during Arabidopsis embryo and seed development. Regulatory interaction maps have been drawn and validated using SBGN-ED and exported via new VANTED export functions for websites with clickable image-maps. The RIMAS service provides access to SBGN network maps, linked literature databases and possibilities to export diagrams in common exchange formats such as GML for modifying network layouts according to individual purposes and further application using tools such as VANTED or other image formats.

	Abstract No. 324
	Weile, J; Pocock, M; Lord, P; Wilkinson, D; Dewar, J; Holstein, E; Cockell, SJ; Hallinan, J; Wipat, A
	Exploring telomere function in Saccharomyces cerevisiae using customisable views in Ondex
	Telomeres play a major role in the process of ageing. The budding yeast Saccharomyces cerevisiae is an ideal organism for studying telomere function. Many different aspects of cellular biology contribute to the complexity of telomere-associated processes. Thus an integrative systems approach to the study of telomere function promises to be highly beneficial to our understanding of these complex processes. We are applying this approach using the graph-based data integration system Ondex. To facilitate the exploration of integrated datasets such as ours, we have developed a concise visualisation technique which applies customisable views. Using this method, we have generated hypotheses that may explain known associations between telomere capping and mitochondrial DNA maintenance, the NMD pathway and cell division control.

	Abstract No. 325
	Witney, AA; Tyler, R; Waldron, D; Hinds, J
	B?G@Sbase ? a microarray database and analysis tool
	The Bacterial Microarray group at St George's, University of London (B?G@S) design, generate and support whole genome microarrays for eleven bacterial genomes. These microarrays are made available through collaborations with research groups within the UK and the rest of the world. B?G@Sbase is a relational database system developed at B?G@S to store and track all information relating to the array process. Collaborating users are able to obtain information about the arrays they have been provided, including PCR products/oligonucleotides, and all arraying details. The system enables users to capture the data required for MIAME compliance and incorporates the Microarray Ontology developed within the MGED group. Data can then be exported in MAGE-ML format to the Microarray data repository ArrayExpress (EBI), and a pipeline has been established between B?G@Sbase and ArrayExpress to allow the easy flow of data. The system has built-in security such that each user has their own personal space within the database to store their experimental data. However, read and write permissions can then be given to other users and groups of users. A web interface has been developed to allow multi-platform access from external collaborators. Currently the database contains data from 170 experiments (60 publicly accessible), consisting of approximately 5000 data files for 15 different species of bacteria; thus providing a uniform platform for powerful cross species genetic analysis. Therefore, a framework has been established that allows R statistical analysis tools to be accessed by users; including R packages such as LIMMA and BioConductor. Also phylogenetic analyses will be available for comparative genomic data using the PHYLIP package. An intuitive user interface has been designed to allow drag-and-drop building of analysis pipelines using the javascript application development library ExtJS.

	Abstract No. 326 View the poster
	Colmsee, C; Flemming, S; Klapperstueck, M; Lange, M; Scholz, U
	Efficient management of high throughput primary lab data
	In modern life science research it is very important to have an efficient management of high throughput primary lab data. To realise such an efficient management four different main aspects have to be handled: (I) storage, (II) security, (III) upload and (IV) retrieval. In this poster we define central requirements for a primary lab data management and discuss aspects of best solutions to realise these requirements. Furthermore we introduce a pipeline that have been implemented to store terabytes of primary lab data in the Leibniz Institute of Plant Genetics and Crop Plant Research (IPK). It comprises: (I) a data storage implementation including a Hierarchical Storage Management system, a relational Database Management System and the BFiler package to store primary lab data and their meta information, (II) the Virtual Private Database implementation for the realisation of data security and the LIMS Light application to (III) upload and (IV) retrieve stored primary lab data.

	Abstract No. 327 View the poster
	Jimenez, R; Reisinger, F; Hermjakob, H
	EnCore, a data integration platform
	EnCore is the integration platform for the ENFIN European Network of Excellence. It aims to integrate an extensive list of database resources and analysis tools in a computationally accessible and extensible manner across different disciplines, facilitating automated data retrieval and processing with a special focus on systems biology. The EnCORE platform is available as a collection of webservices with a common standard format (EnXML), easy to integrate in Workflow management software such as Taverna. Additionally EnCORE services are also accessible thought EnVISION, a web graphical user interface providing elaborated information such as molecular interaction, biological pathways and system biology models. Information about the ENFIN project and the EnCore platform is available in http://www.enfin.org

	Abstract No. 328 View the poster
	Gustavo Salazar, G; Nicola Mulder, N; Edwin Blake, E
	DAS writeback: A Collaborative Annotation System
	We designed and developed a Collaborative Annotation System for Proteins called DAS Writeback, which extends the Distributed Annotation System (DAS) to provide the functionalities of adding, editing and deleting annotations. Several research groups have taken on the task of making protein information accessible to the community; however, the information flow in the opposite direction has not been extensively explored. DAS users are currently passive actors that behave as consumers of sources of protein annotations and they have no practical way to provide feedback to the source. As a solution, we proposed an extension of the DAS protocol, defining communication rules between the client and the writeback server following the Uniform Interface of the RESTful architecture. As proof of concept we implemented a fully functional system integrating the writeback in MyDAS (a DAS Java server) and in Dasty2 (a web-based protein client).

	Abstract No. 329 View the poster
	Villaveces, J; Jimenez, R; Hermjakob, H; Gel Moreno, B
	JSDAS a JavaScript library to query the DAS system
	JSDAS is a JavaScript client library to query biological annotations using the Distributed Annotation System (DAS). It consumes data contained in DAS sources and communicates with the registry to obtain information about available data sources. It provides an extensible interface supporting the DAS 1.6 specification. It is designed as a lightweight library structured in three levels of abstraction to provide different ways of communication. It aims to facilitate DAS web client development and data integration. JDAS is an open source project freely available at http://code.google.com/p/jsdas/

	Abstract No. 331 View the poster
	Vila, P; Rocha, M; Rocha, I
	A Software Tool for Metabolic Engineering Tasks using Integrated Metabolic/Regulatory Models
	OptFlux (http://www.optflux.org) is an open-source platform that includes several tools to support in silico Metabolic Engineering (ME), including functionalities to load genome-scale metabolic models in several formats, to simulate the phenotype of both wild type and mutant strains using steady-state approaches (e.g. Flux Balance Analysis-FBA[1]) and also to perform strain optimization tasks (e.g. finding the best sets of knockouts for the production of a given metabolite). In this work, a novel plug-in for this platform is presented, allowing the use of gene regulatory models, represented as Boolean networks. This plug-in links the regulatory model to its corresponding metabolic model, creating an integrated model and allowing its use for the phenotype simulation and strain optimization tasks. This is the first software application that allows the simulation of integrated regulatory/metabolic models. The phenotype simulation using integrated models is conducted by firstly simulating the Boolean network and, afterwards, identifying the disabled genes in the final state and considering those as ?knockouts? for the metabolic simulation that is conducted using FBA or alternative methods (such as MOMA or ROOM). The user can define the initial state of the Boolean network, use different environmental conditions and also simulate mutants by imposing a set of gene deletions. The strain optimization operation provides interfaces to identify sets of genes deletions that are able to maximize a given objective function related to the production of a given metabolite with industrial interest. Two meta-heuristic optimization methods (Evolutionary Algorithms and Simulated Annealing) are used in this task[2]. [1] K.J. Kauffman, P. Prakash, and J.S. Edwards. Advances in flux balance analysis. Curr Opin Biotechnol, 14:491-496, 2003. [2] M. Rocha, P. Maia, R. Mendes, J.P. Pinto, E.C. Ferreira, J. Nielsen, K.R. Patil,and I. Rocha. Natural computation meta-heuristics for the in silico optimization of microbial strains. BMC Bioinformatics, 9, 2008.

	Abstract No. 333 View the poster
	Jenkinson, AM; Warren, J
	Distribution Annotation System version 1.6
	The Distributed Annotation System (DAS) is an established data integration framework: a network of web services comprising hundreds of geographically distributed data sources and consumed by many independent software applications. The DAS specification, originally published in 2001, previously focused exclusively on the sharing of genomic sequence and annotations of genomic sequence. Over recent years, usage of DAS has expanded to other areas including protein sequence annotation, communication of more diverse data types and improvements to existing capabilities. This has been achieved through unofficial "extensions" to the DAS protocol. 2010 sees the publication of a new DAS specification: version 1.6. This version better reflects modern usage of DAS, clarifying ambiguity and incorporating many new concepts and features inspired by the various extensions in use by the DAS community. This is the first official amendment to the DAS specification since 2002, intended to ensure compatibility between software implementations. It represents the official recognition of several years' work on the protocol and its usage in the "real world". Improvements include the addition of the DAS Registry and a more advanced metadata system, allowing clients to automatically find and configure additional data sources. Provisions for communicating protein 3D structures and multiple sequence alignments are now present, as is a way to vastly improve performance for retrieving high-density data across the network. The final draft of the specification is available for comment: http://www.ebi.ac.uk/~aj/1.6_draft5/documents/spec.html

	Abstract No. 334 View the poster
	Fernandes, M A; Gom?z-Tom, N; Barker, GC
	A Bayesian Model of the S. aureus hazard in the milk chain integrating expert opinion and data.
	Modern food chains are no longer local and static but complex dynamic systems that cross national and international boundaries. The complexity is problematic when tracing the point of origin of biological agents (e.g. bacteria or toxins) in the chain and this hinders the implementation of appropriate interventions. As part of the EU IP BIOTRACER a Bayesian Belief Network has been developed to represent the hazard domain for S. Aureus enterotoxins in the milk chain using software called HUGIN (HUGIN A/S, Aalborg, Denmark). This construction includes the integration of diverse (and multi-disciplinary) published expert opinions and data (growth rates, toxin production, enzyme levels) for the dominant elements of the chain and provides quantitative risk assessment for S, aureus (enterotoxins) in the milk chain. The risk quantification is produced as a causal network that is a factorization of the full joint probability. Advantages include a graphical representation beneficial in communicating and understanding beliefs.. Degrees of uncertainty are included consistently and sensitivity analysis can improve an appreciation of the various factors which contribute to the overall risk. The systematic representation reveals options for including molecular data sources that can decrease uncertainty in source level inference for S. aureus contamination and Bayesian Belief systems provide a valuable framework for evaluation of such new information sources.

	Abstract No. 336 View the poster
	Kalas, M; Puntervoll, P; Joseph, A; Bartaseviciute, E; Ison, J; Liaquat, A; Blanchet, C; Rapacki, K; Jonassen, I
	BioXSD: the canonical XML-Schema data model for everyday bioinformatics Web services
	Although the flat-file formats play an important role in bioinformatics, with growing interest in interoperable services and 'computer-friendly' data formats, the use of XML is highly advantageous. XML Schema (XSD) formally defines the structure of the data. This formal definition is computer-readable, and there are well-proven industrial technologies for processing the XML Schema and XML data. BioXSD offers a set of fine-grained formal definitions of the bioinformatic data formats. It attempts to serve as the common data model for the most widely used biological data exchanged with Web services. BioXSD covers biological sequences, alignments, sequence annotations, and references to databases and ontologies. Transformers between the BioXSD and the main plain or tabular formats are included in the BioXSD development. BioXSD has been developed, and is in the process of further development and refinement, by analysing the requirements of existing Web services, tools, ontologies and data formats, and in a wide and ongoing collaboration within the community. BioXSD allows users to mix-and-match diverse services freely and without a need for 'shims' to translate between the legacy plain-text formats. Designing of analytical workflows becomes faster and cheaper, decreasing the needs for specialised personnel with advanced programming skills.

	Abstract No. 337 View the poster
	Gade, SG; Litman, TL; Ghadimi, MD; Beissbarth, TB
	Graph based data integration of microRNA and gene expression data
	The parallel measurement of different types of high throughput expression data, like gene, microRNA and protein data, gives rise to the new task of data integration. Depending on the nature of problems to adress different strategies have been purposed which perform the integration on different stages of the analysis process. Micro RNAs are small non-coding RNAs which play an important role during cancer development. The details of the microRNA mediated regulation mechanism are not fully understood. Though it is certain that the binding of a microRNA to its target transcript can lead to degradation and consequently to a measurable change in the level of the mRNA. We proposed a graph based workflow to intregate microRNA and gene expression data. By combining the correlation structure of the two data sets with target predictions we are able to construct a bipartite graph with connections between microRNAs and genes. Meta information can be added and are used to distinguish reasonable sub clusters in this graph. This graph structure can be used to perform an overrepresentation analysis of targets and effected pathways. Thus it gives rise to new hypotheses of the role of specific microRNAs and their target genes. By applying this method to a colorectal cancer data set consisting of 97 samples and measurements of microRNA and gene expressions, we are able to indentify clusters of microRNAs with significant correlations to genes in cancer related pathways like cytokine-cytokine receptor interaction and the Jak-STAT signaling.

	Abstract No. 338
	Samarajiwa, SA; Narita, M; Jauhiainen, A; Narita, M; Tavar, S
	SENESCEome: A knowledge discovery environment for integrating, mining and visualizing cellular senescence associated datasets.
	Somatic cells can enter into a state termed cellular senescence where these cells are irreversibly arrested in the G1 phase and enter a viable and metabolically active state incapable of division even in the presence of mitogens. Cells can also undergo senescence in response to stress such as DNA damage and over expression of oncogenes(1) and this process is an important barrier against oncogenic transformation. Increased availability of high throughput datasets and complexity of the senescence system necessitates the use of computational methods to extract meaningful biology and better understand this system. In response we built the SENESCEome, an integrated knowledge discovery environment. Senescence associated genes, pathways and networks were identified by analysing in-house and public microarray datasets and web and database technologies were used in building an integrated knowledgebase. SENESCEome include analysis tools enabling identification of gene signatures, analysis of regulatory motifs, pathways, interactions and expression across tissues, species and gene families. The data mining pipeline consists of purpose built pathway and regulatory analysis methods and supervised and unsupervised methods used for knowledge discovery in databases (KDD). A key feature of this pipeline is the ability to include statistical methods from the R project. Open source visualization software was used to build rich dynamic interfaces capable of displaying genomic datasets using graphs, plots, heat maps, dynamic network representations, and an integrated genome browser. This systems oriented, integrative approach coupled with the knowledge discovery tools included in SENESCEome will provide a basis for increased understanding of the senescence system. Furthermore, the methods and software tools described above are easily configurable for similar applications in other biological systems. References: 1. Hemann MT, Narita M.

	Abstract No. 339
	Tamaddoni-Nezhad, A.; Barton, R.; Hitchen, P.; Kay, E.; Lesk, V.; Turner, F.; Dell, A.; Rawlings, C.; Sternberg, M.; Wren, B.; Muggleton, S.
	An integrative systems biology approach to modeling genotype-phenotype relations in Campylobacter
	We present an exemplar research programme at the Centre for Integrative Systems Biology at Imperial College (CISBIC) which aims at building predictive models of genotype-phenotype relations in Campylobacter jejuni. The modeling techniques include a cycle of hypotheses generation and experimentation which allows the predictions of the models to be tested and the models to be revised. We use logic-based representation & inference (i.e. Abductive ILP) to generate hypotheses about the Polysaccharide structures in Campylobacter from mutants data and a background knowledge which can potentially be incomplete. Lipo-Oligo-Saccharide (LOS) synthesis pathway, which is a well-known and relatively complete pathway, was used as a test case in the initial experiments. The modeling techniques have been also tested on the Capsule bio-synthetic pathway which is a less understood pathway. Using Capsule pathway and relevant mutants data, the learned model suggests explicit hypotheses concerning the functions of unknown genes which can be investigated experimentally. Some of these hypotheses were already verified by mutagenesis and structural analysis and known to be true. However, some of the hypotheses are novel and yet to be tested experimentally in vivo. These include predictions about the involvement of some putative transferases of unknown function in the assembly of the Ribfuranose Monosaccharide of the Capsule structure. Ongoing experiments using NMR/MS techniques aim at testing these hypotheses.

	Abstract No. 340 View the poster
	Lysimachos, Zografos; Andrew, Pocklington; Seth, GN Grant; Douglas, J Armstrong
	COMPARATIVE BIOINFORMATIC STUDIES OF MAMMALIAN AND INSECT POSTSYNAPTIC DENSITY COMPLEXES AND INTERACTION NETWORKS
	The postsynaptic density (PSD) is a multi-protein complex, consisting of over 1000 proteins, which underlies molecular computation in the brain (Collins et al., 2006). We have previously isolated and analysed proteomics datasets from molecular sub-complexes of the mammalian PSD. The two most recent approaches yielded the NMDA receptor complex or MAGUK-associated signalling complex (NRC/MASC) (Husi et al., 2000; Collins et al., 2006; Pocklington et al., 2006) and the PSD-95 complex (Fernandez et al., 2009). Our previous comparative proteomics and interactomics approaches (Emes et al., 2008; Zografos et al., 2009) showed that most types of PSD proteins were present in the early Metazoan synapse and that the changes in signalling complexity, resulting from gene family duplication and diversification, were predominantly added on specific structural modules of the protein interaction network. However, wanting to elucidate the lineage specific differences or similarities in the organization of PSD complexes, we decided to directly compare complexes of the mouse (mPSD) and fruit fly's (D. melanogaster) PSD (fPSD). In this poster we will present the results of the first fPSD purification using tagged protein trap lines (Ryder et al., 2007) as well as a comparison with the mPSD by applying a comparative interactomics workflow that uses network motifs (Milo et al., 2002) and graphlets (Przulj, 2006), in combination with molecular function and biological process annotation, to compare protein interaction networks and convergence or divergence in the organisation of different molecular families within the network?s topology.

	Abstract No. 341
	Ramirez, F; Albrecht, M
	From gene sets to biological knowledge
	Screening technologies like microarrays, RNA interference or yeast two-hybrid produce large quantities of data on genes and proteins. Therefore, the analysis of the resulting sets of genes or proteins has become a ubiquitous task. Commonly, researchers want to know which molecular functions or biological processes are present in a given gene or protein set or which genes or proteins are related to a certain phenotype or disease. In the case of screens producing new protein interaction data, it is also important to identify previously reported interactions in the set of proteins. Although most biological databases are freely accessible through the Internet, the distributed nature of the available knowledge forces the scientist to visit a number of websites to gather the relevant information. Without appropriate computational tools, automated data analyses are very inefficient and time-consuming. To facilitate the analysis and interpretation of human gene sets, we developed a data warehouse that integrates almost 3 million annotations from over 25 major human gene and protein annotation databases. Our online platform BioMyn provides convenient access to the data and analysis tools (http://biomyn.de/). In contrast to other tools that offer Gene Ontology (GO) enrichment analysis only, our platform performs analysis using GO as well as sequence family classifications, protein domain architectures, metabolic and signaling pathways, ortholog species distribution, protein interactions and protein complexes, disease associations, tissue expression and UniProt keywords. Using our data analysis tools, it is also possible to create subsets of genes or proteins based on any annotation, for example, expression in brain or localization in the cell nucleus. Our data warehouse and the analysis platform have already been used successfully in the analysis of various large-scale experimental screens and in the prediction of scaffold proteins.

	Abstract No. 342
	Lysenko, A; Mansoor, S; Hodgman, C; Rawlings, CJ
	Integrated analysis pipeline for study of regulatory and signalising processes in model plant Arabidopsis thaliana.
	Modern biological research produces ever increasing amounts of relevant experimental data. Systems biology approach can be a valuable paradigm for interpreting and mining these data, but in turn requires integration of these disparate information types in a unified framework. Ondex system is one of tools that fulfil this role by facilitating merging of data from a diverse range of biological databases and providing means for visualisation and manipulation of this information by converting it into a graph of concepts and relationships between them, with their properties stored as attributes of elements in the network. To support the study of gene regulation and signalling in Arabidopsis, Ondex software was extended with new functionality for automated construction of co-expression networks from microarray data, clustering methods, ontology-driven analysis. Together with data integration methods of Ondex, these modules comprise the analysis pipeline used to assemble the integrated datasets for Arabidopsis thaliana and to combine it with relevant experimental data. Network analysis and context-sensitive graph mining methods were used to deduce modularity and indentify potentially important groups of genes that mediate diverse range of responses.

	Abstract No. 343 View the poster
	Taschuk, ML; Lister, AL; Wipat, A
	Saint: A Light-weight Model Annotation Integration Environment
	Mathematical simulation in biology and medicine is becoming increasingly important. Currently, many quantitative models in languages such as SBML contain only the formulas and parameters required to allow a model to be simulated. While extremely useful to the modeller who created the model, the lack of biological context limits the potential for re-use as well as the usability of the model for computational and logical reasoning. With the addition of biological information (in the form of annotation), models may be more easily re-used and shared with other modellers. However, manual annotation can be difficult and tedious. The annotation necessary for a model may be found in many separate databases, services, and websites. Not only must the modeller be aware of all of these different annotation sources, he must also be aware of the appropriate method to integrate the information into his model. In very large models, manual annotation is infeasible. Even in smaller models, it can be laborious. Saint is a data integration tool which allows modellers to easily and rapidly add relevant biological annotation to their models. Saint integrates data from across multiple data sources and allows a user to selectively add relevant annotations through an easy-to use interface. No knowledge of the modelling language is required. Saint also can suggest expansions to the model, including potential new reactions or pathways. Modeller may then download a copy of their model full of new annotations and reactions. In this way, models can be more rapidly constructed and distributed.

	Abstract No. 344 View the poster
	Taschuk, ML; Simillion, C; Hallinan, J; Wipat, A
	CID: CISBAN Interactome Database
	Complex biological phenomena such as cellular senescence and ageing are the result of interactions between different pathways and cellular components. Ageing is one of the most complex of these emergent properties, caused by a multitude of factors, including defective mitochondria, production of reactive oxygen species, aberrant proteins and DNA damage. Only by simultaneously studying all these different factors and the interactions between them can the process of ageing be fully understood. A systems-level approach depends upon the reliable identification of the interactome, i.e. the set of molecular interactions, of the organism of interest. Interactome data from model organisms can be used to infer knowledge in less well-studied interactomes. However, the overall quality of the data in an interactome can be highly heterogeneous due to the variety of different methods used to produce interaction data. In addition, more complex pathways and biological systems are rarely equivalent between different species. We present the CISBAN Interactome Database (CID), an integrated resource for comparing interaction networks across different species. This database contains probabilistic functional integrative networks of emprirical eukaryotic interaction data, integrated using a novel Gene Ontology-based Gold standard. We demonstrate that CID can be used to predict the existence of ageing-related pathways using model organism interactomes. These predictions were validated using both expression and knock-out data.

	Abstract No. 345
	Ceh, E; Calin, GA; Dovc, P; Cannistraci, CV; Kunej, T
	Integration of microRNAs into gene networks associated with chronic lymphocytic leukemia
	Network analysis is a highly efficient approach for developing and understanding interactions which involve microRNA (miRNA). These insights are important in elucidating background of many diseases, especially chronic lymphocytic leukemia (CLL) since it was shown that miRNAs are causally involved in this type of leukemia. Recently, the high-throughput genomic techniques identified many CLL candidate miRNAs and their targets, but integratively this knowledge has not been fully exploited. We reviewed the PubMed database (http://www.pubmed.com) for the CLL studies published between 1978 and 2009. The literature search yielded more than 2000 CLL-associated loci from 250 articles discussing cytogenetic, epigenetic and expression studies, linkage and association analysis. Among them, 11 studies describing 197 CLL-associated miRNAs were published. Over 1000 experimentally validated targets of H. sapiens miRNAs were obtained from miRecords database (http://mirecords.biolead.org/). The overlap between all three data sets revealed 96 experimentally tested miRNA-target pairs that were previously associated with CLL. The interactions were entered into MetaLink and visualized against MetaCore's manually curated network. Using Dijkstra's shortest path algorithm the most significant miRNAs in the network were miRNAs belonging to the let-7 family (i.e. let-7a, let-7b, let-7i, let-7f), miR-21 and miR-146a. Microarray data and other experimental approaches confirmed the abnormal expression of these miRNAs in CLL samples and identified miRNAs behaving as oncogenes or tumor suppressors as well as their roles as diagnostic and prognostic markers.

	Abstract No. 346 View the poster
	Mitsopoulos, C; Sims, D; Bursteinas, B; Gao, Q; Jain, E; MacKay, A; Zvelebil, M
	ROCK-BCFG : a Breast Cancer Functional Genomics Resource
	The heterogeneity of breast cancer pathology and response to treatment has instigated a range of molecular profiling approaches to attempt to stratify breast cancer subtypes and patterns of genomic rearrangement. However, gene-specific data from large-scale studies is often difficult to access, and many functional genomic datasets can be mined in different ways from those originally published. We have developed ROCK-BCFG (brcabase.icr.ac.uk) to provide a unique, publicly accessible resource for the integration of breast cancer functional genomic datasets. ROCK-BCFG provides a simple online interface for the navigation, cross-linking and cross-correlation of gene expression, aCGH and RNAi screen data from breast cancer cell lines and tumour samples. It enables the interrogation of gene lists in the context of statistically analysed functional genomic datasets, interaction networks, pathways, GO terms, somatic mutations and drug targets. It provides interactive visualisations of gene expression, aCGH and RNAi datasets as well as interaction networks. ROCK-BCFG collates data from a wealth of breast cancer molecular profiling and functional screening studies into a single portal, where analysed and annotated results can be accessed at the level of a single gene, sample or study. Thus, it enables cancer researchers to quickly assess the potential significance of genes of interest in the context of prior knowledge in the field.

	Abstract No. 349 View the poster
	Hindle, MM; Defoin-Platel, M; Saqi, M; Hodgman, C; Habash, D; Rawlings, C
	Improving sequence annotation: combining data-integration with BLAST and HMM alignment
	Given the inexorable rise in the number of sequences being produced and the inevitable backlog of experimental validation of function, improving the quantity and quality of annotations that can be computationally assigned to a new sequence is an immediate priority in bioinformatics. We demonstrate a gene annotation workflow built within the Ondex data-integration system. We focus on annotation to the Gene Ontology and Enzyme Commission Nomenclature and demonstrate how integration of AraCyc, KEGG, ENZYME, and other databases can provide mutual enrichment of annotation information. Alignment of the sequences is done using conjoint BLAST and HMMER, where annotations are inferred from annotations of similar proteins or domains. The system is demonstrated for the wheat Affymetrix microarray and the results are compared to annotation pipelines based on sequence alignment. Our observations confirm the utility of improved data integration to sequence annotation and identify three areas for optimization in current pipelines: sequence alignment, data-integration, and annotation scoring metrics.

	Abstract No. 350 View the poster
	Mazza, Davide
	A Haptic Framework for Molecular Interactions Study in Drug Design
	Molecular interactions analysis has acquired a great important in last decades. The study of the interactions among molecules and the ways in which they can be aggregated to obtain new chemical compounds with desired conformations and properites is at the basis of the supramolecular chemistry field. Especially in designing innovative and more effective drugs, it is fundamental to obtain a deep and precise understanding of the inter-molecular forces that govern this kind of process. Haptic tecnhnologies (i.e., that makes user able to feel forces) can greatly help here because the manipulation through the sense of touch of the involved forces can be used to feel and demonstrate many of the properties of studies molecules. In this work we present the realization of a framework exhibiting a virtual environment coupled with a haptic device, where the user can sense the electrostatic field around a shown molecule simulating the movement of a probing electric charge in the surrounding space. The tool can be applied on different molecules, taken from a repository, of medium complexity in terms of number of atoms and involved ligands. Molecules with particular importance in the biological and pharmaceutical field have been considered. Data and information used to provide a replication of the phenomena as precise as possible are obtained thought the usage of a computational chemistry tool widely used in research.

	Abstract No. 358
	T?ndel, K.; Indahl, U. I.; Martens, H.; Omholt, S. W.
	Modelling time series data from mathematical models using multivariate analysis
	A deterministic mathematical model for the mammalian circadian clock has been established earlier by Leloup and Goldbeter [1]. This model is based on the positive and negative regulatory loops involving the Per, Cry, Bmal1, and Clock genes, and estimates circadian oscillations in conditions of continuous darkness. In order to identify sources of epistatic effects in the system, we have compared the effects of the various parameters and investigated possible interaction effects using multivariate analysis. A similar analysis has been carried out using data from a mathematical model of the mouse heart cell [2]. Using Partial Least Squares Regression we are able to obtain good predictions of the various time dependent outputs (e.g. action potentials) from the deterministic mathematical models directly based on the specified parameter values. This is attractive both with respect to understanding the dynamics of the system and improving the computational efficiency of the corresponding deterministic models. 1. Leloup J-C, Goldbeter A (2004) Modeling the mammalian circadian clock: sensitivity analysis and multiplicity of oscillatory mechanisms. J Theor Biol, 230:541-562 2. Bondarenko VE, Szigeti GP, Bett GCL, Kim S-J, Rasmusson RL (2004) Computer model of action potential of mouse ventricular myocytes. Am J Physiol Heart Circ Physiol 287(3):H1378?403

	Abstract No. 359 View the poster
	Tanoh, F; Bhagat, J; Nzuobontane, E; Laurent, T; Wolstencroft, K; Stevens, R; Pettifer, S; Lopez, R; Goble, C
	BioCatalogue-The Life Science Web Service Registry
	Web services are crucial for access and analysis of data and resources in Life Science. Many institutions and organisations such as the European Bioinformatics Institute, the National Center for Biotechnology Information and the DNA Data Bank of Japan provide Web services to access and analyse their resources. Despite their importance, Web services are hard to find, poorly described and poorly documented, making them hard to use. To address these issues and enable a wide adoption of Web services, the authors have recently launched the BioCatalogue (http://www.biocatalogue.org/). The BioCatalogue is a registry of Life Science Web services. It is a place where the community can find, annotate, understand and monitor Web services. It is also a platform for Web service providers to publish and advertise their services. The goals of the BioCatalogue are to provide a centralised and curated catalogue of Life Science web services, and to build a collaborative environment where the community can find, contact and meet the experts and maintainers of these services. The BioCatalogue has adopted a community approach in the sense that services can be submitted, annotated or commented on by the providers, the users or expert curators. Web services in the catalogue are constantly monitored for their availability and reliability. The BioCatalogue currently holds over 1100 web services and has more than 200 registered users. The BioCatalogue is the product of collaboration between the myGrid project at the University of Manchester and the European Bioinformatics Institute.

	Abstract No. 361 View the poster
	Love, CG; Andongabo, A; Wang, J; Carion, P; King, G; Verrier, P
	BrassEnsembl: Displaying Brassica data within the Ensembl framework
	BrassEnsembl is a key component in a suite of tools developed at Rothamsted Research for associating genetic, genomic and phenotypic information in Brassica. The resource uses the Ensembl genome browser as a basis for displaying the reference genome sequence and annotation, whilst providing links to genetic marker data and subsequent phenotype information as well as comparative analysis with Arabidopsis. Fundamental to the design of the tools is the ability to trace the analysis back to the underlying parameters and sources of data, allowing researchers to have confidence in the associations identified. Interlinking these tools provides a valuable resource which simplifies the ability to browse between genomic sequence, genetic and phenotypic data in Brassica. BrassEnsembl is accessible from http://www.brassica.info/BrassEnsembl

	Abstract No. 362 View the poster
	Love, CG; Andongabo, A; Castells-Brooke, N; King, GJ; Verrier, P; Mitchell, RAC
	PpETS: A tool to identify locus specific transcript sequences in polyploidy crop species
	Polyploid Estimated Transcript Server (PpETS) is a major expansion and improvement of the successful WhETS tool. The tool is designed to assist biologists seeking to design locus-specific PCR primers for polyploidy crop species that lack sequenced genomes, using publicly available EST and mRNA sequences. PpETS utilises the closest matching gene model from related diploid species with a fully-sequenced genome as a basis for separating homoeologs through identification of sequence polymorphisms. Modified assemblies are aligned back to the gene model and presented graphically displaying information related to assembly components with links to download the analysis. PpETS is accessible from http://www4.rothamsted.bbsrc.ac.uk/ppets

	Abstract No. 363 View the poster
	Rohn, H; Klukas, C; Schreiber, F
	Integrative Visualization of Multimodal Biological Data
	Many approaches in Systems Biology take into account information on metabolite-, enzymatic and gene-regulatory levels, but most neglect spatial (except compartmental) information of their distribution. In our approach we combine standard Systems Biology visualization with manifold two-, three- and four-dimensional information, such as high-resolution NMR-volumes and spatio-temporal metabolite distribution derived from stacks of in situ cross-sections. Based on a Java3D volume renderer and VANTED, a tool supporting visualization of *omics data in the context of networks, we developed a platform integrating and visualizing these different types of data. The approach allows to visualize the data separately, map the data from different domains together and visualize the mappings using views, such as 2D graph view, image stack view and 3D view. These methods allow to explore and understand the relations between the data in an intuitive matter by taking into account all available data using a system wide approach.

Comments on the content and suggestions for improvements to these pages are always welcome.
Webmaster: Paul.Verrier@bbsrc.ac.uk.