Search Options | Help | Site Map | Cultivate Web Site
		Search Disabled

The ORIEL Project - biological information management from a new perspective

By Les Grivell- November 2002

Les Grivell reports on the EU-funded ORIEL Project, a project focused on digital information management in the biosciences.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Introduction

Starting on 1st January 2002 and running for three years, the European Commission is funding ORIEL, a project focused on digital information management in the biosciences (Contract number IST-2001-32688). ORIEL (Online Research Information Environment for the Life Sciences) was formulated as a response to the need of biologists to deal with a growing stream of information generated by genomics, imaging and other data-intensive technologies. ORIEL aims to provide biological research communities with tools to manage large, complex, multimedia datasets and to navigate through an increasingly intricate and potentially confusing information landscape.

Writing in 1985 in a committee report for the US National Academy of Sciences, Morowitz argued that biological research had reached a point where "new generalizations and higher order biological laws are being approached, but may be obscured by the simple mass of data" [1]. His warning was perhaps at that time a little premature, but it has turned out to be truly prophetic. Biology is currently undergoing a revolution.

The development that probably has had most impact on research methodology and perspectives is Genomics. This young and rapidly expanding discipline focuses primarily on the determination and interpretation of the DNA sequences of the entire genetic blueprint (genome) of an organism. Genomics research is characterised by the production of vast amounts of raw and derived data obtained from high throughput technologies that include whole genome gene and protein analysis, systematic protein 3D-structure determination, and real time molecular and cellular imaging. Each of these areas shares a number of features:

• data-intensive character
• dependence on use of the Internet for the communication of results and the presentation of large datasets and of Web resources that allow their further interpretation or visualisation
• application of computational methods developed within the new discipline of bio-informatics

To give some impression of the scale of the problem, as of 10th October 2002, the number of DNA sequences deposited in the combined international nucleotide sequence databanks formed by DDBJ (DNA Data Bank of Japan; CIB, Mishima, Japan), EMBL (EBI, Hinxton, UK) and GenBank (NCBI, Bethesda, USA) amounted to a staggering 29.9 gigabases in around 19.7 million records [2]. In 1985, the year of the Morowitz report, the total number of DNA sequence records in the EBI-EMBL nucleotide database was around 5000, while in 2001, the number of entries added to the database per day was around five times this number (see Figure 1 below).

Figure 1 : Exponential growth of DNA sequence data held in the joint EMBL- GenBank-DDBJ databanks, with a number of landmarks in genomic sequencing

The integration of the exponentially growing amounts of these data and associated information in the scientific literature is presenting one of the most demanding current challenges to both biologists and information technologists. It is as a response to this challenge to develop improved tools for information management that the ORIEL project [3] was conceived. ORIEL is an acronym for Online Research Information Environment in the Life Sciences. In architecture, an Oriel window is one that projects from an upper floor of a building. Such windows were popular in buildings of the late Gothic and Tudor periods and offered a particularly good view of their surroundings.

With EMBO, the European Molecular Biology Organisation, as coordinator, the project combines the skills of a number of partner institutions and organisations in Europe (Table 1), with focus on development of:

• technologies that effectively link a wide range of biological digital information sources, including molecular, genomic and multi-dimensional image databases thereby facilitating cross-database navigation
• methods that implement novel technologies for information extraction and representation, thereby encouraging creative exploration and discovery within the information landscape
• toolkits that facilitate the construction of adaptive interfaces, thereby meeting the needs of individual users for presentation of data and visualisation of information

Project Components

Individual components of the project consist of:

• The BioImage Database: building on progress made in an earlier EU-funded project (European Commission IV Framework Biotechnology Infrastructure Project BIO4-CT96-0472 "BIOIMAGE: A European Database of Biological Volume Images), this will be a repository for multi-dimensional biological images [4]. Members of the Image Bioinformatics Laboratory (University of Oxford) together with members of Ingenta's ORIEL project team will work to implement existing and emerging standards for both multimedia and semantic web technologies which will allow efficient search and retrieval of scientific images and their integration with relevant information from other datasets
• Tools for knowledge representation and transformation: work is being undertaken by research groups at IGH and INRIA (Montpelier) to define strict standards of representation and exchange of the biological factual information (genome annotations, metabolic pathways, gene interaction networks) and to develop tools for these processes
• Integration of gene mining toolkits and interfaces to literature: starting from principles developed within the Genebuilder package [5] research groups at the Istituto di Tecnologie Biomediche Avanzate and Istituto di Biologia Cellulare (CNR) will develop within ORIEL toolkits and modules that will enhance the data mining capabilities of programs used for gene and protein structure/function prediction and facilitate automatic searches of gene relationships in the published literature. Artificial neural networks and other non-parametric pattern recognition methods will be further developed and applied in text categorisation and data mining
• Improved information retrieval from literature articles; linkage to factual databases: a considerable portion of our knowledge of living organisms is present in free text form in the published literature. To be useful, this information needs to be incorporated into databases and associated with the corresponding factual data. Although readily assimilated (within limits) by the human brain, the freedom of natural language makes computer extraction of information from this rich source a very complex matter. Given the dramatic increase in genomics-related technologies, one particularly pressing problem is that of unambiguously recognising gene symbols within scientific text. Researchers have seldom followed logical rules in the construction of these symbols. The result is that name ambiguities abound, caused by use of synonyms (different proteins in the same organism with the same name), homonyms (proteins in different organisms that have the same name, but often with different function) and orthographic variations (e.g. IL 6, IL-6, IL6) that are difficult to deal with in a general way. Building on prototype analytical methods developed by the Protein Design Group of the National Center of Biotechnology (CNB), Madrid [6] [7] and drawing on both established and novel technologies in the field of text processing in collaboration with the Bio-Semantics Group of the Department of Medical Informatics, University of Rotterdam [8], a library of software tools for literature analysis in the biological sciences will be developed and applied to semantic interpretation of literature articles and database records
• More efficient information retrieval by implementation of user-adaptable interfaces and task-oriented agents: working together with other partners in ORIEL, LIRMM will develop modules that aim at enrichment of traditional Knowledge Representation (KR) tools with machine learning. The aim will be to improve existing ontologies through implementation of rules gained from experience. In parallel, a multi-agent platform based on a model for a social structure will be developed. The agents are task-oriented; they make use of ontologies to perform activities such as modelling and simulation
• Streamlined user navigation facilitated through application of user passport technology: based on the premise that content owners will continue to exert access controls on information resources maintained by them, the project will initiate the development of 'passport' technologies to allow users to navigate from one service to another while maintaining their user details and rights information. This part of the project will be carried out by Ingenta and in the first instance, the technology will be tested and evaluated within the context of The BioImage Database, since in many instances this will hold only the relevant metadata for images stored at their original locations

ORIEL and the E-BioSci platform

A unique feature of the ORIEL project is its degree of integration with the E-BioSci initiative [9] that is funded by the EC as a research infrastructure under its Quality of Life Programme (contract no. QLRI-CT-2001-30266). The E-BioSci platform aims to provide services that will help biologists to find and bring together many different kinds of digital information. It will focus primarily on interlinkage between genomics-related, image and molecular datasets and the relevant research literature, implementing novel technologies for retrieval as these become available. Given the complexity of the biological knowledge base, solutions to many of the issues involved will require significant research effort. ORIEL will play an important role in focusing and inputting such effort during this developmental process. Conversely, the E-BioSci platform will offer ORIEL partners opportunity for rigorous testing of prototype toolkits and other project components in a controlled, live environment, with critical feedback from selected groups of test users.

Conclusion

As mentioned in the introduction to this brief overview, biology is undergoing a revolution. Driven by advances in technology and huge amounts of new information, this revolution is dramatically changing the way in which researchers formulate and test concepts and make their experimental data known to the research community. The tools developed within ORIEL will help biologists meet the challenge of turning these data into useful knowledge.

References

1. Morowitz, H. (1985) Models for biomedical research: a new perspective,
   Committee on Models for Biomedical Research, National Academy Press
2. EMBL nucleotide database statistics,
   URL: < http://www3.ebi.ac.uk/Services/DBStats/>
3. ORIEL,
   URL: < http://www.oriel.org/>
4. BioImage,
   URL: < http://www.bioimage.org/>
5. Milanesi, L., D. D'Angelo, and I.B. Rogozin (1999) GeneBuilder: interactive in silico prediction of gene structure,
   Bioinformatics, 15, 612-21.
6. Blaschke, C., et al. (1999) Automatic extraction of biological information from scientific text: protein-protein interactions,
   in Proc Int Conf Intell Syst Mol Biol. p. 60-7.
7. Blaschke, C. and A. Valencia (2001) The potential use of SUISEKI as a protein interaction discovery tool,
   Genome Inform Ser Workshop Genome Inform, 12, 123-34.
8. van Mulligen, E.M., et al. (2000) Facilitating networks of information,
   in Proc AMIA Symp. p. 868-72.
9. E-BioSci ,
   URL: < http://www.e-biosci.org/>

Author Details

Les Grivell
ORIEL Project Coordinator
European Molecular Biology Organisation
Meyerhofstrasse 1
69117 Heidelberg
Germany

URL: <http://www.oriel.org/>
URL: <http://www.e-biosci.org/>
Email: les.grivell@embo.org
Phone: +49 6221 889 1501
Fax: +49 6221 889 1210

Les Grivell is manager of EMBO's Electronic Information Programme and coordinates both ORIEL and E-BioSci projects.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

For citation purposes:
Grivell, L. "The ORIEL Project - biological information management from a new perspective", Cultivate Interactive, issue 8, 15 November 2002
URL: <http://www.cultivate-int.org/issue8/oriel/>

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Related articles:
If you would like to view similar articles to this one click on a key word below:

< - ORIEL - life sciences - biology - information management - genomics - bioinformatics - multi-dimensional images - sequence databases - text retrieval - ontologies - >

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Copyright ©2000 - 2001 Cultivate. | Published by UKOLN | Design by ILRT | Contact Us