Limits on the computational expressivity of non-equilibrium biophysical processes
Many biological decision-making tasks, such as those involved in processing the glycan code, require classifying high-dimensional chemical states into one of potentially hundreds of different categories. The biophysical and computational mechanisms that enable these feats of classification remain enigmatic. In this work, using Markov jump processes as an abstraction of general biochemical networks, we reveal several unanticipated and universal limitations on the classification ability of generic biochemical processes. These limits arise from a fundamental non-equilibrium thermodynamic constraint that we derive. Importantly, we show that these limitations can be overcome using common biochemical mechanisms that we term “input multiplicity,'' examples of which include enzymes acting on multiple targets. Analogous to how increasing depth enhances the expressivity and classification ability of neural networks, our work demonstrates how tuning input multiplicity can potentially enable an exponential increase in a biological system's ability to classify and process information. Our non-equilibrium thermodynamic analysis places broad limits on the flexibility and sharpness of decision boundaries, as well as on the number of different classes these networks can distinguish. These findings shed new light on how cells interpret and understand their environment through chemical reaction-based information processing.
Hosted by Professor Gene Hall