7/28/2019 Sea Life Leaflet

1/2

ICT for Health Resource book of eHealth Projects FP6

Sealife:A Semantic Grid Browser for the Life Sciences

applied to the study of Infectious DiseasesSealife will develop a browser, which will link the existing Web to the currently emerging eScienceinfrastructure. The Sealife browser will be demonstrated within three application scenarios in

infectious diseases relating to evidence-based medicine, literature mining, and molecular biology.

Objectives of the project

Problem or Context:

Currently, much effort is spent on creating a new

computational and data infrastructure tofacilitate eScience, the cooperation of

geographically distributed organisations, whichtransparently integrate their computational anddata resources at a structural and semanticlevel. Progress has been made with standards

for grid computing and semantic representationsfor life science data with many projects creatinga host of grid-enabled services for the lifesciences.

How can the researcher in the lab benefit fromthis new infra-structure to science? Atechnology is needed to transparently bringsuch services to the desks of the scientists. The

Web started with a browser and a handful ofWeb pages. The vision of eScience with anunderlying Grid and Semantic Web will only takeoff with the development of a Semantic Gridbrowser.

Project:

The Sealife project is filling this gap bydeveloping such a semantic grid browser. Thesebrowsers will operate on top of the existing Web,

but they introduce an additional semantic level,thus implementing a Semantic Web. Usingontologies as background knowledge thebrowsers can automatically identify entities such

as protein and gene names, molecularprocesses, diseases, types of tissue, etc. andthe relationships between them, in any Web

document, they collect these entities and thenapply further analyses to them using applicableWeb and Grid services. The browser will beevaluated in three applications relating to the

study of infectious diseases

Project Description

Sealife will solve the following problems toachieve its objectives:

1. Ontologies: Design and integration ofontologies and associated infrastructure,which can serve as background knowledgefor a Semantic Grid browser geared towardslife science applications ranging from themolecular level to the person level.

2. Concept Mapping: Bridging the gap betweenthe free text on the current Web and theontology-based mark-up for the SemanticWeb and Grid by developing automatedmark-up modules for free text, which arebased on textmining and natural languageprocessing technologies.

3. Service Composition: Bridging the gapbetween the ontologies of the SemanticWeb and the services of the Grid by linkingsuitable ontology mark-up to applicableservices and by supporting the interactivecreation of such mappings for complexservices.

The Sealife browser will be demonstrated withinthree application scenarios in evidence-based

medicine, literature and patent mining, andmolecular biology, all relating to the st\udy ofinfectious diseases. The three applicationsvertically integrate the molecule/cell, th\etissue/organ and the patient/population level bycovering the analysis of high-throughputscreening data for endocytosis (the molecular

entry pathway into the cell), the expression ofproteins in the spatial context of tissue andorgans, and a high-level library on infectiousdiseases designed for clinicians and their

patients.

To illustrate the power of this vision consider the

following applications:

Thematic AreaICT for Health

7/28/2019 Sea Life Leaflet

2/2

Evidence-based medicine: Consider aclinician, who consults the nationalelectronic library of infections to get trusted

information on infections. The user visits thesite and finds an interesting page on hipatitisand its treatment: "Ribavirin with or without

alpha interferon for chronic hepatitis C".Using its background knowledge, the Sealifebrowser identifies hipatitis as disease andinterferon as a immunologic factor. With thisknowledge the browser automatically offersthe user the ability to query the biomedicaldatabases Ensmbl and PDB to learn more.

Literature and Patent Mining: Getting a quickoverview over a field is often vital . In

browsing a patent database, a researchercomes across the patent entitled "Animproved infant formula is described whichincludes a phospholipids supplement in

order to more closely resemble thecomposition of human milk.". The Sealifebrowser identifies the term "phospholipid

metabolism" and offers the followingdefinition: "The chemical reactions andphysical changes involving phospholipids,any lipid containing phosphoric acid as a

mono- or diester.". It also identifies human inits taxonomy. The user decides that this isrelevant and wishes to learn more aboutphospholipids. The Sealife browserautomatically offers the service of showingall human proteins in the UniProt database,

which are involved phospholipidmetabolism.

Molecular Biology: Consider a biologist, whoencountered the statement ``Rabaptin-5interacts with the small GTPase Rab5 and is

an essential component of the fusionmachinery for targeting endocytic vesicles toearly endosomes''. The Sealife browser

identifies ``Rabaptin-5'' and ``Rab5'' asprotein names, ``endocytosis'' as biologicalprocess, and ``early endosome'' as cellularcomponent. When the user moves the mouse

over "Rab5", the browser offers to searchsequence databases for Rab5 proteins. Atthe same time, it offers to move the proteinsequence of Rab5 to a shopping cart. After

some time of browsing, the user decides tovisit his/her shopping cart and takes a lookat the proteins he/she has collected in the

web-session. The \slb now offers to performa series of services on the proteinsequences in the cart.

Expected Results & Impacts

Sealife builds on a number of relevant systemsalready developed by the partners:

GoPubMed.org, an ontology-based literaturesearch engine

MyGrid, a Grid computing platform,

Corese, a concept resource search engine,

NeLI, the National electronic library of infectiousdiseases,

Edinburgh Mouse Atlas.

These systems will be advanced through Sealife andwill ensure a link to a user base. Additionally, Sealifehas set up an advisory board with members fromPfizer, AstraZeneca, Unilever, and others. Dresdenhas spun-off Transinsight.com, which is dedicated tointelligent search for life sciences. Transinsight has

secured seed funding by the German High-techGrnderfonds and has obtained an award by thefederal ministry for economic affairs.

TransinsightSealife

Project title: A semantic grid browser for the lifesciences applied to the study of infectious diseasesProject co-ordinator: TU Dresden

Contact: Prof. Michael Schroeder

Tel: 0049 351 463 400 60

Fax: 0049 351 463 400 61

Email: ms@biotec.tu-dresden.de

Web: www.biotec.tu-dresden.de/sealife

Partners:

1. TU Dresden, Germany

2. Hariot-Watt University, Edinburgh, UK

3. City University, London, UK

4. University of Manchester, UK

5. Scionics GmbH, Dresden, Germany

6. Inria, Sophia-Antipolis, France

Timetable: from 4/2006 to 3/2009

Total cost: 2.6M

EC funding: 2.2M

Instrument: STREP

Project Identifier: FP6-2006-IST-027269

Keywords: Grid, semantic web, molecular biology,healthcare, bioinformatics