Editors' ChoiceSystems Biology

Decisions, Decisions

+ See all authors and affiliations

Science Signaling  19 Feb 2008:
Vol. 1, Issue 7, pp. ec63
DOI: 10.1126/stke.17ec63

By connecting cellular response systems to external and internal stimuli, complex signaling networks appear to provide not only a communication system but also an information processing system. Such decision-making on the basis of sophisticated pattern recognition of noisy inputs is not unlike human face recognition--a task relatively easy for the human brain but not trivial for a computer. Helikar et al. provide evidence that cellular signaling systems do indeed have the necessary characteristics for such complex information processing. The authors constructed a Boolean model of signal transduction by receptor tyrosine kinases, G protein (heterotrimeric guanine nucleotide-binding protein)-coupled receptors, and integrins. They derived the logical instruction sets (the rules that govern activity of a particular protein or protein state, given the states of its interacting regulators) for 130 protein nodes in the model network from a set of 800 papers and then sampled the response of the system to 10,000 random inputs. They found that noisy random inputs were partitioned into a limited number of global cellular responses that appeared to be biologically relevant. Thus, the system demonstrated a basic characteristic of pattern recognition machines. The observations support the reasoning that characteristics of complex cellular signaling systems are not an accident of evolution but rather reflect the necessary structure of a pattern recognition system that allows cells to sample random inputs from hormones and growth factors and the like and then to decide on an appropriate response that ensures or enhances organismal survival.

T. Helikar, J. Konvalina, J. Heidel, J. A. Rogers, Emergent decision-making in biological signal transduction networks. Proc. Natl. Acad. Sci. U.S.A. 105, 1913-1918 (2008). [Abstract] [Full Text]

Related Content