Dissecting Flux Balances to Measure Energetic Costs in Cell Biology: Techniques and Challenges

Abstract

Life is a nonequilibrium phenomenon: Metabolism provides a continuous supply of energy that drives nearly all cellular processes. However, very little is known about how much energy different cellular processes use, i.e., their energetic costs. The most direct experimental measurements of these costs involve modulating the activity of cellular processes and determining the resulting changes in energetic fluxes. In this review, we present a flux balance framework to aid in the design and interpretation of such experiments and discuss the challenges associated with measuring the relevant metabolic fluxes. We then describe selected techniques that enable measurement of these fluxes. Finally, we review prior experimental and theoretical work that has employed techniques from biochemistry and nonequilibrium physics to determine the energetic costs of cellular processes. Expected final online publication date for the Annual Review of Condensed Matter Physics, Volume 14 is March 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Boosting Fitness Costs Associated with Antibiotic Resistance in the Gut: On the Way to Biorestoration of Susceptible Populations

“…Fitness costs associated with the expression of antibiotic resistance mechanisms and/or with the carriage of mobile genetic elements could be unbearable for certain resistant bacterial populations, favoring their replacement with the susceptible ones. Stated another way, given that life is a nonequilibrium phenomenon, we propose to act against resistance by modifying energetic flux balances [102], thus altering the relative fitness costs of susceptible and resistant populations. As shown in Figure 1, the supply of energy (ATP-producing processes) has a cost, which is balanced by the benefits of energy investment (ATP-consuming processes), resulting in bacterial replication.…”

Making Sense of Neural Networks in the Light of Evolutionary Optimization

“…Nonequilibrium thermodynamics offers a different approach to this problem, providing general mechanisminsensitive expressions that may turn out to be relevant for a general theory of neural networks. 16,18,19,23,31,32,[39][40][41][42][43] A nervous system is dissipative and, as any dissipative system, its thermodynamic cost can be expressed as the rate of entropy production 𝑆 ̇𝑡, which can be calculated in the steady state as…”

New Restrictions on AI from Physics: The Most Reliable Way to Predict AGI future?

“…While fitness is a biological concept and cannot be directly applied to artificial neural networks, a growing understanding is linking fitness to Nonequilibrium Thermodynamics concepts, such as the growth or reproduction rates, entropy production rate, energy flows, dynamic stability or resilience and other related quantities. [51][52][53][54][55][56][57][58][59][60][61] Contrarily, previous literature on fundamental theory of neural networks assumed models minimized other functionals, for example, root-mean-square error of predictions. [62][63][64][65][66] We expect that fitness and its generalizations within Nonequilibrium Thermodynamics will provide a more solid basis for defining an AGI goal function than those previously proposed.…”

“…50 By interpreting the entropy production rate as the cost of a dissipative process, successful attempts have been made in the literature to apply this perspective to several biological systems. [51][52][53][54][55][56][57][58][59][60][61] However, this approach is yet to be used with a nervous system. Extending this to a neural network is not straightforward, as we have previously demonstrated.…”