Cells, especially those of a complex multicellular organism, have to act and react to
each other and to external influences in a well concerted manner. Thus, if we want to
understand cellular behaviour and its responses to external signals, or want to
influence it in a predictable manner, we have to understand the pathways through
which these signals are mediated into and within the cell. Knowledge about the
principle mechanisms of signal transduction and regulation mechanisms of individual
macromolecules in signaling pathways has multiplied in the last decade. It is now
growing at a rate that makes it difficult to keep up with [Krauss, 1997]. In most cases
changes in cell behaviour involve the execution of transcriptional events, which are
specific for each signal in its cellular context [Hill and Treisman, 1995]. Biological
signaling pathways also interact with each other to form complex networks. These
networks show emergent properties like signal integration accross multiple time
scales or selfsustaining feedback loops, which are not present in the isolated
pathways [Bhalla and Iyengar, 1999].
The huge and ever more rapid growing amount of signal transduction data demands
for a database that stores and organizes this knowledge, providing simple and fast
access to the information. The complexity created by the crosstalk between pathways
makes it virtually impossible to infer by hand all the consequences that follow after
one modifies one part of the network. To this end, computeraided simulation will
have to be used. It can only be successful on the basis of a comprehensive and
detailed dataset.
At the moment, there are at least two databases providing information on signal
transduction in general which are publicly available worldwide, CSNDB and KEGG.
CSNDB (http://geo.nihs.go.jp/csndb.html) specializes in the semantic view of
signaling pathways, incorporating a pathway viewer [TakaiIgarashi and Kaminuma,
1998]. KEGG (http://www.genome.ad.jp/kegg/kegg.html) focuses on metabolic
pathways, but also contains some signal pathways. These consist of maps of
interacting molecules or genes. The signal pathways do not contain data on the
reaction mechanisms. [Ogata et al., 1998].
TRANSPATH (http://transpath.gbf.de) is an
information system on gene regulatory pathways, and a member of the TRANSFAC
family of databases [Heinemeyer et. al., 1999]. The core functionality of the
information system is to provide a structured repository for pathwayrelated data.
Also the data should be fit to be used in signaling network simulations.
TRANSPATH focuses on pathways involved in the regulation of transcription factors
in different mammalian species, mainly human, mouse and rat, but aims at a
comprehensive data collection. Elements of the relevant signal transduction pathways
like hormones, receptors, enzymes and transcription factors are persistently stored
together with information about their interaction and references in an objectoriented
database. Objects are a natural choice to represent those elements as data. Molecules
with similar structure and function can be grouped in families, providing a way to
abstract their common signaling behaviour.
Interactions are modeled as reactions with reactants and products and a single
enzyme. To enable the system to be used as the basis for simulation it is neccesary to
include rate constants in the reactions and different entries for different states of a
molecule. Alternatively to this mechanistic view, interactions can be stored as
activation and inhibition pointers providing a semantic view (similar to that provided
by CSNDB) which corresponds to the schematic drawings familiar from the literature.
Queries to the database are conducted via the Internet by submitting names of
transcription factors or other signal molecules. The user can choose to view either the
encyclopaedic information for the requested molecule or the reaction cascades
starting from the molecule. All information is validated with references to the
original publications. Also, references to other databases are provided (TRANSFAC,
CYTOMER, SwissProt, EMBL, PubMed and CSNDB). The TRANSPATH interface
has also been used as a gateway to the CSNDB data.
This work has been supported by a grant of the German Ministry of Education, Science, Research and Technology (BMBF; 01 KW 9629/7). The complete data set of CSNDB was generously provided by T. TakaiIgarashi.