Society for Mathematical Biology nautilus logo

International Conference on Mathematical Biology and

Annual Meeting of The Society for Mathematical Biology,

July 27-30, 2009

University of British Columbia, Vancouver

.

Program

Poster PS27B
Theodore Roman
Case Western Reserve University
Title In Silico Modeling of the Wnt Signaling Pathway
Abstract The Wnt signaling pathway plays important roles in multicellular development and human cancers. A better understanding of the components and interactions of this signaling pathway will enable better modeling of these processes, also potentially resulting in the development of novel medical treatments for various cancers. The Wnt pathway is better understood as a network of interactions rather than a linear progression of events. While some of the key players in this pathway and their interactions are well characterized, knowledge of many components of the pathway is still limited. Building on this existing knowledge of Wnt signaling, we aim to construct a comprehensive model of the Wnt signaling network in order to illuminate potentially novel pathway regulators and network motifs. Compared to existing models, our model will be more comprehensive because we will curate and integrate multiple types of data from multiple sources including: gene-expression data (e.g. microarrays and RNAi screens), protein-DNA binding data (e.g. transcription factor binding motifs), high-throughput protein-protein interaction data (e.g. affinity purification with mass spectrometry), and low throughput information from the literature. Literature references provided us with a well-annotated core of Wnt pathway interactors. Our own affinity purification-mass spectrometry (AP-MS) data and other published protein-protein interaction (PPI) data has provided us with a list of potential Wnt pathway interactors. We have developed an in silico model of the pathway from the core to enrich our understanding of these potential Wnt pathway interactors. This model consists of three layers: the core set of seed proteins--first-level proteins; all direct interactors with the core--second-level proteins; and all direct interactors with the second-level proteins. We scored the non-core nodes of the network based on the length of the shortest path to the core, degree of connectivity to the core, and promiscuity. To ensure our scoring did not occur simply due to chance, we compared the Wnt modeled network to a collection of randomized networks, which were generated by preserving the topology and degree distribution of the original human PPI network. In our scheme, a high score for a node indicates that the node (protein) is highly connected to only the known Wnt signaling proteins, and therefore may play a role in Wnt signaling. This new in silico method of building a select set of candidate proteins to be investigated in the wet lab will result in a more focused, more efficient experimental investigation of potential Wnt signaling proteins.
CoauthorsAlex Galante, Sudipto Saha, Mehmet Koyuturk, Rob Ewing
LocationWoodward Lobby (Wednesday-Thursday)