Graphical rule-based representation of signal-transduction networks

• Authors: James R. Faeder;Michael L. Blinov;William S. Hlavacek
• Venue: Proceedings of the 2005 ACM symposium on Applied computing
• Year: 2005

The process by which a cell senses and responds to its environment, as in signal transduction, is often mediated by a network of protein-protein interactions, in which proteins combine to form complexes and undergo post-translational modifications, which regulate their enzymatic and binding activities. A typical signaling protein contains multiple sites of protein interaction and modification and may contain catalytic domains. As a result, interactions of signaling proteins have the potential to generate a combinatorially large number of complexes and modified states, and representing signal-transduction networks can be challenging. Representation, in the form of a diagram or model, usually involves a tradeoff between comprehensibility and precision: comprehensible representations tend to be ambiguous or incomplete, whereas precise representations, such as a long list of chemical species and reactions in a network, tend to be incomprehensible. Here, we develop conventions for representing signal-transduction networks that are both comprehensible and precise. Labeled nodes represent components of proteins and their states, and edges represent bonds between components. Binding and enzymatic reactions are described by reaction rules, in which left graphs define the properties of reactants and right graphs define the products that result from transformations of reactants. The reaction rules can be evaluated to derive a mathematical model.