A Pathway DB: Annotating Signal Transduction Pathways with bio-processes using hierarchical multi-layered structures.

Ken Ichiro Fukuda1, Yuki Yamagata2, Toshihisa Takagi
1fukuda-cbrc@aist.go.jp, CBRC, AIST; 2snowfox@hgc.jp, BIRD, JST

This presentation describes our ongoing project that formalizes Signal Transduction Pathways (STPs) and other functional relation knowledge into a database from scientific literatures. STPs explain the underlying mechanisms of various biological phenomena in terms of interactions of biological processes and bio-molecules. Due to the diversity of topics that the term "signal" covers, the kind of biological entities that constitutes STPs is highly diverse and ranges from metal ion, DNA sequences, and protein to biological concepts such as cellular responses in general. In addition, the meaning of interactions that relate biological entities are likewise diverse, including causality relations, mediations, transportation of proteins, chemical reactions, or secretion of ions. Typically, these heterogeneous concepts are mixed-up in a single diagram. A simple graph theoretical analogy to these pathways is not suitable or sufficient to handle the knowledge described in scientific articles and a strongly structured ontology-aware representation have to be developed. Besides, as biological pathways can be decomposed into arbitral levels that refer to processes of different granularities, a recursively decomposable representation structure is needed. In this project, we adopted a compound graph based data model, which allows us to describe pathways in a hierarchical way. The database focuses on annotating pathways or sub-pathways according to their related biological processes and taxonomies of biological entities such as molecular families, cellular localizations, and species. Every process, sub-process, interaction or element in a pathway has a pointer to a specific category of the taxonomies or ontologies, such as GO, and one can query mechanisms of (sub-) processes, interactions, molecules by using them. The database is an XML database with server-side and client-side applications, which includes a pathway diagram database, a set of ontologies, a web interface, and clients for data registration.