Comprehensive analysis of gene regulatory pathways and mechanistic model building through an integrated computational platform for Saccharomyces cerevisiae.
Xufei Qian (1), Shubhada Godbole (2), Michael Baitaluk (1), Erdem Kurul (1), Amarnath Gupta (1), Animesh Ray (2)
(1) San Diego Supercomputer Center, University of California San Diego, La Jolla, California;
(2) Systems Biology, Keck Graduate Institute, 535 Watson Drive, Claremont, CA, 91711, USA
A challenge in post-genomic biology is to develop an integrated computational platform that allows intelligent retrieval and analysis of all genomic-scale and single-gene information. Current databases do not allow this ability because of several difficulties, including those of integrating heterogeneous data types, non-standard annotation, problems with updating of existing databases, and because of disparate confidence levels of different data. A solution has been to provide a common web interface to specialized databases for sequential analysis. Here we describe an integrated computational platform that dynamically assimilates high and low throughput data from MIPS, SGD, TRANSFAC, BIND, preBIND, datasets for global protein-protein, protein-DNA and genetic interactions, and microarray-based gene expression. Over this integration platform, we have developed a query engine that allows graph queries. Using this, for example, we can generate an integrated network of protein-protein, protein-DNA and genetic interactions for all genes that show a specific mRNA induction pattern during a defined period in sporulation, analyze selected subgraphs of this integrated network using Gene Ontology classification for functional module discovery and analyze statistical associations among network components by data-mining. Thus we have developed a powerful knowledge-discovery tool that allows both 'hypothesis-first' and 'data-mining' approaches on the same computational platform.
Return to YGM 2004 Home at SGD