Optimal Prediction by Cellular Signaling Networks

Abstract

Living cells can enhance their fitness by anticipating environmental change. We study how accurately linear signaling networks in cells can predict future signals. We find that maximal predictive power results from a combination of input-noise suppression, linear extrapolation, and selective readout of correlated past signal values. Single-layer networks generate exponential response kernels, which suffice to predict Markovian signals optimally. Multilayer networks allow oscillatory kernels that can optimally predict non-Markovian signals. At low noise, these kernels exploit the signal derivative for extrapolation, while at high noise, they capitalize on signal values in the past that are strongly correlated with the future signal. We show how the common motifs of negative feedback and incoherent feed-forward can implement these optimal response functions. Simulations reveal that E. coli can reliably predict concentration changes for chemotaxis, and that the integration time of its response kernel arises from a trade-off between rapid response and noise suppression.

5 pages, 4 figures; 15 supplementary pages with 12 figures

References in corpus (4)

• Spectral solutions to stochastic models of gene expression with bursts and regulation
• A stochastic spectral analysis of transcriptional regulatory cascades
• Information and fitness
• Telling time with an intrinsically noisy clock

Cited by in corpus (27)

• Nonequilibrium Physics in Biology
• E. coli chemotaxis is information-limited
• How a well-adapting immune system remembers
• Biochemical machines for the interconversion of mutual information and work
• Modeling somatic computation with non-neural bioelectric networks
• Receptor crosstalk improves concentration sensing of multiple ligands
• Trading bits in the readout from a genetic network
• Information thermodynamics for interacting stochastic systems without bipartite structure
• Noise Filtering and Prediction in Biological Signaling Networks
• What we learn from the learning rate
• A connection between bacterial chemotactic network and optimal filtering
• Robustness of clocks to input noise
• The behaviour of information flow near criticality
• Optimal entrainment of circadian clocks in the presence of noise
• The accuracy of telling time via oscillatory signals
• Biochemical Szilard engines for memory-limited inference
• Multi-dimensional biochemical information processing of dynamical patterns
• Optimal information transfer in enzymatic networks: A field theoretic formulation
• Theory for Optimal Estimation and Control under Resource Limitations and Its Applications to Biological Information Processing and Decision-Making
• Fluctuation Theorem of Information Exchange between Subsystems that Co-Evolve in Time
• Resource Limitations induce Phase Transitions in Biological Information Processing
• A biological circuit to anticipate trend
• Diffusion-dynamics laws in stochastic reaction networks
• Fluctuation Theorem of Information Exchange within an Ensemble of Paths Conditioned on Correlated-Microstates
• Chemotaxing E. coli do not count single molecules
• Theoretical Analysis of Resource-Induced Phase Transitions in Estimation Strategies
• The Inertial Spin Model of flocking with position-dependent forces