Joint CMU-Pitt Ph.D. Program in Computational Biology/Department of Computational and Systems Biology Seminar


Understanding the limits of information transmission in cellular signaling networks

Signal transduction networks allow cells to interpret and respond to environmental changes.  These networks are subject to various forms of noise, including the intrinsic biochemical noise due to the stochasticity of reaction events.  Information theory enables quantification of how much information a signaling network is capable of transmitting in the presence of noise.  In some cases, the noise in these networks prevents individual cells from making reliable decisions.  For example, the signaling network responsible for mediating apoptosis transmits less than the 1 bit of information needed to reliably respond to an apoptotic signal.  Interestingly, the noise in this network also enables a graded, fractional survival in a population of cells, enabling transmission of over 3 bits of information to the population.  However, not all signaling networks are so limited in supporting decision making for individual cells.  Upon developing a consistent methodology for calculating information theoretic quantities from dose-response data, we systematically analyzed a number of common signaling motifs to gain an understanding of the limits of intracellular information transmission.  Stochastic models of the simplest networks can transmit over 6 bits of information, and even complex multi-level kinase cascades can encode significantly more information than has been observed in vivo.  Our work suggests that the low levels of information transmission in eukaryotic signaling networks are not the result of biophysical limitations.  Indeed, more work is needed to understand the sources of noise in signaling systems and what purpose they may serve in regulating cellular information processing in metazoan organisms.

Faculty Host: James Faeder (Pitt)

For More Information, Please Contact: