Benjamin Kuznets-Speck with Michael Hinzcewski Andrew Eckford and Peter Thomas

Cut your losses:
A rate-distortion-thermodynamic approach to molecular communication Benjamin Kuznets-Speck

Molecular communication is a central facet of problems ranging from deciphering the principles governing physiological signaling (occurring constantly in cellular and neuronal transduction networks), to optimization in synthetic biology, and foundational questions in evolutionary biology. These problems are often illuminated by a quantitative description of the information transmitted in a given system and the relationship between said systems characteristic timescales and its efficiency of information processing. Information theory provides an abstract formalism in which to study communication channels, while thermodynamics accounts for the physical constraints inevitably present in biological information engines, though a general unification of these dual theories remains incomplete. In particular, when measuring the cost of reliable information transfer, there exists two separate, and possibly compatible, perspectives: rate-distortion theory dictates a specific functional form that cost be measured in for a channel to perform optimally while thermodynamics requires this cost be framed by an entropy production. In this study, we will explore both of these formalisms for a model of synaptic signaling and examine the physiological consequences of unifying these two perspectives using a technique known as probabilistic matching.