Pith. sign in

REVIEW 3 major objections 9 minor 73 references

Ecosystems as adaptive living circuits

T0 review · 3 major / 9 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that living circuits—adaptive networks that set their own energy coupling—switch abruptly from complete collapse to a functioning dissipative state once the driving potential crosses a critical threshold, in contrast to…

desk verdict Clean theory paper: adaptive dissipation gives a first-order transition to a dissipative state, but the central growth rule is postulated and the ecosystem interpretation is conditional on it. read the letter →

arxiv 2506.22017 v1 pith:TFZ2LFOL submitted 2025-06-27 q-bio.PE cond-mat.dis-nncond-mat.stat-mech

Ankit Dhanuka , Avi I. Flamholz , Arvind Murugan , Akshit Goyal This is my paper · ORCID
classification q-bio.PEcond-mat.dis-nncond-mat.stat-mech MSC 92D4082C31
keywords livingcircuitsadaptivedissipationnonequilibriumsteadystatephasetransitionmicrobialecosystemsredoxmetabolismsavetheweakestmaximumprinciple
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to show that adaptive systems—systems that control their own coupling to an energy source—cannot be described by the usual framework of driven nonequilibrium physics. Using microbial ecosystems as a model, the authors map the flow of electrons through redox metabolism onto an electrical circuit whose edge conductances are species abundances that grow with the energy they dissipate and decay with a maintenance cost. They find that such "living circuits" undergo a first-order phase transition: below a critical driving potential the circuit collapses to an equilibrium state with no dissipation, while above it the circuit jumps into a nonequilibrium steady state that dissipates energy. If this is right, ecosystems can switch abruptly from death to a functioning, energy-consuming state through purely local rules, and they may do so at near-maximal dissipation without any global objective.

What carries the argument

The machinery is the mapping of a microbial ecosystem onto a Markov chain and electrical circuit, combined with the adaptive conductance rule $\frac{d\Lambda_{ij}}{dt} = \sigma_{ij} - \sigma_{\rm maint}\Lambda_{ij}$, where $\sigma_{ij} = (W_{ij}p_j - W_{ji}p_i)\log(W_{ij}p_j/W_{ji}p_i)$ is the dissipation on the edge and $\sigma_{\rm maint}$ is a fixed per-capita maintenance cost. Species abundances become edge conductances, the master equation for electron densities becomes Kirchhoff's current law, and autotrophs act as batteries that break detailed balance with driving potential $\mu$. The load-bearing identity is the criticality condition $(e^{\mu_{\rm crit}}-1)\cdot \mu_{\rm crit} = \sigma_{\rm maint}/\chi(k)$, which ties the collapse threshold to the ratio of maintenance cost and an emergent chemical current. The "save the weakest" effect is carried by a quadratic effective potential for a weak edge, $d\Lambda_{\rm weak}/dt \approx \frac{d}{d\Lambda_{\rm weak}}(\tfrac{1}{2}\Xi_{\rm eff}\Lambda_{\rm weak}^2)$, whose curvature changes sign precisely at the phase transition.

What would settle it

A controlled experiment could settle the central rule directly: grow a defined microbial community on a redox pair with known potentials, measure each species' per-capita growth rate against the thermodynamic dissipation on its catalyzed edge across several driving potentials; if growth does not track dissipation, the model's mechanism fails. Alternatively, a closed community with known redox kinetics and maintenance cost that persists at a light drive below the $\mu_{\rm crit}$ computed from $(e^{\mu_{\rm crit}}-1)\mu_{\rm crit} = \sigma_{\rm maint}/\chi(k)$ would falsify the predicted collapse threshold.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a network whose links grow according to the local thermodynamic dissipation on each link—$\frac{d\Lambda_{ij}}{dt} = \sigma_{ij} - \sigma_{\rm maint}\Lambda_{ij}$—does not behave like an ordinary driven circuit. Unlike conventional nonequilibrium circuits, which dissipate for any nonzero drive, living circuits only sustain a nonequilibrium steady state beyond a critical driving potential $\mu_{\rm crit}$, obeying $(e^{\mu_{\rm crit}}-1)\cdot \mu_{\rm crit} = \sigma_{\rm maint}/\chi(k)$, where $\chi(k)$ is an emergent current set by the redox kinetics. Right at the threshold the dissipation jumps discontinuously, signaling a first-order phase transition between complete collapse and a dissipative state in which every surviving edge dissipates more than its maintenance cost. The same local rule produces a "save the weakest" feedback that rescues nearly extinct edges by routing transient dissipation through them, drives the circuit to near-maximal total dissipation without any global optimization principle, and yields richer topologies at higher drive.

Load-bearing premise

The results rest on the assumption that a species' abundance grows in proportion to the thermodynamic energy dissipated on the metabolic edge it catalyzes, minus a fixed per-capita maintenance cost; if real growth tracks biomass yield, resource concentration, or other fluxes instead of dissipation, the phase transition and its consequences could change or vanish.

Editorial extensions

If this is right

  • Any closed ecosystem with insufficient energy input should collapse entirely to an equilibrium state of zero dissipation, rather than persisting at low activity.
  • Above the critical drive, surviving species dissipate more than their maintenance cost, and the final network topology is the same across initial conditions: different starting circuits converge to the same nonequilibrium steady state.
  • Because the weakest edges are transiently the most dissipative, near-extinct species can be rescued by community-level feedback, equalizing dissipation across species and stabilizing the community against perturbations.
  • Locally selfish, dissipation-seeking growth rules drive circuits to near-maximal total dissipation and, at higher drive, to more complex surviving topologies, with complexity saturating near $4/N$ for large single-battery circuits.
  • Multiple energy sources (autotrophs) can cooperate or compete depending on their placement, with cooperation being the typical case in random networks.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If real microbial growth follows biomass yield or cross-feeding fluxes instead of dissipation, the sharp phase transition and near-maximal dissipation could persist in modified form or disappear; a yield-based growth rule is the most direct stress test of the framework.
  • The $4/N$ saturation suggests a testable ecological prediction: a single dominant energy source can sustain at most a small fraction of possible metabolic strategies in a large community, so biodiversity should scale weakly with species pool size in energy-limited closed systems.
  • The same local rule could produce collapse-to-function transitions in other adaptive structures, such as neural networks with dissipation-driven synaptic plasticity; the paper motivates but does not test this generalization.
  • The explicit formula for $\mu_{\rm crit}$ could be used quantitatively: with measured redox potentials, maintenance costs, and light input, one could predict which Winogradsky columns or phototrophic mats should collapse versus persist.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 9 minor

Summary. This paper introduces 'living circuits,' a mapping of microbial redox ecosystems onto electrical circuits whose edge conductances Λ_ij (species abundances) adapt according to dΛ_ij/dt = σ_ij − σ_maint Λ_ij, with σ_ij the thermodynamic dissipation on the edge (Eqs. 2–3). The authors report four main results: (i) a first-order phase transition from complete collapse to a dissipative nonequilibrium steady state (NESS) at a critical drive μ_crit satisfying (e^{μ_crit} − 1) μ_crit = σ_maint/χ(k) (Eq. 4); (ii) a 'save-the-weakest' feedback that rescues weak edges; (iii) near-maximal total dissipation reached without global optimization; and (iv) a drive-complexity relationship that saturates at ≈4/N for low-variance kinetics. Support comes from numerical integration of circuits with up to 10 nodes and 45 edges, analytic calculations for n-cycles (Appendices D–J), and an independently parameterized redox-tower model (Appendix C, Fig. S2).

Significance. If correct, the framework is a valuable contribution to nonequilibrium physics of adaptive systems: it provides a tractable model in which the system's distance from equilibrium is itself a dynamical variable, with concrete analytic predictions (Eq. 4; the 4/N complexity saturation; the two-battery phase diagrams) and a mechanism qualitatively opposite to the pruning rules of vascular and Physarum networks. Strengths of the manuscript include the explicit statement of approximations and free parameters, the clean threshold calculation for cycles (Appendix D), the consistency between random-k_ij and redox-tower parameterizations (Fig. S2), and the authors' candid identification of the model's limitations in the Discussion. The main limitation is that all central results are established for a single postulated growth rule, and the paper itself flags flux-based growth rules as more appropriate for many ecosystems; whether the phase transition, save-the-weakest, and near-maximal dissipation survive that change is untested. The quantitative near-optimality claim for complex circuits also depends on an incompletely documented global-maximization computation.

major comments (3)
  1. [Discussion; Eqs. (2)–(3), (6); Appendix H] The ecosystem-level reading of the central results is conditioned on the dissipation-based growth rule of Eq. (2), and the Discussion explicitly concedes that for ecosystems in which a substantial fraction of resource flux is diverted into biomass, 'flux-based local growth rules and more realistic redox chemistry ... may be more appropriate.' No test is given for whether any of the four central claims survive a flux-based rule dΛ_ij/dt = J_ij − σ_maint Λ_ij. I note that the specific stress-test mechanism suggested in a skeptical reading — that the logarithmic factor in Eq. (3) makes per-capita dissipation σ_ij/Λ_ij diverge as Λ_ij → 0 — does not actually occur: σ_ij/Λ_ij = (k_ij p_j − k_ji p_i) log(k_ij p_j/k_ji p_i) stays finite in that limit because the quasi-steady-state p_i are set by the rest of the circuit. The non-monotonicity in Fig. 3b is therefore controlled by the flux per unit conductance, a quantity a flux-based rule also contains, so a threshold and a rescue mechanism may well survive. However, the near-maximal dissipation result (Eq. 6 and Appendix H) is derived specifically from the logarithmic dissipation structure, and it is not evident that a flux-based rule would maximize dissipation rather than total flux. I request either (a) a supplementary numerical test of the flux-based rule covering the claims of Figs. 2–5, or (b) a revision that explicitly conditions the Abstract's claim that the work 'establishes ecosystems as paradigmatic examples of living circuits' on the dissipation-based rule.
  2. [Fig. 4b,d; near-optimal dissipation section; Appendix H] The near-maximal dissipation claim is quantified as '≈ 7x closer to the global maximum' (Fig. 4d) and 'much closer to the max than expected' (Fig. 4b). For the 10-node, 45-edge circuits, the global maximum is a constrained optimization problem over a ~45-dimensional conductance simplex, but the manuscript does not state how the blue histograms were computed — which optimizer, what multi-start strategy, what convergence tolerance, and how the authors verified that the computed value is the global rather than a local maximum. If the blue distribution is produced by a local optimizer, the near-optimality claim could be overstated or initialization-dependent. Please describe the optimization procedure; for the 3-node circuits (Fig. 4a) the exact maximum can be verified by grid search, and a similar verification on small N would establish the method's reliability.
  3. [Fig. 2c,d; Appendix E] The first-order nature of the transition and the absence of hysteresis are central to the contrast with conventional nonequilibrium circuits. The discontinuity itself is well supported for cycles by the static existence calculation in Appendix E and by Fig. 2c. However, the claim of convergence from 'different initial conditions' to the same NESS (Fig. 2d) is reported without specifying the number or distribution of initial conditions, and the stability of the collapsed state for μ > μ_crit is not analyzed: Appendix E is a static search over the conductance simplex for configurations in which every edge's per-capita dissipation exceeds σ_maint, not a dynamical stability analysis of the collapse boundary. Since a discontinuous jump in the order parameter without a bistable window is an unusual and interesting feature, please report the tested ensemble and, at least for the 3-cycle, confirm analytically or numerically that the collapsed state is unstable for all μ > μ_crit.
minor comments (9)
  1. [Eq. (1); Eq. (A2); Eq. (E1)] The displayed master equation dpi/dt = Σ_j W_ij p_j omits the loss term −Σ_j W_ji p_i; Appendix B writes the correct form, and the flux and dissipation formulas in Eqs. (3) and (A4) are consistent with the complete equation. Please correct the displayed equation in all three locations, since a reader reproducing the quasi-steady-state reduction in Appendix A would otherwise obtain the wrong starting point.
  2. [Appendix A (Methods)] The construction of the detailed-balance rate matrix is incompletely specified: sampling 'the upper triangular part and the lower penultimate diagonal' does not determine a unique algorithm, and the Gaussian (mean 1, std 0.3) can draw negative rates. Please provide an explicit algorithm (e.g., sample node potentials and symmetric amplitudes, then set k_ij/k_ji = exp(q_j − q_i)) and state how negative draws are handled.
  3. [Reproducibility] No code or data availability statement is given, and the numbers of realizations and initial conditions are not reported for Figs. 2c, 2d, 4b–d, and 5b. Please add these details, preferably with a code repository, so the simulation claims can be independently verified.
  4. [Eq. (4); Fig. 5b] Eq. (4) relates μ_crit to the free maintenance parameter σ_maint; it is a threshold consistency condition rather than an ab initio prediction. The caption of Fig. 5b describes a '1-parameter fit to theoretical predictions' — please state which parameter is fitted and list explicitly which quantities are predicted versus fitted.
  5. [Appendix H] The statement that the dynamics 'lack any obvious global optimization principles such as a Lyapunov function' should be reconciled with Eq. (6), which shows the ratio dynamics approximate gradient ascent on log σ_tot with a state-dependent metric; clarify whether the claim is that no exact Lyapunov function is known, since an approximate gradient structure is displayed.
  6. [Fig. 4b,d] Please define the 'typical dissipation' (green) sampling distribution — presumably uniformly random normalized conductances on the simplex — and state how many random configurations were drawn per realization.
  7. [Fig. 1 caption; Abstract] The phrases 'exact mapping' and 'rigorously map' are stronger than what Appendix B establishes, given the linear-regime (Michaelis–Menten, substrate ≪ K_M) assumption and the arbitrary choices in Appendix C; please qualify the claim accordingly.
  8. [Appendix A (Eq. A5)] The claim that the logistic cutoff α is inconsequential can be justified directly: the factor (1 − α Σ Λ) multiplies every dΛ_ij/dt by the same scalar, so on the simplex of conductance ratios it only reparameterizes time and does not shift the fixed points of the per-capita dissipation. Adding this one-line argument would remove the need for the reader to take the claim on faith.
  9. [Throughout] Please fix the duplicated 'use use' in the section on nonequilibrium steady states and the garbled author names 'Pavsko vZupanovi´c' in refs. [47], [62], and [64].

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's results are internal mathematical consequences of its explicitly postulated adaptive rule, not reimported assumptions.

full rationale

The paper defines a model in Eq. (2), dLambda_ij/dt = sigma_ij - sigma_maint*Lambda_ij, and then derives consequences: the critical drive in Eq. (4), the save-the-weakest nonlinearity, near-maximal dissipation, and the drive-complexity relation. Each of these is obtained from the model equations by analysis and simulation, not by fitting a parameter to the target result or by defining one quantity in terms of another. The critical-drive formula (Eq. 4; Appendix D) is a threshold consistency condition: it states the drive at which the weakest edge's per-capita dissipation equals the maintenance parameter sigma_maint. That is a derived consequence of the assumed growth law, not a circular redefinition of the transition. The self-citations to refs. [39] and [70] are used for motivation and analogy ('This is analogous to ecological species growth dynamics based on energy acquisition [39]') rather than as load-bearing proofs; the model explicitly states its growth rule as an assumption and even invites alternatives in the Discussion: 'in ecosystems where a substantial fraction of resource flux is diverted into biomass rather than energy dissipation, including flux-based local growth rules and more realistic redox chemistry, may be more appropriate.' Because the central claims are properties of the model's own dynamics against no external data, there is no fitted-input-called-prediction step and no self-definitional reduction. The main scientific risk is biological realism of Eq. (2), not circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claims rest on the adaptive rule in Eq. (2), a timescale separation, and a positive maintenance threshold. These are plausible modeling choices rather than derived ecological facts, and the phase transition property is sensitive to at least the maintenance parameter. No new physical entities are introduced; the 'batteries' are a mapping of autotroph edges to external drives, not a new mediator.

free parameters (4)
  • sigma_maint (maintenance dissipation) = 0.05
    Chosen by hand in Appendix A and Appendix C ('chosen arbitrarily'). Every central phenomenon, including the existence of a critical drive and the collapse transition, depends on this positive threshold.
  • random kinetic rate distribution = k_ij = 1 + 0.3 eta_ij, eta ~ N(0,1)
    Appendix A samples rates from this Gaussian; main-text phenomenology is demonstrated for this distribution and is not derived for general k.
  • logistic carrying capacity alpha = 1e-4 (total conductance cap 1e4)
    Introduced in Appendix A to stop unbounded growth of conductances. Authors assert the value does not affect normalized results, but no proof or sensitivity scan is shown.
  • redox parameterization constants = k0 = 1e10 s^-1, beta = 1/2.5 V^-1, sigma_maint = 5e-2
    Appendix C states 'These parameters are chosen arbitrarily' for the redox-tower example; the example is used only for qualitative comparison.
assumptions (5)
  • domain assumption Master-equation dynamics for electron densities with linear rates W_ij = Lambda_ij k_ij (linear kinetics)
    Assumed in the mapping (Appendix B, 'Ohm's law or linear approximation'); the authors note nonlinear departures would yield non-Ohmic behavior, so the central results are only for linear kinetics.
  • domain assumption Fast relaxation of electron densities relative to conductance dynamics (quasi-steady state)
    Appendix A step 3 sets dp/dt = 0 and solves p as a function of Lambda ratios; the phase transition analysis depends on this timescale separation.
  • domain assumption Species (edges) grow according to edge dissipation minus a constant per-capita maintenance cost
    Equation (2)-(3) postulates dLambda/dt = sigma_ij - sigma_maint Lambda. This is the adaptive rule from which all reported phenomena follow; it is motivated ecologically but not derived from a mechanistic population model.
  • domain assumption Each species catalyzes one net redox reaction; at most one edge per node pair
    Appendix B explicitly assumes a single net reaction per species and avoids multiple paths between nodes, simplifying the Markov chain; real ecosystems can have multi-reaction species.
  • ad hoc to paper Logistic saturation term regularizes otherwise scale-invariant dynamics
    Appendix A introduces the factor (1 - alpha sum Lambda) to stop runaway growth; the authors claim it does not affect normalized outcomes, but this is stated rather than demonstrated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ecosystems as adaptive living circuits." pith.science (2026). https://pith.science/paper/TFZ2LFOL

@misc{pith2026250622017,
  author       = {Pith},
  title        = {Pith review of: Ecosystems as adaptive living circuits},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TFZ2LFOL}},
  note         = {Machine review of arXiv:2506.22017}
}
read the original abstract

Unlike many physical nonequilibrium systems, in biological systems, the coupling to external energy sources is not a fixed parameter but adaptively controlled by the system itself. We do not have theoretical frameworks that allow for such adaptability. As a result, we cannot understand emergent behavior in living systems where structure formation and non-equilibrium drive coevolve. Here, using ecosystems as a model of adaptive systems, we develop a framework of living circuits whose architecture changes adaptively with the energy dissipated in each circuit edge. We find that unlike traditional nonequilibrium systems, living circuits exhibit a phase transition from equilibrium death to a nonequilibrium dissipative state beyond a critical driving potential. This transition emerges through a feedback mechanism that saves the weakest edges by routing dissipation through them, even though the adaptive rule locally rewards the strongest dissipating edges. Despite lacking any global optimization principle, living circuits achieve near-maximal dissipation, with higher drive promoting more complex circuits. Our work establishes ecosystems as paradigmatic examples of living circuits whose structure and dissipation are tuned through local adaptive rules.

Figures

Figures reproduced from arXiv: 2506.22017 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: a demonstrates this phenomenon for a specific redox tower with 3 nodes (redox states) and 3 edges (liv￾ing species), where all circuit configurations can be vi￾sualized on a simplex of normalized conductances. The heatmap overlay shows the landscape of global dissipa￾t…
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

73 extracted references · 69 canonical work pages

  1. [1]

    Nonequilibrium physics in biology

    Xiaona Fang, Karsten Kruse, Ting Lu, and Jin Wang. Nonequilibrium physics in biology. Reviews of Modern Physics, 91(4):045004, 2019

    2019

  2. [2]

    Efficiency of molecu- lar motors at maximum power

    Tim Schmiedl and Udo Seifert. Efficiency of molecu- lar motors at maximum power. Europhysics Letters, 83(3):30005, 2008

    work page 2008

  3. [3]

    Theoretical per- spectives on biological machines

    Mauro L Mugnai, Changbong Hyeon, Michael Hinczewski, and D Thirumalai. Theoretical per- spectives on biological machines. Reviews of Modern Physics, 92(2):025001, 2020

    work page 2020

  4. [4]

    Simulating a chemi- cally fueled molecular motor with nonequilibrium molec- ular dynamics

    Alex Albaugh and Todd R Gingrich. Simulating a chemi- cally fueled molecular motor with nonequilibrium molec- ular dynamics. Nature communications, 13(1):2204, 2022

    work page 2022

  5. [5]

    Nonequilib- rium generation of information in copolymerization pro- cesses

    David Andrieux and Pierre Gaspard. Nonequilib- rium generation of information in copolymerization pro- cesses. Proceedings of the National Academy of Sciences, 105(28):9516–9521, 2008

    work page 2008

  6. [6]

    Design principles for nonequilibrium self-assembly

    Michael Nguyen and Suriyanarayanan Vaikuntanathan. Design principles for nonequilibrium self-assembly. Proceedings of the National Academy of Sciences , 113(50):14231–14236, 2016

    work page 2016

  7. [7]

    Self- assembling outside equilibrium: emergence of structures mediated by dissipation

    A Arango-Restrepo, D Barrag´ an, and JM Rubi. Self- assembling outside equilibrium: emergence of structures mediated by dissipation. Physical Chemistry Chemical Physics, 21(32):17475–17493, 2019

    work page 2019

  8. [8]

    Free energy dissipation enhances spatial accu- racy and robustness of self-positioned turing pattern in small biochemical systems

    Dongliang Zhang, Chenghao Zhang, Qi Ouyang, and Yuhai Tu. Free energy dissipation enhances spatial accu- racy and robustness of self-positioned turing pattern in small biochemical systems. Journal of the Royal Society Interface, 20(204):20230276, 2023

    work page 2023

Show all 73 references
  1. [9]

    Minimum energetic cost to maintain a target nonequilib- rium state

    Jordan M Horowitz, Kevin Zhou, and Jeremy L England. Minimum energetic cost to maintain a target nonequilib- rium state. Physical Review E, 95(4):042102, 2017

    2017

  2. [10]

    Pattern forma- tion outside of equilibrium

    Mark C Cross and Pierre C Hohenberg. Pattern forma- tion outside of equilibrium. Reviews of modern physics, 65(3):851, 1993

    1993

  3. [11]

    Kinetic proofreading: a new mechanism for reducing errors in biosynthetic processes requiring high specificity

    John J Hopfield. Kinetic proofreading: a new mechanism for reducing errors in biosynthetic processes requiring high specificity. Proceedings of the National Academy of Sciences, 71(10):4135–4139, 1974

    1974

  4. [12]

    Kinetic amplification of enzyme discrim- ination

    Jacques Ninio. Kinetic amplification of enzyme discrim- ination. Biochimie, 57(5):587–595, 1975

    1975

  5. [13]

    Discriminatory proofreading regimes in nonequilibrium systems

    Arvind Murugan, David A Huse, and Stanislas Leibler. Discriminatory proofreading regimes in nonequilibrium systems. Physical Review X, 4(2):021016, 2014

    2014

  6. [14]

    Size limits the sensitivity of kinetic schemes

    Jeremy A Owen and Jordan M Horowitz. Size limits the sensitivity of kinetic schemes. Nature communications, 14(1):1280, 2023

    2023

  7. [15]

    Energetic costs of cellular computation

    Pankaj Mehta and David J Schwab. Energetic costs of cellular computation. Proceedings of the National Academy of Sciences, 109(44):17978–17982, 2012

    2012

  8. [16]

    Thermodynamics of computa- tional copying in biochemical systems

    Thomas E Ouldridge, Christopher C Govern, and Pieter Rein ten Wolde. Thermodynamics of computa- tional copying in biochemical systems. Physical Review X, 7(2):021004, 2017

    2017

  9. [17]

    Universal energy-accuracy tradeoffs in nonequilibrium cellular sensing

    Sarah E Harvey, Subhaneil Lahiri, and Surya Ganguli. Universal energy-accuracy tradeoffs in nonequilibrium cellular sensing. Physical Review E, 108(1):014403, 2023

    2023

  10. [18]

    Flexibil- ity and sensitivity in gene regulation out of equilib- rium

    Sara D Mahdavi, Gabriel L Salmon, Patill Dagh- lian, Hernan G Garcia, and Rob Phillips. Flexibil- ity and sensitivity in gene regulation out of equilib- rium. Proceedings of the National Academy of Sciences, 121(46):e2411395121, 2024

    2024

  11. [19]

    Limits on the compu- tational expressivity of non-equilibrium biophysical pro- cesses

    Carlos Floyd, Aaron R Dinner, Arvind Murugan, and Suriyanarayanan Vaikuntanathan. Limits on the compu- tational expressivity of non-equilibrium biophysical pro- cesses. arXiv preprint arXiv:2409.05827, 2024

    2024 arXiv

  12. [20]

    Phosphorylation energy hypothesis: open chemical systems and their biological functions

    Hong Qian. Phosphorylation energy hypothesis: open chemical systems and their biological functions. Annu. Rev. Phys. Chem., 58(1):113–142, 2007

    2007

  13. [21]

    Free energy transduction and biochemical cycle kinetics

    Terrell L Hill. Free energy transduction and biochemical cycle kinetics. Courier Corporation, 2005

    2005

  14. [22]

    Nonequilibrium mechanics of active cytoskeletal networks

    Daisuke Mizuno, Catherine Tardin, Christoph F Schmidt, and Frederik C MacKintosh. Nonequilibrium mechanics of active cytoskeletal networks. Science, 315(5810):370–373, 2007

    2007

  15. [23]

    Spontaneous motion in hierarchically assembled active matter

    Tim Sanchez, Daniel TN Chen, Stephen J DeCamp, Michael Heymann, and Zvonimir Dogic. Spontaneous motion in hierarchically assembled active matter. Na- ture, 491(7424):431–434, 2012

    2012

  16. [24]

    F-actin ar- chitecture determines the conversion of chemical en- ergy into mechanical work

    Ryota Sakamoto and Michael P Murrell. F-actin ar- chitecture determines the conversion of chemical en- ergy into mechanical work. Nature Communications, 15(1):3444, 2024

    2024

  17. [25]

    Brock biology of microorganisms, volume 11

    Michael T Madigan, John M Martinko, Jack Parker, et al. Brock biology of microorganisms, volume 11. Pren- tice hall Upper Saddle River, NJ, 1997

    1997

  18. [26]

    Temporal and spatial distribution of the mi- crobial community of winogradsky columns

    David J Esteban, Bledi Hysa, and Casey Bartow- McKenney. Temporal and spatial distribution of the mi- crobial community of winogradsky columns. PLoS One, 10(8):e0134588, 2015

    2015

  19. [27]

    Introduction to a submolecular biology

    Albert Szent-Gyorgyi. Introduction to a submolecular biology. Elsevier, 2012

    2012

  20. [28]

    Bacterial competition: surviving and thriving in the microbial jungle

    Michael E Hibbing, Clay Fuqua, Matthew R Parsek, and S Brook Peterson. Bacterial competition: surviving and thriving in the microbial jungle. Nature reviews micro- biology, 8(1):15–25, 2010

    2010

  21. [29]

    The genotypic view of social interactions in microbial communities

    Sara Mitri and Kevin Richard Foster. The genotypic view of social interactions in microbial communities. An- nual review of genetics, 47(1):247–273, 2013

    2013

  22. [30]

    Horizontal gene transfer: building the web of life

    Shannon M Soucy, Jinling Huang, and Johann Peter 10 Gogarten. Horizontal gene transfer: building the web of life. Nature Reviews Genetics, 16(8):472–482, 2015

    2015

  23. [31]

    Antibiotics and the art of bacterial war

    Daniel M Cornforth and Kevin R Foster. Antibiotics and the art of bacterial war. Proceedings of the national academy of sciences, 112(35):10827–10828, 2015

    2015

  24. [32]

    Communication in bacteria: an ecological and evolutionary perspective

    Laurent Keller and Michael G Surette. Communication in bacteria: an ecological and evolutionary perspective. Nature Reviews Microbiology, 4(4):249–258, 2006

    2006

  25. [33]

    The microbial engines that drive earth’s biogeochemical cycles

    Paul G Falkowski, Tom Fenchel, and Edward F Delong. The microbial engines that drive earth’s biogeochemical cycles. science, 320(5879):1034–1039, 2008

    2008

  26. [34]

    Consistent microbial dynamics and functional commu- nity patterns derived from first principles

    Hadrien Delattre, Elie Desmond-Le Qu´ em´ ener, Chris- tian Duquennoi, Ahlem Filali, and Th´ eodore Bouchez. Consistent microbial dynamics and functional commu- nity patterns derived from first principles. The ISME Journal, 13(2):263–276, 2019

    2019

  27. [35]

    Does microbial life al- ways feed on negative entropy? thermodynamic anal- ysis of microbial growth

    U Von Stockar and J-S Liu. Does microbial life al- ways feed on negative entropy? thermodynamic anal- ysis of microbial growth. Biochimica et Biophysica Acta (BBA)-Bioenergetics, 1412(3):191–211, 1999

    1999

  28. [36]

    Thermodynamics of mi- crobial growth and metabolism: an analysis of the cur- rent situation

    Urs Von Stockar, Thomas Maskow, Jingsong Liu, Ian W Marison, and Rodrigo Patino. Thermodynamics of mi- crobial growth and metabolism: an analysis of the cur- rent situation. Journal of Biotechnology, 121(4):517–533, 2006

    2006

  29. [37]

    Lehninger principles of biochemistry

    Albert L Lehninger, David L Nelson, and Michael M Cox. Lehninger principles of biochemistry. Macmillan, 2005

    2005

  30. [38]

    Bioener- getics of microbial growth

    Joseph J Heijnen and Robbert Kleerebezem. Bioener- getics of microbial growth. Encyclopedia of bioprocess technology: Fermentation, biocatalysis, and biosepara- tion, 1:267–291, 1999

    1999

  31. [39]

    Closed ecosystems extract energy through self-organized nutrient cycles

    Akshit Goyal, Avi I Flamholz, Alexander P Petroff, and Arvind Murugan. Closed ecosystems extract energy through self-organized nutrient cycles. arXiv preprint arXiv:2305.19102, 2023

    2023 arXiv

  32. [40]

    Network theory of microscopic and macroscopic behavior of master equation systems

    J¨ urgen Schnakenberg. Network theory of microscopic and macroscopic behavior of master equation systems. Reviews of Modern physics, 48(4):571, 1976

    1976

  33. [41]

    Markov chains

    James R Norris. Markov chains. Number 2. Cambridge university press, 1998

    1998

  34. [42]

    Global op- timization, local adaptation, and the role of growth in distribution networks

    Henrik Ronellenfitsch and Eleni Katifori. Global op- timization, local adaptation, and the role of growth in distribution networks. Physical review letters , 117(13):138301, 2016

    2016

  35. [43]

    Random net- work peristalsis in physarum polycephalum organizes fluid flows across an individual

    Karen Alim, Gabriel Amselem, Fran¸ cois Peaudecerf, Michael P Brenner, and Anne Pringle. Random net- work peristalsis in physarum polycephalum organizes fluid flows across an individual. Proceedings of the Na- tional Academy of Sciences, 110(33):13306–13311, 2013

    2013

  36. [44]

    Adaptation and optimization of biological transport networks

    Dan Hu and David Cai. Adaptation and optimization of biological transport networks. Physical review letters, 111(13):138701, 2013

    2013

  37. [45]

    Fluctuations and redundancy in op- timal transport networks

    Francis Corson. Fluctuations and redundancy in op- timal transport networks. Physical Review Letters, 104(4):048703, 2010

    2010

  38. [46]

    Martyushev and Vladimir D

    Leonid M. Martyushev and Vladimir D. Seleznev. Maxi- mum entropy production principle in physics, chemistry and biology. Physics Reports, 426(1):1–45, 2006

    2006

  39. [48]

    Paltridge

    Garth W. Paltridge. Climate and thermodynamic sys- tems of maximum dissipation. Nature, 279(5714):630– 631, 1979

    1979

  40. [49]

    Wiley-Interscience, 1977

    Gr´ egoire Nicolis and Ilya Prigogine.Self-organisation in nonequilibrium systems. Wiley-Interscience, 1977

    1977

  41. [50]

    Statistical physics of self-replication

    Jeremy L England. Statistical physics of self-replication. The Journal of chemical physics, 139(12), 2013

    2013

  42. [51]

    Dissipative adaptation in driven self- assembly

    Jeremy L England. Dissipative adaptation in driven self- assembly. Nature nanotechnology, 10(11):919–923, 2015

    2015

  43. [52]

    Biological order, structure and instabilities1

    Ilya Prigogine and Gr´ egoire Nicolis. Biological order, structure and instabilities1. Quarterly reviews of bio- physics, 4(2-3):107–148, 1971

    1971

  44. [53]

    Statistical physics of adaptation

    Nikolay Perunov, Robert A Marsland, and Jeremy L England. Statistical physics of adaptation. Physical Re- view X, 6(2):021036, 2016

    2016

  45. [54]

    Rules for bio- logically inspired adaptive network design

    Atsushi Tero, Seiji Takagi, Tetsu Saigusa, Kentaro Ito, Dan P Bebber, Mark D Fricker, Kenji Yumiki, Ryo Kobayashi, and Toshiyuki Nakagaki. Rules for bio- logically inspired adaptive network design. Science, 327(5964):439–442, 2010

    2010

  46. [55]

    Demonstration of decentralized physics- driven learning

    Sam Dillavou, Menachem Stern, Andrea J Liu, and Dou- glas J Durian. Demonstration of decentralized physics- driven learning. Physical Review Applied, 18(1):014040, 2022

    2022

  47. [56]

    Physical learning of power-efficient solutions

    Menachem Stern, Sam Dillavou, Dinesh Jayaraman, Douglas J Durian, and Andrea J Liu. Physical learning of power-efficient solutions. arXiv preprint arXiv:2310.10437, 2023

    2023 arXiv

  48. [57]

    Damage and fluctuations induce loops in optimal transport networks

    Eleni Katifori, Gergely J Sz¨ oll˝ osi, and Marcelo O Magnasco. Damage and fluctuations induce loops in optimal transport networks. Physical review letters, 104(4):048704, 2010

    2010

  49. [58]

    The units of selection

    Richard C Lewontin. The units of selection. Annual review of ecology and systematics, pages 1–18, 1970

    1970

  50. [59]

    Contribution to the energetics of evolu- tion

    Alfred J Lotka. Contribution to the energetics of evolu- tion. Proceedings of the National Academy of Sciences, 8(6):147–151, 1922

    1922

  51. [60]

    Non-equilibrium thermodynamics, max- imum entropy production and earth-system evolu- tion

    Axel Kleidon. Non-equilibrium thermodynamics, max- imum entropy production and earth-system evolu- tion. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 368(1910):181–196, 2010

    1910

  52. [61]

    Maximum entropy production and the fluctuation theorem

    Roderick C Dewar. Maximum entropy production and the fluctuation theorem. Journal of Physics A: Mathe- matical and General, 38(21):L371, 2005

    2005

  53. [62]

    Photosynthetic models with maximum entropy production in irreversible charge transfer steps

    Davor Jureti´ c and Pavsko vZupanovi´ c. Photosynthetic models with maximum entropy production in irreversible charge transfer steps. Computational Biology and Chem- istry, 27(6):541–553, 2003

    2003

  54. [63]

    Energy flow in ecosystems: a historical review

    Eugene P Odum. Energy flow in ecosystems: a historical review. American Zoologist, 8(1):11–18, 1968

    1968

  55. [64]

    Energy, ecology, and economics

    Howard T Odum. Energy, ecology, and economics. Am- bio, pages 220–227, 1973

    1973

  56. [65]

    Kirchhoff’s loop law and the maximum entropy produc- tion principle

    Pa vsko vZupanovi´ c, Davor Jureti´ c, and Sre´ cko Botri´ c. Kirchhoff’s loop law and the maximum entropy produc- tion principle. Physical Review E—Statistical, Nonlin- ear, and Soft Matter Physics, 70(5):056108, 2004

    2004

  57. [66]

    Biogeochemical gradients and microbial communities in winogradsky columns established with polluted wetland sediments

    Izabella Babcs´ anyi, Fatima Meite, and Gwena¨ el Imfeld. Biogeochemical gradients and microbial communities in winogradsky columns established with polluted wetland sediments. FEMS microbiology ecology, 93(8):fix089, 2017

    2017

  58. [67]

    Oxic-anoxic regime shifts mediated by feedbacks between biogeo- chemical processes and microbial community dynamics

    Timothy Bush, Muhe Diao, Rosalind J Allen, Ruben Sinnige, Gerard Muyzer, and Jef Huisman. Oxic-anoxic regime shifts mediated by feedbacks between biogeo- chemical processes and microbial community dynamics. Nature Communications, 8(1):789, 2017. 11

    2017

  59. [68]

    Microbial mats: a joint venture

    Hans Van Gemerden. Microbial mats: a joint venture. Marine Geology, 113(1-2):3–25, 1993

    1993

  60. [69]

    Bacterial proliferation pattern formation

    John S Chuang, Riccardo Rao, and Stanislas Leibler. Bacterial proliferation pattern formation. PRX Life, 3(1):013007, 2025

    2025

  61. [70]

    Energy-ordered resource stratification as an agnostic signature of life

    Akshit Goyal and Mikhail Tikhonov. Energy-ordered resource stratification as an agnostic signature of life. Nature communications, 16(1):2867, 2025

    2025

  62. [71]

    Synaptic energy use and supply

    Julia J Harris, Renaud Jolivet, and David Attwell. Synaptic energy use and supply. Neuron, 75(5):762–777, 2012

    2012

  63. [72]

    Energy efficient synaptic plasticity

    Ho Ling Li and Mark Cw Van Rossum. Energy efficient synaptic plasticity. Elife, 9:e50804, 2020

    2020

  64. [73]

    How a well-adapted im- mune system is organized

    Andreas Mayer, Vijay Balasubramanian, Thierry Mora, and Aleksandra M Walczak. How a well-adapted im- mune system is organized. Proceedings of the National Academy of Sciences, 112(19):5950–5955, 2015

    2015

  65. [74]

    save the weakest

    Avi I Flamholz, Akshit Goyal, Woodward W Fischer, Dianne K Newman, and Rob Phillips. The proteome is a terminal electron acceptor. Proceedings of the National Academy of Sciences, 122(1):e2404048121, 2025. 12 Appendices and Supplementary Material Ecosystems as adaptive living ...

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.