Infrastructure for Open Source Modeling

DESCRIPTION (provided by applicant): Our knowledge of the combinatorial complexity and context-dependency of signaling pathways is both fragmented and rapidly evolving. The scientific community needs to organize this knowledge in a way that can be conveniently shared, discussed, and revised, and that will support quantitative modeling efforts through collaboration. Achieving this goal starts with a formal language that can represent mechanistic details of protein-protein interactions at a level that molecular biologists find natural. We will create a web-based environment in which users build, share, discuss and revise a database of molecular parts and interaction rules expressed in a concise, visual, detailed, and executable language. We will seed this environment with a state of the art description of the EGFR pathway to illustrate and test the capabilities of our platform and tools. If successful, this environment will grow into a dynamic, community-curated knowledge base of lasting relevance that can be extended to any pathways of interest. Our knowledge and understanding of molecular networks underlying complex diseases is rapidly evolving while remaining highly fragmented across large industrial and academic research communities. Models will greatly enhance our reasoning about such networks through precision and explication, thereby enabling computer-assisted storage, retrieval, critique, evaluation, communication and manipulation of formalized stories. We will provide a wiki-based platform to represent and edit biological knowledge in an executable fashion that lends itself to the collaborative construction and critique of models, thereby giving voice to data.