Special Issue: Artificial Intelligence in the Life Sciences
Guest Editors:
Prof Werner
Dubitzky, University of Ulster, Coleraine, Northern Ireland, URL
Dr Francisco
Azuaje*, University of Dublin, Trinity College Dublin, Ireland
Journal Editor:
Prof Paul Mc Kevitt, University
of Ulster, Magee, Derry/Londonderry, Northern Ireland
*
Dr Azuaje will shortly be joining the University of Ulster
Table of Contents
Call for Papers
Databases
Call for Papers
Bioinformatics and Life Science Informatics
are emerging disciplines, which are concerned with the organization, modeling,
analysis, and interpretation of data and information arising from research
and development within the life sciences. A great deal of this work is
directed towards understanding the structure and function of proteins,
cells, the genome, and the process of natural evolution. The considerable
"algorithmic complexity" of the underlying biological systems requires
a huge amount of detailed information for their complete description. Very
rapidly, vast amounts of biological information are becoming available
in public information repositories, including
DNA/RNA and protein sequence and structure databases, metabolic and signaling
pathway databases, biomedical literature repositories, image databases,
and many others. The key to advances in life science research and development
lies in these information repositories and future biological experiments
and studies.
The principal approach to analyzing,
modeling, and interpreting biological data is to abstract them into logical
structures that support and incrementally promote the development of a
more general conceptual framework for characterizing, explaining, and predicting
processes in living systems. The available data and the need for making
sense of them presents vast opportunities for the application of a wide
range of artificial intelligence techniques, ranging from intelligent image
analysis, machine learning, data mining, and text mining to knowledge representation
and management, automatic reasoning, uncertainty management, and computational
creativity. So far, these opportunities have not been extensively exploited.
This Special Issue is intended
to draw together those who are developing and applying artificial intelligence
methods to bioinformatics and life science informatics problems. Submissions
are particularly encouraged from those developing and applying artificial
intelligence techniques to genome, transcriptome, proteome, and metabolome
research, systems biology, drug discovery/design, neuroinformatics, and
artificial life research. With this Special Issue we hope to raise the
awareness and appreciation of artificial intelligence techniques in the
life science and bioinformatics communities. It is likely that this Special
Issue will also be produced as a book.
Submissions of full papers in
RTF, MS WORD, PDF or POSTSCRIPT format accompanied with contacting author
name/affiliation should be sent to w.dubitzky@ulster.ac.uk
before January 1, 2003. Papers should be no more than 30 printed pages
(approximately 15,000 words) with a 12-point font and 18-point spacing,
including figures and tables. Full details on the Artificial Intelligence
Review (including instructions for authors) are available at http://www.wkap.nl/journals/aire.
Important Dates:
Papers due: January 1, 2003
Acceptance notification: March
1, 2003
Final manuscript due: June 1,
2003
Date of issue: Summer/Autumn
2003
Download this Call for Papers
as PDF file.
Life Science Data and Information Repositories
One reason, why we believe the life
science constitute a formidable challenge and scenario for artificial intelligence
research and applications is the rich global information ecosystem
that comes with the life sciences. In this environment we distinguish three
types of information/knowledge resources -- databases,
information
bases, and knowledge bases. Some of these
resources can be reached via the links below.
Databases
Historically, data and information
storage/access technologies within the life sciences are generally taken
for granted. However, new scientific instruments generate new types of
experimental
data. These data must be modeled and integrated into the existing conceptual
and technical infrastructure. With the fast-growing globalization of life
science, a wide range of research issues for data handling and management
in the life sciences emerges – data models, data annotation, Grid technologies,
computer networking, distributed and federated databases, multimedia data
stores (signals, images, video), data security, real-time access, data
quality, data compression, data visualization, and more.
Information
Bases (or Summary Databases)
Not too long ago, data and information
analysis and modeling was mostly carried out by the life scientists themselves
or by statisticians, either manually or by using simple software. Today's
life scientists are faced by four phenomena that render their classical
approach increasingly difficult, if not impossible. First, the physical
data volumes and the logical data structures are becoming increasingly
complex. This is mainly due to the detail with which the underlying life
processes are measured. For example, functional magnetic resonance image
(fMRI) scans of the brain often result in several gigabytes of image data.
Second, the available machinery of analysis and modeling methods and tools
is incredibly large and complex. Modern statistics, machine learning, data
mining, soft-computing, artificial intelligence, and algorithmic approaches
are nowadays employed to tackle the analytical challenge posed by the experimental
data and the complexity of the underlying phenomena. Third, in trying to
make sense of the experimental data, it is increasingly necessary to carry
out the analysis in the light of what information and knowledge (see knowledge
bases below) already exists about the phenomena under study. This information
is held in information bases around the world – scientific article resources
(paper-based or electronic), ontology and thesaurus systems, information
systems like DNA, protein, and brain databases, Web sites, and so on. These
systems contain summarized information, as opposed to experimental or observational
data. Approximately one thousand information repositories of this kind
exist in the field of biology alone. The unaided human brain stands no
chance in processing these resources with sufficient precision and speed.
Fourth, due to the global proliferation of life science information bases,
emerging Grid solutions are becoming increasingly important. The Grid dimension
adds yet more conceptual and technological “baggage” to the already highly
intricate life science scenario.
Knowledge
Bases
In addition to data and information
repositories, there are a growing number of executable knowledge-based
systems and simulation systems that are added to the global
pool of life science knowledge. Such systems capture (a) the knowledge
of a well-defined scientific domain in the life sciences, or (b) the systemic
properties of the biological entity or system of interest. They differ
from data and information repositories in that they are able to respond
intelligently to queries falling within the jurisdiction of their knowledge
domain. Such systems are also called computational theories or models,
which are based on methods from the field computational intelligence and
numerical simulation respectively. Systems biology, for example, is a new
scientific discipline attempting to develop computational models for different
biological systems ranging from sub-cellular systems and cells to entire
organs and organisms. It is expected that systems of this kind are becoming
commonplace and develop into a “bread-and-butter” instrument for life scientists.
At the far end, we envisage that the creative element in life science is
more and more supported by computational approaches capable of generating
novel, useful, and innovative ideas and hypotheses. Such systems could
be called intuition bases.
[ Back to Top ]