Conference Report: The ESF Programme on Integrated Approaches for Functional Genomics ...downloads.hindawi.com/journals/ijg/2002/154652.pdf · 2019. 8. 1. · Genomics. Workshop on

Feature

Conference Report: The ESF Programme onIntegrated Approaches for FunctionalGenomics. Workshop on ‘Data Integrationin Functional Genomics and Proteomics’October 15–17th, 2001; Swiss Institute of Bioinformatics, Geneva, Switzerland

Pierre-Alain Binz1*, Annette Martin2, Mike Taussig2 and Antoine de Daruvar31 Swiss Institute of Bioinformatics, Switzerland2 The Babraham Institute, Cambridge, UK3 LION bioscience, Bordeaux, France

*Correspondence to:Swiss Institute of Bioinformatics,Proteome Informatics Group,1 Rue Michel Servet, CH-1211Geneve 4, Switzerland.E-mail:[email protected]

Published online:

2 January 2002

Introduction

Beyond an analysis of gene function at the genomelevel, the ultimate goal of functional genomics is tounderstand the organisation and coordinated oper-ation of the cell. Integration of data and informa-tion is an essential feature of all the steps leadingfrom the production of experimental results tothe modelling of a complete cell. The Geneva work-shop, organised within the framework of theEuropean Science Foundation (ESF) Programmeon Integrated Approaches for Functional Genomics(http://www.functionalgenomics.org.uk), provided anexcellent opportunity to review the issue of dataintegration from various perspectives. It broughttogether scientists with different backgrounds (bio-logists, bioinformaticians), who currently participatein consortia involving or requiring the integrationof heterogeneous biological data. The workshopdedicated equal time to presentations and discus-sions in order to optimise knowledge distributionand sharing.

The first session was devoted to existing func-tional genomics projects, where diverse sets ofexperimental approaches are applied to a modelorganism or given project. For these, both the scien-tific objectives and data management strategies weredescribed. Another set of presentations focused on‘data integration requirements’ in proteomics. Inthis rapidly evolving area, data integration is a keyissue for technological developments that are beingcarried out in both academic and industrial con-texts. Data integration approaches were then pre-sented by scientists involved in the design andmaintenance of public database resources. The lastsession was devoted to bioinformatics and gave anoverview of state of the art approaches for buildingup information systems with good capabilities withrespect to data integration. The workshop con-cluded with an open discussion on bottlenecks andperspectives that emerged from the presentations.

As a number of speakers have submitted reviewsbased on their presentations that are appearing inthis issue of CFG and in the next one, in this report

Comparative and Functional Genomics

Comp Funct Genom 2002; 3: 16–21.DOI: 10.1002 / cfg.134

Copyright # 2002 John Wiley & Sons, Ltd.

we only briefly mention the content of their con-tributions at the workshop.

Sessions and subsequent discussions

European functional genomics projects:scientific issues and status

Uwe Kaerst (Gesellschaft fur BiotechnologischeForschung, Braunschweig, Germany) introducedthe European Union funded REALIS project(Molecular strategies for adaptation and survival:Post-genomic analysis of the lifestyles of Listeriamonocytogenes (LM) in the environment and theinfected host) which commenced in February 2000(see Kaerst et al., in this issue). As the LM genomeis already known, the project aims to completelydecipher all the genes required for survival andadaptation of LM within an infected host and forresponses to the external environment. Usinggenomic and post-genomic tools, REALIS alsoseeks to precisely address questions regarding theevolutionary relationships between pathogenic andnon-pathogenic Listeria and to define the qualitiesof particularly successful clonal pathovariantswhich cause disease. Workpackages are focused ontranscriptomics, proteomics, large-scale generationof mutants, central regulon analysis, comparativegenomics and bioinformatics. The latter workpack-age aims to develop an integrated bioinformaticsdatabase, based on the SRS6 system.

Peter Rice (LION bioscience, UK) describedRIBDB, an integrated database of Listeria experi-mental data. Based on SRS6 (http://srs.ebi.ac.uk), itis hosted on a central website and acts as theREALIS consortium support for internal queries. Itis currently populated with public database seq-uences of Listeria, gene lists, annotations for L.monocytogenes, expression data and proteomic spotannotations (see Rice et al., in this issue).

Philippe Glaser (Institut Pasteur, Paris, France)presented the comparative genomics activity thatwill be implemented as part of the REALIS project.He depicted phylogenetic relationships betweenvarious Listeria spp., Bacillus subtilis and Mycobact-erium genitalium. Symmetry, similarities and struct-ure conservation were discussed, based on analysisof genomic sequences.

Javier Paz Ares (Centro Nacional de Biotecnolo-gia, Madrid, Spain) presented the aims and struct-ures of the REGIA (Regulatory Gene Initiative in

Arabidopsis) project (see the review in our nextissue), which started in February 2000 with twoyears of funding. It is mainly focused on thecharacterisation of transcription factors (TFs) andtheir relevance to plant breeding programs. Theseven workpackages include expression patterns,identification of mutant and TF loci, ectopicexpression, phenotype analysis using high through-put metabolic profiling, interaction using two-hybrid screening (Y2H) and bioinformatics. Pro-teomics is not included in this project. About 1500TFs are recognised, from which, about 420 zincfinger proteins have thus far been predicted.

The information storage, analysis and webrepresentation of REGIA (see the review in ournext issue)were presented by Alfonso Valencia(Centro Nacional de Biotecnologia, Madrid, Spain).About 30 groups interact with the database. Thearchitecture uses XML to transfer the data to thecentral relational database. No raw data areintroduced. While the metabolic data and Y2Hdata are formalised and normalised, there is nosuch standard for transcriptional and mutantanalyses. There is a fundamental difficulty inhomogenising the experimental and analytical pro-cedures, when various groups do not employ thesame approaches. This is yet to be completelyresolved.

Wilhelm Stiekema (Wageningen University, TheNetherlands) presented the goals of the recentlyfunded EU PLANET (European Plant GenomeDatabase Network). With 8 partner groups, theworkpackages include topics such as ontology,high-density maps, data mining tools, haplomaps,diversity maps, metabolome profiling, genomics,and proteomics. The study involves analysis ofgene function for plant protection. Much of theavailable data, partially that coming from Germ-plasm DB (6 million accessions, including moleculargenetic fingerprints of more than 250 markers),will be combined with data generated by the con-sortium to build an annotation pipeline. This willprobably be based on flat files and automated usingPerl scripts. He also mentioned EXOTIC (the ExonTrapping Insert Consortium), the aim of which isto detect the spatial and temporal expressionpatterns of approximately 5000 Arabidopsis genes,and to identify and characterise their regulatoryelements.

Colin Harwood (Newcastle University, UK) pre-sented the activities of BACELL (From generegulation to gene function: regulatory networks in

Conference Report 17

Copyright # 2002 John Wiley & Sons, Ltd. Comp Funct Genom 2002; 3: 16–21.

the model Gram-positive bacterium Bacillus subtilis,see Harwood and Moszer in this issue). The 11groups forming this EU funded consortium havedivided their activities into four workpackages.Transcriptomic and proteomic approaches are usedto observe responses to environmental changes(stimulons and regulons) and to characterise reg-ulatory proteins. Global regulatory networks will bebuilt using data integration and existing databases,such as MICADO (http://locus.jouy.inra.fr/cgi-bin/genmic/madbase/progs/madbase.operl/) and Sub2D(http://microbio2.biologie.uni-greifswald.de : 8880/).

Peter Jungblut (Max-Planck-Institute for Infec-

tion Biology, Berlin, Germany) discussed the EBPthematic network (Comparative analysis of pro-teome modulation in human pathogenic bacteriafor the identification of new vaccines, diagnosticsand antibacterial drug targets). It comprises 10workpackages including studies on 6 pathogenicbacteria. Besides its functional goals, the networkaims to build a database system, incorporatingmainly proteomic but also transcriptomic data (seePleissner et al., in this issue). Two complementaryalternative approaches have been proposed to theconsortium members: a centralised database inBerlin (see Pleissner et al., in our next issue) and asystem based on the federated model of World-2DPAGE (http://www.expasy.org/ch2d/2d-index.html).

Clive Edwards (University of Liverpool, UK)introduced a field not frequently considered instandard approaches to organism study, namelymolecular ecology. He highlighted the importanceof monitoring active prokaryotes in their naturalenvironments, as these organisms are involved inkey biogeochemical cycles and may have pathogenicincidences. In the last 10 to 15 years, molecularbiology methods and particularly sequencing of 16SrRNA have been used to detect ‘‘functional genes’’in prokaryotes. The discovery of such genes doesnot necessarily reflect actual activity, however.DNA array technology could provide a possiblealternative for these approaches. One major techni-cal difficulty is that of culturing the vast majority ofmarine prokaryotes. Proteomic approaches areproposed in order to identify active prokaryotesand environmentally induced active protein synth-esis, using two-dimensional polyacrylamide gelelectrophoresis (2-D PAGE) coupled with pulse-labelling experiments and mass spectrometry (MS).The cellular origin of these proteins could beobserved via fluorescently labelled probes.

Session on proteomics

Manfredo Quadroni (University of Lausanne,Switzerland) addressed the proteomics issues ofgoals, technologies and quality requirements.While defining the first phase of proteomics analysisas the production of experimental data, he dis-cussed the difficulty of capturing and comparing2-D PAGE and MS results when there exists such adegree of protocol heterogeneity between differentlabs. The second obstacle corresponds to datahandling and analysis. He pointed out somerequirements related to data management, databaseusefulness and data quality assessment. Here thereis a need for detailed descriptions of sample originand preparation, relationship to expression levels,subcellular localisation, 3-D structure, biologicalactivity, description of post-translational modifica-tions and interpretation of protein interactionmaps, among others. He also discussed the qualitydifference between curated and archived databases.

Ron Appel (Swiss Institute of Bioinformatics,Geneva, Switzerland) focused on quality issuesapplied to 2-D PAGE image analysis and relateddatabases. At the user level, efforts have to be madeto describe all parameters describing the generationof 2-D PAGE images. These include sample choiceand preparation, the protocols for the 2-D PAGEseparation, the gel staining and digitising steps. Theimage analysis software should, on its side, deliverinformation on the type of algorithm used for spotdetection and matching, together with associatedparameters and a measure of confidence. Theprinciple ‘garbage in, garbage out’ can be thereforeunderstood as ‘quality in, quality out’. He presentedan application of, and the related challenges for,the Melanie image analysis software (http://www.expasy.org/melanie), and the advantages and issuesof the Federated Model of 2-D PAGE databases.

Joel Vandekerckhove (University of Ghent,Belgium) presented alternatives to 2-D PAGEbased protein identification methodologies. Besidesthe ‘multi-dimension liquid separation protein iden-tification technique’ (MudPIT) and the ICATtechnique, he introduced a new methodology invol-ving specific amino acid labelling and diagonalchromatography. He illustrated the advantages ofthis technique, compared to 2-D PAGE basedidentification, in enhancing sensitivity.

Denis Hochstrasser (Geneva University, GenevaUniversity Hospital, GeneProt, Switzerland) intro-duced the principle and the capacities of the

18 Conference Report


molecular scanner, a parallel, high-throughputmethod of analysing 2-D PAGE separated pro-tein samples. He then presented the industrialapproach that GeneProt (http://www.geneprot.com)has chosen to analyse entire proteomes with highsensitivity.

Session on databases

Amos Bairoch (Swiss Institute of Bioinformatics,Geneva, Switzerland) presented the current statusof the SWISS-PROT knowledge base (http://www.expasy.org/sprot) and related databases. As theintegrated data comes from various types ofsources, he pointed out that the maintenance ofthe information and of the cross-references toexternal databases is difficult. In the case of cross-references to Medline/PubMed, current stability isgood. However, it is difficult to link directly to anarticle as the journals use different systems. SWISS-PROT has cross-links to 46 databases and 21 extracreated on the fly. There are links to 38 2D-PAGEdatabases on the web that are providing annotationfor about 200 images. However there are currentlyno public MS databases available on the web. TheDR lines (Cross-references) are in general difficultto maintain, as various databases have unstableunique identifiers, which necessitate manual cura-tion. The CC database is used to link informationresources relevant to only a small number ofproteins. These types of annotation are particularlylabour intensive. A new feature is the OX line,which links to the NCBI taxonomy. As taxonomyfrequently changes, these lines have to be updatedfrequently, almost every week. In order to helpannotation, controlled vocabularies are used forseveral topics, including keywords, journal abbre-viations, tissue, plasmid name, catalytic activity, etc.At the sequence level, up to 12 000 variants havebeen validated and annotated. There are also linksto other information, including synonyms, MSdata, cofactors, and pathways. In the future,SWISS-PROT will also be distributed in XML.

Henning Hermjakob (European BioinformaticsInstitute, Hinxton, UK) introduced present andfuture projects of the European BioinformaticsInstitute (EBI, http://www.ebi.ac.uk). Temblor,accepted in May 2001 for three years, is a 25-member consortium (see Hermjakob and Apweiler,in this issue). It will provide a highly integratedview of genomic and proteomic data (Integr8)by drawing on databases maintained at major

bioinformatics centres in Europe, and by creatingnew and important resources for protein-proteininteractions (IntAct), structural (EMSD) andmicroarray (DESPRAD) data.

Ivan Moszer (Institut Pasteur, Paris, France)presented the Subtilist DB (http://genolist.pasteur.fr/SubtiList/), dedicated to Bacillus subtilis. Histori-cally based on the MICADO and Sub2D databases,the main features of Subtilist include enhanced andverified information on contigs, genes and proteins,EMBL entries and bibliographic references. As wellas sequence correction, applied to more than 200genes, a significant number of ‘unknown’ geneshave been labelled and associated with functions.Subtilist is currently being adapted to handletranscriptomics data that are produced in theframework of the BACELL project (above). Thedatabase design is done very carefully in order tostore all the information, which will be useful tosupport quality control and expression data analy-sis. The system can be extended with additionaltools and links and can also be generalised for usewith other organisms.

Session on bioinformatics

Robert Stevens (Manchester University, UK) pre-sented a talk on ontologies, a field which isincreasingly used within the world of genomicsand functional genomics. He presented the compo-nents of an ontology and the process of buildingone in order to capture knowledge about a domain(see Roberts, in this issue).

Paul Van der Vet (University of Twente,Netherlands) talked about C2M, a configurablechemical middleware. The current difficulty experi-enced in exchanging information from heteroge-neous resources is related to format multiplicity.Ideally, users should have several tools at theirdisposal, such as a wrapper (data converter) gener-ator that uses high-level format description, a codegenerator that turns these descriptions into anappropriate program code, a compiler, and a docu-mentor that turns these descriptions into a humanreadable documentation. C2M is a prototype thataims to meet these required specifications (see Vander Vet, in issue 2/6).

Anne Morgat (INRIA, Montbonnot-Saint Martin,France) introduced Panoramix, a two-year projectaimed at federating a set of knowledge basesdedicated to microbial genome annotation (see thepaper in our next issue). From 49 known microbial



genomes, two have been chosen as referenceorganisms. Four activities are linked to Panoramix:Genomix (genomic information from genomesequences), Proteix (protein information, based onextraction from SWISS-PROT), Organix (organiccompounds involved in biological processes) andMetabolix, (dealing with the description of meta-bolic pathways). Metabolix faces challenges asso-ciated with heterogeneous user points of view (eg.,chemists see chemical compounds, biochemists seeenzymes, geneticists see genes encoding proteins,computer scientists see graphs encoding reactions).Anne highlighted the difficulty of integrating incon-sistencies and incomplete data with existing general-ised and specialised databases. An example is thecomparison of chemical compound names as CASnumbers and/or structural representations.

Denis Shields (Royal College of Surgeons, Ireland)presented an approach for integrating genotypingdata. He talked about the EU prospective cardio-logy project, MORGAM, and the World HealthOrganisation’s MONICA project (http://www.ktl.fi/monica/index.html). The idea of the bioinformaticsactivity here is to integrate genetic variationinformation coming from phenotypic, metabolic,proteomic, transcriptomic and genetic data (seeShields and O’Halloran, in this issue). The principleof the planned architecture is to use a LaboratoryInformation Management System (LIMS) to handleraw data and to centralise them in a database forlater analysis using statistical and data mining tools.The design of the centralised database is still notfinalised and will depend on the types of datagenerated.

Bottlenecks and perspectives

The last session took the format of an opendiscussion based on a number of predefined topics.

The possibility of sharing data and databaseschema. There is an obvious need to compare andperhaps exchange database schema betweenongoing projects. As presented here, most of theprojects involve creation of a database and alldevelopers had encountered the difficulty of hetero-geneity of format and data description in existingdatabases. The discussion highlighted the practicaldifficulties of sharing database schema from differ-ent networks, owing to several reasons. Somedatabases are not at a stage suitable for sharinginformation or are not stable yet. In addition, thereare intellectual property issues and restrictions from

contracted partners (commercial or otherwise). Atthis point it was noted that the EU has onlyrecently started to fund bioinformatics activity inmost of these projects. It was suggested that pub-lic funding ought to be related to public access tothe information in the databases and their schema.This is particularly important as numerous publica-tions are based on data that are not publiclyavailable.

Post-project maintenance of the databases is alsorequired, which needs additional funding. Even iffunding of sequence databases, such as SWISS-PROT or EMBL, is more or less stable today, thisis not the case for the newer types of functionalgenomics related databases. There is currently nosource of EU funding for curating these highlyimportant data. Currently, at the end of the grantedperiod, there is no rule and no funding assigned tothe maintenance of the data collected, hence posinga considerable threat that all of this valuable datamay be lost. This topic has to be further discussedand practical solutions have to be reached.

A consensus was agreed amongst the projectsthat have been running for more than one year thatinteraction between biologists and bioinformaticians

has to be improved. The current lack of interactionrelates to the difficulty bioinformaticians face inpredicting the needs of biologists and evaluating thefuture possibilities of this rapidly evolving techno-logy. The current tactic uses a pragmatic approachof matching a ‘prototype’ model. This then requiresspecific redesigning and optimisation. It is onlyrecently that the MGET and EU PLANET projectshave been able to define a workpackage specificto dealing with this issue. Clearly, biologists need tounderstand the capabilities of bioinformatics toextract and interpret the masses of data that aregenerated by their research to assist them further(see Van der Vet et al., in issue 2/6).

Status of modelling and systems biology. A newand far-reaching area, this topic was generally notrepresented in the various projects here. The issuepertains to the collection of large amounts of datafrom extraneous sources, which is not alwayspossible within these projects. In addition, the lackof common crossover between databases is cur-rently a limiting factor. A proposal has beensubmitted for an ESF programme workshop in2002, focusing on this issue. The same argumentsare true when initiating data mining activities; here,a workshop may be scheduled for 2003.

20 Conference Report


Conclusions and future

The workshop, as an exploratory experience, wasconsidered very interesting and successful by a vastmajority of the participants. The broad range ofindividual expertise raised interesting issues andhighlighted the need for multidisciplinary crossoverbetween the partners of EU functional genomicsprojects. This workshop was aimed at assemblingcombined knowledge about the current status ofthese European projects, which was a unique oppor-tunity. In addition to the comments made above,some further proposals have been made for futureactivities and improvements to the format of work-shops such as this. Now that an initial status on thevarious projects has been assessed, further shorterworkshops focusing on more specific topics shouldbe organized, perhaps as satellites of conferences, inorder to maintain the momentum of sharing andintegration begun here. The current ESF pro-gramme may finance several training courses orpersonnel exchanges, within the programme areas,in order to facilitate crossover and integration ofbioinformatics and biology. Efforts are still being

made to facilitate and enhance the content (qualityand format) of databases and to provide the endusers with powerful and transparent, optimizedtools. This may be made possible by encouragingbetter communication between established andfuture projects.

Acknowledgement

REALIS, REGIA, BACELL, EBP are funded under the

Fifth Framework Programme by the Quality of Life and

Management of Living Resources Programme (http://europa.

eu.int/comm/research/quality-of-life.html) of the European

Commission. Description of the projects can be found on

the Community R&D Information Service (CORDIS) server

(http://www.cordis.lu/).

The organisers would like to thank the ESF for financing

the workshop and Geneva Bioinformatics, GeneProt and

LION bioscience for generous sponsorship. We are also

grateful to the Medical Faculty of the Geneva University,

which provided the conference rooms and infrastructure.

Thanks also go to Vanessa Lenglart who entertained the

participants with a piano recital in Cartigny, a charming

village in the Geneva countryside.



Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation http://www.hindawi.com

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttp://www.hindawi.com

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research


International Journal of

Microbiology

Documents

Conference Report: The ESF Programme on Integrated Approaches for Functional Genomics ...downloads.hindawi.com/journals/ijg/2002/154652.pdf · 2019. 8. 1. · Genomics. Workshop on