Nonequilibrium Theory for Molecular Machine Design

arXiv:2605.09105 · cond-mat.stat-mech, physics.bio-ph · Submitted 2026-05-09 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Nonequilibrium Theory for Molecular Machine Design".

Mira: Here's the gist of the paper. Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production because this alone is not sufficient for design, optimization and evolution because it doesn’t treat cost-benefit tradeoffs, or small-system misflows (backsteps, futile cycles, ineffective actions), or differential properties for flow design.

Kai: Network Flow Design Principles The paper develops CFT Design based on Caliber Force Theory to provide a general framework for optimizing nonequilibrium flow networks.

Mira: This framework addresses the need to treat costs and benefits of flows, time costs due to misflows, and differential properties for flow design.

Kai: It establishes observable-force conjugacy for NEQ dynamics to formalize a framework of flow design based on Caliber Force Theory

13, 14: .

Title and authors: Kai: So, we're looking at this paper, "Nonequilibrium Theory for Molecular Machine Design," and it’s about moving beyond just counting node populations or edge fluxes with Master Equations.

Mira: Right. It says that just looking at those things isn't enough for designing or optimizing these machines because you miss the cost-benefit tradeoffs, like when you have small misflows—things like backsteps or futile cycles—or how different parts of the network behave differently.

Lev: From an engineering standpoint, that means standard Master Equation simulations might give you a picture of the dynamics but not a clear path to actually build something that performs well under real constraints.

Kai: What this paper proposes is using something called CFT Design, which uses Caliber Force Theory to create a general framework for optimizing these nonequilibrium flow networks.

Mira: It establishes this framework by formalizing an observable-force conjugacy for nonequilibrium dynamics based on Caliber Force Theory, which helps you treat those costs and benefits and the time costs from misflows.

Lev: So it’s trying to give us a way to define what a good design looks like by treating the forces in terms of affinities for node dwelling, edge exchange, and cycle completion.

Kai: Exactly. And it uses these forces to solve constrained optimization problems using conjugate coordinates, which is how you actually get the design decisions made.

Mira: They derive these conjugate forces by maximizing path entropy, what they call Maximum Caliber of the Markov dynamics thirty-three <ref:2605.09105#pg3>. This gives us a way to define what’s desirable in terms of flow organization.

Lev: And then they show how node energy acts as an isotropic scaler—perturbing it scales all outgoing transition rates proportionally, but it doesn't change the ratios between different flows.

Kai: That's important because if you can scale things proportionally, you know that changing the energy doesn't fundamentally alter the structure of what’s happening in terms of flow distribution.

Mira: They also look at kinetic barriers as topological routers; tuning a barrier only affects traffic on that specific edge if its net flux is zero, which lets them establish design limits for things like enzyme inhibitors.

Title and authors: Lev: That sounds useful for hardware because it suggests we can route errors or unwanted flows by tuning the barriers, and if an edge has no net flow, you just tune its own traffic without messing up the rest of the system’s observables.

Kai: The paper shows that speed, accuracy, and cost trade-offs aren't absolute limits but boundary constraints within kinetic proofreading networks when you use CFT’s coordinate systems.

Mira: So instead of seeing those metrics as fixed rules, they suggest you can decouple them when you use the coordinates like pi tau rem S epsilon C r prime Fcycle.

Lev: That decoupling is key for real hardware because it means we can potentially optimize for speed without necessarily sacrificing accuracy or incurring prohibitive energy costs.

Kai: They also look at enzyme inhibition specifically, and they distinguish between competitive and uncompetitive inhibitors using this CFT framework.

Mira: Competitive inhibitors act like topological dead-ends optimized just by node energies, while noncompetitive ones form leaky loops that need kinetic barrier routing for optimization.

Lev: That distinction is concrete; it tells us whether we need to focus on energy landscape tuning or barrier routing when designing an inhibitor.

Kai: Then they introduce the Inhibitor Equal Traffic Principle, which states that for noncompetitive inhibition, you need a balanced traffic distribution across the leak loop to maximize production.

Mira: Maximizing that production happens at the traffic upper bound where a bottleneck severs that compensatory leak, allowing you to reach the maximum capacity set by the primary binding traffic.

Lev: That gives us a clear target for optimization in inhibitor design—find that bottleneck point on the flow network.

Kai: Moving into their algorithmic details, they use an inverse design protocol driven along a linear sequence parameterized by lambda.

Mira: At each step of this process, if any edge traffic exceeds one times its transition rate, the local kinetic bandwidth is expanded by multiplying that edge's traffic by ten.

Lev: That sounds like a heuristic for exploration; it’s an iterative way to push the system toward a better flow configuration when you don't have an exact analytical solution.

Kai: And they reconstruct the complete set of microscopic transition rates using the fundamental CFT mapping, which gives us kij equal to tau ij plus Jij divided by two pi i.

Title and authors: Mira: That explicit reconstruction helps illuminate what they call the pi-degeneracy in inverse design, showing how multiple rate configurations can lead to the same dynamics.

Lev: So you can see exactly how different microscopic rates map onto the same macroscopic flow behavior, which is helpful for validation.

Kai: To translate an operational target point back into those microscopic transition rates, they evaluate four fundamental cycle fluxes and compute local net fluxes Jij on every edge.

Mira: Then they substitute fixed cycle forces and the remaining tree traffics into equation S59 to get the final rate configuration.

Lev: That systematic sequence is what allows you to go from a desired performance goal back down to the actual kinetic rates of your machine.

Kai: They also establish traffic bounds, saying that Jij less than or equal to tau ij imposes a strict constraint on the product of one-way probability fluxes.

Mira: And they mention that when you drive the system into regimes of negative proofreading dissipation, a minimum achievable error floor is established by this constraint.

Lev: That gives us a physical limit on how accurate we can get before we hit these thermodynamic bounds in certain operating conditions.

Kai: To wrap up the implications of this paper, the CFT Design framework provides a comprehensive system for molecular-machine inverse design, solving constrained optimal rates via conjugate coordinates.

Mira: It moves beyond just calculating fluxes to providing actual design principles for optimizing nonequilibrium flow networks based on Caliber Force Theory.

Lev: For us working on real hardware, this means we have a mathematical tool to predict the physical performance bounds of proofreading networks before we even start building them.

Kai: So, that’s what they are doing with "Nonequilibrium Theory for Molecular Machine Design," giving us tools to design faster motors and better inhibitors.

Mira: It’s a solid framework for treating the trade-offs between speed, accuracy, and cost in these complex systems.

Lev: We'll keep an eye on how these kinetic barriers act as routers when we try to implement them in actual devices.

Kai: That’s the gist of it—it gives us a language to talk about flow design that accounts for all those subtleties.

The paper's summary: Kai: So, we’re looking at how this paper summarizes its whole approach to designing molecular machines using what they call CFT Design based on Caliber Force Theory, right?

Mira: Yeah, it boils down to treating these machines not just as simple flow networks with Master Equations, but as systems where you have to account for the actual costs and benefits of every single flow.

Lev: So they’re moving past just counting node populations and edge fluxes to actually designing something that works under real constraints.

Kai: Exactly. It says they established a framework by formalizing an observable-force conjugacy for nonequilibrium dynamics, which is basically a mathematical way to treat the machine's performance goals as forces you can optimize against.

Mira: They use these forces to define things like affinities for node dwelling and edge exchange, which lets them solve constrained optimization problems using conjugate coordinates, giving us concrete design rules.

Lev: It’s interesting how they handle things like kinetic barriers; tuning a barrier only affects its own traffic if its net flux is zero, which gives them a way to establish routing constraints.

Kai: And then they tackle the trade-offs between speed and accuracy by showing those metrics aren't absolute limits, but rather boundary constraints when you use their specific coordinate system.

Mira: They map out these trade-offs using coordinates like pi tau rem S epsilon C r prime Fcycle, which lets you decouple them for optimization purposes.

Lev: That decoupling is important because it shows us where the real overhead lands on actual hardware when we try to implement these proofreading cycles in a physical system.

Kai: The paper then details how they use an inverse design protocol that drives target functional variables along a linear sequence parameterized by lambda, and if traffic gets too high, they expand the kinetic bandwidth.

Mira: They also show how you can reconstruct the complete set of microscopic transition rates from the final performance point using a mapping called kij equal to tau ij plus Jij divided by two pi i.

Lev: That reconstruction is powerful because it shows you exactly what those microscopic rates need to be, which helps validate if the design is actually physically realizable.

Kai: They also provide a systematic sequence for translating an operational target point back into those specific microscopic transition rates by evaluating cycle fluxes and local net fluxes on every edge.

Mira: And they set physical limits, saying that traffic less than or equal to tau ij imposes a strict bound on the product of one-way probability fluxes.

Lev: That constraint means when you run into negative proofreading dissipation regimes, there's a minimum error floor you have to deal with in the system.

Kai: So what this means for us is they’ve given us a mathematical language to design faster motors and better inhibitors by accounting for those flow subtleties.

Mira: It moves the discussion from just simulation results to actual design principles based on Caliber Force Theory, which is a big step forward in treating these complex nonequilibrium systems.

Lev: We need to keep an eye on how they handle the kinetic capacity bound; figuring out that bottleneck point is where we’ll focus our efforts when building proofreading hardware.

The paper's improvements: Kai: So, we just talked about how the paper sets up this whole flow network optimization thing using Caliber Force Theory, and now they’re talking about what they actually suggest improving in that framework itself.

Mira: Right. They’re not just giving us a static design tool; they are suggesting ways to make the entire system more robust for real-world application, like molecular motors or proofreading systems.

Lev: What do you mean by improvements? Are we talking about adding new terms to the coordinate set, or something deeper about how they handle those kinetic bottlenecks?

Kai: It’s more than just adding terms. They propose using "traffic control" for molecular motors, essentially routing flows through kinetic barriers to actively correct futile backstepping errors when things go wrong.

Mira: That means instead of just letting the system run, you're actively steering the traffic to avoid dead ends caused by those small misflows we talked about earlier.

Lev: From an engineering standpoint, that sounds like dynamic control, which is what we need for real hardware overhead management. Does this mean lower energy dissipation overall?

Kai: It’s aimed at making the design more efficient by minimizing those time costs associated with misflows, which is a major hurdle in kinetic proofreading.

Mira: And they suggest decoupling speed, accuracy, and cost even further when using their specific coordinate system. That means you can optimize for one metric without having to sacrifice the others in the interior of the network's operation.

Lev: If we can truly decouple those things, it changes how we budget resources for proofreading cycles on a chip or in a lab setup.

Kai: They also refine how they analyze enzyme inhibition by distinguishing between competitive and noncompetitive inhibitors using this CFT structure, which lets you choose the right optimization strategy based on the inhibitor type.

Mira: So for competitive inhibitors, you focus purely on node energy tuning to create those dead-ends, whereas for noncompetitive ones, you need that kinetic barrier routing we discussed earlier to get a good result.

Lev: That distinction is practical; it tells us exactly which tool—the energy scaler or the barrier router—to pull out of the toolbox depending on what we’re trying to build.

Kai: They also provide an algorithmic way, a linear exploration protocol, where they expand the kinetic bandwidth if any edge traffic exceeds one times its transition rate.

Mira: That’s a practical heuristic for inverse design; it’s an iterative way to push the system toward a better flow configuration when you aren't sure where the optimal point is located analytically.

Lev: That expansion factor of ten is quite aggressive, so we need to check how that scales when we move from theoretical models to actual simulation runs.

Kai: The paper also clarifies that the reconstruction of microscopic transition rates isn't just a mathematical exercise; it’s a systematic sequence involving four fundamental cycle fluxes and local net fluxes on every edge.

Mira: So, they give us a concrete recipe to go from a desired performance point back down to the actual numbers needed for the machine's operation.

Lev: That systematic translation is what I need when I’m trying to verify if a design we propose actually works on real hardware parameters. So these improvements give us more specific design rules and operational recipes, but they also flag where the model stops working—specifically, they only provide insights for networks that fit their current flow topology assumptions.

Conclusion: Kai: So we’ve covered how this paper sets up CFT Design to optimize nonequilibrium flow networks for machines, and now we’re wrapping up with what this all means in practice and what’s next.

Mira: Basically, the "Nonequilibrium Theory for Molecular Machine Design" gives us a mathematical language to move beyond just simulation results toward actual design principles for molecular machines.

Lev: It really does provide that systematic sequence to go from a target performance point back down to the microscopic transition rates, which is crucial because it’s what we need for hardware verification.

Kai: Right. The implications are that we can predict physical performance bounds before we even start building proofreading networks, which helps us manage expectations early on.

Mira: It’s a big shift from just looking at what happens to understanding how to actively steer those systems using forces like Caliber Force Theory.

Lev: I think the most immediate change is that we can finally quantify the overhead of those kinetic barriers and see exactly where the computational cost lands on a real system.

Kai: We’re moving toward a design process where you can explicitly trade speed, accuracy, and energy costs using those decoupling coordinates they introduced.

Mira: That suggests a more nuanced approach to proofreading design, depending on whether you are dealing with competitive or noncompetitive inhibitors in your system.

Lev: I think the explicit mapping of the rate reconstruction is what makes this framework useful for our error-correction work because it’s a direct path to parameter space exploration.

Kai: It’s a solid tool for moving from abstract dynamics to concrete engineering constraints on molecular machines, which is what we’re aiming for in our own quantum hardware experiments.

Mira: So, the "Nonequilibrium Theory for Molecular Machine Design" gives us a comprehensive system for optimizing nonequilibrium flow networks based on Caliber Force Theory.

Lev: It’s a framework that really helps us see the physical limits of proofreading cycles and where we need to focus our error-correction efforts next.

Ying-Jen Yang, Ken A. Dill

Laufer Center of Physical and Quantitative Biology at Stony Brook University · Department of Physics and Astronomy at Stony Brook University · Department of Chemistry at Stony Brook University

cond-mat.stat-mech, physics.bio-ph

Submitted: 2026-05-09

Updated: 2026-05-09

DOI: 10.1103/c8kc-rklj

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 89/100

The gist: Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy…

Key concepts

Caliber Forces
These are forces derived by maximizing the path entropy of the machine's dynamics. They act as affinities that guide the design process: one force relates to how often a node dwells, another to edge exchange, and a third to cycle completion. These forces enable constrained optimization for machine design.
Kinetic Barrier as a Topological Router
Kinetic barriers on edges act like routers in the network. Tuning these barriers allows designers to control traffic flow; if an edge has no net flux, changing its barrier only affects its own traffic, making it a key element for routing flows and optimizing network behavior.
Decoupling Speed, Accuracy, and Cost
The theory shows that speed-accuracy-cost trade-offs are not absolute limits but rather boundary constraints within the machine's operating region. By using specific coordinate systems derived from Caliber Force Theory, these three metrics can be treated independently during optimization.
Inhibitor Classification (Dead-Ends vs. Leaky Loops)
The theory classifies enzyme inhibitors based on their topology. Competitive and uncompetitive inhibitors act as 'topological dead-ends' optimized by node energies, while noncompetitive inhibitors form 'leaky loops' that require kinetic barrier routing for effective optimization.

Terminology

Summary

Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production because this alone is not sufficient for design, optimization and evolution because it doesn’t treat cost-benefit tradeoffs, or small-system misflows (backsteps, futile cycles, ineffective actions), or differential properties for flow design. p. 0

Where the paper says it

Dill Laufer Center of Physical and Quantitative Biology, Stony Brook University Department of Physics and Astronomy, Stony Brook University and Department of Chemistry, Stony Brook University (Dated: May 12, 2026) Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production. But this alone is not sufficient for design, optimization and evolution because it doesn’t treat cost-benefit tradeoffs, or small-system misflows (backsteps, futile cycles, ineffective actions), or differential properties for flow design.

Network Flow Design Principles

The paper develops CFT Design based on Caliber Force Theory to provide a general framework for optimizing nonequilibrium flow networks. This framework addresses the need to treat costs and benefits of flows, time costs due to misflows, and differential properties for flow design. p. 0

Where the paper says it

First, we need a way to treat the costs and benefits of flows. Whereas equilibrium (EQ) thermodynamics often balances the tradeoffs of heat and work, for example in Carnot cycles, network flow streams can have a much broader range of desirability properties.

It establishes observable-force conjugacy for NEQ dynamics to formalize a framework of flow design based on Caliber Force Theory [13, 14]. p. 0
Where the paper says it

Here, we develop principles for the design, optimization, and evolution of flows under constraints. Based on the recent Caliber Force Theory (CFT) [13, 14], we establish the observable-force conjugacy for NEQ dynamics to formalize a framework of flow design.

Caliber Forces and Response Rules

CFT derives conjugate forces in terms of the maximization of the path entropy (Maximum Caliber) of the Markov dynamics. The forces conjugate to node distribution, edge traffic, and cycle fluxes represent affinities for node dwelling, edge exchange, and cycle completion respectively. p. 1

Where the paper says it

CFT derives the conjugate forces in terms of the maximization of the path entropy (Maximum Caliber) of the Markov dynamics [13, 32, 33]. The forces conjugate to the node distribution πn, the edge traffic τij , and the cycle fluxes Jc represent the affinities for node dwelling, edge exchange, and cycle completion, respectively (Fig.

These forces enable machine design by solving constrained optimization using conjugate coordinates. p. 1
Where the paper says it

(3) CFT handles the LDB constraint naturally as Fcycle,c is half of the cycle affinity, and these forces enable machine design. Conjugate coordinates solve constrained optimization.

Node Energy as an Isotropic Scaler

Perturbing the energy En of state n modifies all outgoing transition rates from that state proportionally, leading to the relation ∂x¯α/∂En = πn x¯α. This shows that node energies act purely as isotropic scalers, scaling the global magnitude of all network flows proportionally but cannot alter flux ratios. p. 1

Where the paper says it

Node Energies affect the scaling of fluxes: CFT shows that node energy perturbations isotropically scale the average value of any flux variable φ¯ ∈ (τij , Jc) (or their linear combinations): ∂φ¯ ∂En = πn φ. ¯ (4) The state (node) with highest πn is the most effective scaler, but pure En perturbations cannot alter flux ratios.

Kinetic Barrier as a Topological Router

Perturbing the symmetric kinetic barrier Bij on an edge modifies both forward and backward rates equally and oppositely in logarithmic space. For all other observables, the response is given by ∂x¯α/∂Bij = −Jij [A−1]α,(ij). p. 2

Where the paper says it

3 governing how node energy modulates the entropy production rate (EPR) of the network: ∂En EPR ≡ ∂En [ P c Jc (2Fcycle,c)] = πn EPR where we have used 2Fcycle,c = P (ij)∈c ln(kij/kji) = P (ij)∈c arctanh(Jij/τij ), confirming that its derivative with respect to En vanishes. Kinetic Barriers affect the routing of fluxes: Conversely, barrier perturbations Bij route fluxes, mediated by the edge’s net flux Jij .

If an edge has zero net flux (Jij = 0), tuning its barrier alters only its own traffic while leaving all other network observables invariant, establishing kinetic barriers as the exclusive routers of the network. p. 2
Where the paper says it

(3) [14, 31]. If an edge has zero net flux (Jij = 0)—such as those leading to deadend pathways—tuning its barrier alters only its own traffic τij while leaving all other network observables invariant.

Decoupling Speed, Accuracy, and Cost in Kinetic Proofreading

CFT reveals that speed-accuracy-cost trade-offs are boundary constraints rather than absolute limits within the operational interior of kinetic proofreading networks. The framework allows for the decoupling of these metrics when using CFT’s coordinate systems, such as (π, τ rem, S, ε, C, r′, Fcycle). p. 1

Where the paper says it

CFT solves this using its conjugate coordinate systems and Legendre transformations [13]. By shifting from the observable coordinate (π, τ , J) to the observable-force coordinate (π, τ , Fcycle) (see SI Sec.

Stronger Enzyme Inhibition: Dead-Ends Versus Leaky Loops

In enzyme inhibition analysis using CFT for non-invertible networks, competitive and uncompetitive inhibitors act as topological dead-ends optimized solely by node energies. Conversely, noncompetitive inhibitors form leaky loops that require kinetic barrier routing for optimization. p. 6

Where the paper says it

optimize dead-end inhibitors, designers cannot rely on barrier routing; they must instead deepen the node energy (En) of the bound state, maximizing the binding affinity purely by suppressing koff. In contrast, noncompetitive inhibitors form leaky loops that require kinetic barrier routing for optimization.

Inhibitor Equal Traffic Principle

For noncompetitive inhibition, the Inhibitor Equal Traffic Principle (iETP) dictates that maximizing inhibition requires a balanced traffic distribution across the leak loop. The maximum catalytic production occurs at the traffic upper bound where a bottleneck severs the compensatory leak, allowing production to reach its maximum capacity dictated by the primary binding traffic. p. 6

Where the paper says it

Therefore, maximizing inhibition requires a balanced traffic distribution across the leak loop. Conversely, the maximum catalytic production occurs at the traffic upper bound (J max cat = Jleak + τE,ES), where the reverse one-way flux pES,E vanishes.

Algorithmic Details of Linear Exploration

An inverse design optimization protocol is employed that drives target functional variables along a linear sequence parameterized by λ. At each step, if any edge collides with the physical boundary (Jij /τij ≥ 1. p. 4

Where the paper says it

This parameterization yields physical performance bounds dictated by the non-negativity of one-way transition fluxes (pij = [τij + Jij ]/2 ≥ 0). For example, the kinetic capacity of the initial binding step of the Right substrate imposes the constraint JR(S, ε) + J ′ R (S, C, r′ ) ≤ τE,ER (i.

0), the local kinetic bandwidth is expanded by multiplying the traffic of that specific edge by 10. p. 2
Where the paper says it

Kinetic Barriers affect the routing of fluxes: Conversely, barrier perturbations Bij route fluxes, mediated by the edge’s net flux Jij . For any distribution or flux observable x¯α ∈ (πn, τab, Jc) (except the local traffic τij , which has an additional term in the equation [31]): ∂x¯α ∂Bij = −Jij A −1 α,(ij) (5) where A−1 is the inverse of the force-rate Jacobian matrix A(ij),α ≡ kij∂kijFα, which can be easily evaluated based on Eqs.

Rate Reconstruction and π-Degeneracy

The complete set of microscopic transition rates can be unambiguously reconstructed via the fundamental CFT mapping, where kij = τij + Jij / 2πi. This explicit reconstruction illuminates the π-degeneracy in inverse design. p. 6

Where the paper says it

Traditional thermodynamic models focus on energetic drives, net fluxes, and correlating machine performance with entropy production, but they lack explicit handles for the topological misflows that degrade machine efficiency—such as futile backstepping in the F1 motor, branched misrouting in kinetic proofreading, or dead-ends or leaks in enzyme inhibition. By pairing dynamic observables with conjugate path-entropic forces, Caliber Force Theory (CFT) establishes a dual design framework: inverse design to map performance targets to optimal transition rate configurations, and forward design to scale or route flows with node energies or kinetic barriers.

Systematic Rate Reconstruction

The sequence to translate an operational target point back to microscopic transition rates involves evaluating the four fundamental cycle fluxes, computing local net fluxes Jij on every edge, and substituting fixed cycle forces and remaining tree traffics into Eq. (S59).

Traffic Bounds and Anti-Proofreading Regimes

The physical constraint Jij ≤ τij imposes a strict boundary on the product of one-way probability fluxes pij = (τij + Jij)/2 ≥ 0. When the system is driven into regimes of negative proofreading dissipation, a minimum achievable error floor is established. p. 8

Where the paper says it

[41] We define “anti-proofreading” by the net reversal of the proofreading cycle, which operates via a different physical mechanism than the state-occupancy or speed-driven antiproofreading regimes discussed previously by Murugan et al. Peliti, Thermodynamics of accuracy in kinetic proofreading: dissipation and efficiency trade-offs, J.

Code Availability

The codes used to produce the Figures are available at DOI: 10.5281/zenodo. p. 6

Where the paper says it

CODE AVAILABILITY The codes used to produce the Figures are available at DOI: 10. 5281/zenodo.

p. 6
Where the paper says it

CODE AVAILABILITY The codes used to produce the Figures are available at DOI: 10. 5281/zenodo.

20057619.

Acknowledgments

We are grateful for the financial support from the Laufer Center for Physical and Quantitative Biology at Stony Brook, the John Templeton Foundation (Grant ID 62564), and NIH (Grant RM1-GM135136).

Author Contributions

YJY and KD conceptualized the project together, YJY developed the framework and the results, and KD supervised all developments.

References

[1] H. Qian, Phosphorylation Energy Hypothesis: Open Chemical Systems and Their Biological Functions, Annual Review of Physical Chemistry 58, 113 (2007). p. 7

Where the paper says it

Both authors contributed equally to writing and editing the manuscript. Qian, Phosphorylation Energy Hypothesis: Open Chemical Systems and Their Biological Functions, Annual Review of Physical Chemistry 58, 113 (2007).

[2] J. M. Horowitz and M. p. 7

Where the paper says it

Qian, Phosphorylation Energy Hypothesis: Open Chemical Systems and Their Biological Functions, Annual Review of Physical Chemistry 58, 113 (2007). Horowitz and M.

Esposito, Thermodynamics with Continuous Information Flow, Phys. p. 7
Where the paper says it

Horowitz and M. Esposito, Thermodynamics with Continuous Information Flow, Phys.

Rev. X 4, 031015 (2014). p. 7
Where the paper says it

Esposito, Thermodynamics with Continuous Information Flow, Phys. X 4, 031015 (2014).

[3] J. M. R. Parrondo, J. p. 7

Where the paper says it

X 4, 031015 (2014). Parrondo, J.

M. Horowitz, and T. p. 7
Where the paper says it

Qian, Phosphorylation Energy Hypothesis: Open Chemical Systems and Their Biological Functions, Annual Review of Physical Chemistry 58, 113 (2007). Horowitz and M.

Sagawa, Thermodynamics of information, Nature Phys 11, 131 (2015). p. 7
Where the paper says it

Horowitz, and T. Sagawa, Thermodynamics of information, Nature Phys 11, 131 (2015).

[4] U. Seifert, From Stochastic Thermodynamics to Thermodynamic Inference, Annual Review of Condensed Matter Physics 10, 171 (2023). p. 7

Where the paper says it

Sagawa, Thermodynamics of information, Nature Phys 11, 131 (2015). Seifert, From Stochastic Thermodynamics to Thermodynamic Inference, Annual Review of Condensed Matter Physics 10, 171 (2019).

[5] L. Peliti and S. p. 7

Where the paper says it

Seifert, From Stochastic Thermodynamics to Thermodynamic Inference, Annual Review of Condensed Matter Physics 10, 171 (2019). Peliti and S.

Pigolotti, Stochastic Thermodynamics: An Introduction (Princeton University Press, 2021). p. 7
Where the paper says it

Peliti and S. Pigolotti, Stochastic Thermodynamics: An Introduction (Princeton University Press, 2021).

[6] Y. Cao and S. Liang, Stochastic thermodynamics for biological functions, Quantitative Biology 13, e75 (2025), eprint: https://onlinelibrary. p. 7

Where the paper says it

Pigolotti, Stochastic Thermodynamics: An Introduction (Princeton University Press, 2021). Liang, Stochastic thermodynamics for biological functions, Quantitative Biology 13, e75 (2025), _eprint: https://onlinelibrary.

p. 7
Where the paper says it

Pigolotti, Stochastic Thermodynamics: An Introduction (Princeton University Press, 2021). Liang, Stochastic thermodynamics for biological functions, Quantitative Biology 13, e75 (2025), _eprint: https://onlinelibrary.

wiley.com/doi/pdf/10. p. 7
Where the paper says it

Liang, Stochastic thermodynamics for biological functions, Quantitative Biology 13, e75 (2025), _eprint: https://onlinelibrary. com/doi/pdf/10.

p. 7
Where the paper says it

Liang, Stochastic thermodynamics for biological functions, Quantitative Biology 13, e75 (2025), _eprint: https://onlinelibrary. com/doi/pdf/10.

1002/qub2.75.

[7] M. P. Leighton and D. p. 7

Where the paper says it

com/doi/pdf/10. Leighton and D.

A. Sivak, Flow of Energy and Information in Molecular Machines, Annual Review of Physical Chemistry 76, 379 (2025). p. 7
Where the paper says it

Leighton and D. Sivak, Flow of Energy and Information in Molecular Machines, Annual Review of Physical Chemistry 76, 379 (2025).

[8] C. Floyd, A. R. Dinner, A. Murugan, and S. p. 7

Where the paper says it

Sivak, Flow of Energy and Information in Molecular Machines, Annual Review of Physical Chemistry 76, 379 (2025). Murugan, and S.

Vaikuntanathan, Limits on the computational expressivity of non-equilibrium biophysical processes, Nat Commun 16, 7184 (2025). p. 7
Where the paper says it

Murugan, and S. Vaikuntanathan, Limits on the computational expressivity of non-equilibrium biophysical processes, Nat Commun 16, 7184 (2025).

[9] Y. Tu, Nonequilibrium Thermodynamics of Biochemical Networks: Energetics of Cellular Functions, Annual Reviews of Biophysics 10. p. 7

Where the paper says it

Vaikuntanathan, Limits on the computational expressivity of non-equilibrium biophysical processes, Nat Commun 16, 7184 (2025). Tu, Nonequilibrium Thermodynamics of Biochemical Networks: Energetics of Cellular Functions, Annual Reviews of Biophysics 10.

1146/annurev-biophys-021424-112247 (2026). p. 7
Where the paper says it

Tu, Nonequilibrium Thermodynamics of Biochemical Networks: Energetics of Cellular Functions, Annual Reviews of Biophysics 10. 1146/annurev-biophys-021424-112247 (2026).

[10] H. Qian and D. p. 7

Where the paper says it

Both authors contributed equally to writing and editing the manuscript. Qian, Phosphorylation Energy Hypothesis: Open Chemical Systems and Their Biological Functions, Annual Review of Physical Chemistry 58, 113 (2007).

A. Beard, Metabolic futile cycles and their functions: a systems analysis of energy and control, Syst Biol (Stevenage) 153, 192 (2006). p. 7
Where the paper says it

Qian and D. Beard, Metabolic futile cycles and their functions: a systems analysis of energy and control, Syst Biol (Stevenage) 153, 192 (2006).

[11] M. Baiesi and C. p. 7

Where the paper says it

Beard, Metabolic futile cycles and their functions: a systems analysis of energy and control, Syst Biol (Stevenage) 153, 192 (2006). Baiesi and C.

Maes, Life efficiency does not always increase with the dissipation rate, J. p. 7
Where the paper says it

Baiesi and C. Maes, Life efficiency does not always increase with the dissipation rate, J.

Phys. Commun. 2, 045017 (2018). p. 7
Where the paper says it

Maes, Life efficiency does not always increase with the dissipation rate, J. 2, 045017 (2018).

[12] C. Maes, Frenesy: Time-symmetric dynamical activity in nonequilibria, Physics Reports 850, 1 (2020). p. 7

Where the paper says it

2, 045017 (2018). Maes, Frenesy: Time-symmetric dynamical activity in nonequilibria, Physics Reports 850, 1 (2020).

[13] Y.-J. Yang and K. p. 0

Where the paper says it

Nonequilibrium Theory for Molecular Machine Design Ying-Jen Yang∗ Laufer Center of Physical and Quantitative Biology, Stony Brook University Ken A. Dill Laufer Center of Physical and Quantitative Biology, Stony Brook University Department of Physics and Astronomy, Stony Brook University and Department of Chemistry, Stony Brook University (Dated: May 12, 2026) Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production.

p. 0
Where the paper says it

Nonequilibrium Theory for Molecular Machine Design Ying-Jen Yang∗ Laufer Center of Physical and Quantitative Biology, Stony Brook University Ken A. Dill Laufer Center of Physical and Quantitative Biology, Stony Brook University Department of Physics and Astronomy, Stony Brook University and Department of Chemistry, Stony Brook University (Dated: May 12, 2026) Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production.

A. Dill, A principled basis for nonequilibrium network flows, Nat Commun 10. p. 7
Where the paper says it

Yang and K. Dill, A principled basis for nonequilibrium network flows, Nat Commun 10.

1038/s41467-026-71239-9 (2026). p. 7
Where the paper says it

Dill, A principled basis for nonequilibrium network flows, Nat Commun 10. 1038/s41467-026-71239-9 (2026).

[14] Y.-J. Yang and K. A. Dill, Fluctuation-Response Design Rules for Nonequilibrium Flows (2026), arXiv:2602. p. 7

Where the paper says it

Yang and K. Dill, Fluctuation-Response Design Rules for Nonequilibrium Flows (2026), arXiv:2602.

10957 [condmat]. p. 7
Where the paper says it

Dill, Fluctuation-Response Design Rules for Nonequilibrium Flows (2026), arXiv:2602. 10957 [condmat].

[15] E. Gerritsma and P. p. 2

Where the paper says it

To model the in vitro F1 motor assay (Fig. 2a), we adopt the 2-state, 2-edge network from Gerritsma and Gaspard [15] (Fig.

Gaspard, Chemomechanical Coupling and Stochastic Thermodynamics of the F1-ATPase Molecular Motor with an applied external toque, Biophys. p. 7
Where the paper says it

Gerritsma and P. Gaspard, Chemomechanical Coupling and Stochastic Thermodynamics of the F1 -ATPase Molecular Motor with an applied external toque, Biophys.

Rev. Lett. 05, 163 (2010). p. 7
Where the paper says it

Gaspard, Chemomechanical Coupling and Stochastic Thermodynamics of the F1 -ATPase Molecular Motor with an applied external toque, Biophys. 05, 163 (2010).

[16] J. Ninio, Kinetic amplification of enzyme discrimination, Biochimie 57, 587 (1975). p. 7

Where the paper says it

05, 163 (2010). Ninio, Kinetic amplification of enzyme discrimination, Biochimie 57, 587 (1975).

[17] J. J. Hopfield, Kinetic Proofreading: A New Mechanism for Reducing Errors in Biosynthetic Processes Requiring High Specificity, PNAS 71, 4135 (1974). p. 7

Where the paper says it

Ninio, Kinetic amplification of enzyme discrimination, Biochimie 57, 587 (1975). Hopfield, Kinetic Proofreading: A New Mechanism for Reducing Errors in Biosynthetic Processes Requiring High Specificity, PNAS 71, 4135 (1974).

[18] K. Banerjee, A. p. 7

Where the paper says it

Hopfield, Kinetic Proofreading: A New Mechanism for Reducing Errors in Biosynthetic Processes Requiring High Specificity, PNAS 71, 4135 (1974). Banerjee, A.

B. Kolomeisky, and O. p. 7
Where the paper says it

Banerjee, A. Kolomeisky, and O.

p. 7
Where the paper says it

Banerjee, A. Kolomeisky, and O.

p. 7
Where the paper says it

Banerjee, A. Kolomeisky, and O.

A. Igoshin, Elucidating interplay of speed and accuracy in biological error correction, Proceedings of the National Academy of Sciences 114, 5183 (2017). p. 7
Where the paper says it

Kolomeisky, and O. Igoshin, Elucidating interplay of speed and accuracy in biological error correction, Proceedings of the National Academy of Sciences 114, 5183 (2017).

[19] J. D. Mallory, A. p. 7

Where the paper says it

Igoshin, Elucidating interplay of speed and accuracy in biological error correction, Proceedings of the National Academy of Sciences 114, 5183 (2017). Mallory, A.

p. 7
Where the paper says it

Igoshin, Elucidating interplay of speed and accuracy in biological error correction, Proceedings of the National Academy of Sciences 114, 5183 (2017). Mallory, A.

B. Kolomeisky, and O. A. Igoshin, TradeOffs between Error, Speed, Noise, and Energy Dissipation in Biological Processes with Proofreading, J. p. 7
Where the paper says it

Kolomeisky, and O. Igoshin, TradeOffs between Error, Speed, Noise, and Energy Dissipation in Biological Processes with Proofreading, J.

Phys. Chem. B 123, 4718 (2019). p. 7
Where the paper says it

Igoshin, TradeOffs between Error, Speed, Noise, and Energy Dissipation in Biological Processes with Proofreading, J. B 123, 4718 (2019).

[20] W. W. Cleland, The kinetics of enzyme-catalyzed reactions with two or more substrates or products: II. p. 7

Where the paper says it

B 123, 4718 (2019). Cleland, The kinetics of enzyme-catalyzed reactions with two or more substrates or products: II.

Inhibition: Nomenclature and theory, Biochimica et Biophysica Acta (BBA) - Specialized Section on Enzymological Subjects 67, 173 (1963). p. 7
Where the paper says it

Cleland, The kinetics of enzyme-catalyzed reactions with two or more substrates or products: II. Inhibition: Nomenclature and theory, Biochimica et Biophysica Acta (BBA) - Specialized Section on Enzymological Subjects 67, 173 (1963).

[21] R. A. Copeland, Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacologists (Wiley-Interscience, Hoboken, N. p. 7

Where the paper says it

Inhibition: Nomenclature and theory, Biochimica et Biophysica Acta (BBA) - Specialized Section on Enzymological Subjects 67, 173 (1963). Copeland, Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacologists (Wiley-Interscience, Hoboken, N.

J, 2005).

[22] A. Cornish-Bowden, Fundamentals of Enzyme Kinetics (WileyVCH, Weinheim, 2012). p. 7

Where the paper says it

Copeland, Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacologists (Wiley-Interscience, Hoboken, N. Cornish-Bowden, Fundamentals of Enzyme Kinetics (WileyVCH, Weinheim, 2012).

[23] J. Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Rev. p. 7

Where the paper says it

Cornish-Bowden, Fundamentals of Enzyme Kinetics (WileyVCH, Weinheim, 2012). Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Rev.

Mod. Phys. 48, 571 (1976). p. 7
Where the paper says it

Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Rev. 48, 571 (1976).

[24] T. L. Hill, Free Energy Transduction and Biochemical Cycle Kinetics (Springer-Verlag, New York, 1989). p. 7

Where the paper says it

48, 571 (1976). Hill, Free Energy Transduction and Biochemical Cycle Kinetics (Springer-Verlag, New York, 1989).

[25] U. Seifert, Stochastic thermodynamics: From principles to the cost of precision, Physica A: Statistical Mechanics and its Applications 504, 176 (2018). p. 7

Where the paper says it

Hill, Free Energy Transduction and Biochemical Cycle Kinetics (Springer-Verlag, New York, 1989). Seifert, Stochastic thermodynamics: From principles to the cost of precision, Physica A: Statistical Mechanics and its Applications 504, 176 (2018).

[26] C. Maes, Local detailed balance, SciPost Phys. p. 7

Where the paper says it

Seifert, Stochastic thermodynamics: From principles to the cost of precision, Physica A: Statistical Mechanics and its Applications 504, 176 (2018). Maes, Local detailed balance, SciPost Phys.

Lect. Notes, 32 (2021). p. 7
Where the paper says it

Maes, Local detailed balance, SciPost Phys. Notes , 32 (2021).

[27] G. Falasco and M. p. 7

Where the paper says it

Notes , 32 (2021). Falasco and M.

Esposito, Local detailed balance across scales: From diffusions to jump processes and beyond, Phys. p. 7
Where the paper says it

Falasco and M. Esposito, Local detailed balance across scales: From diffusions to jump processes and beyond, Phys.

Rev. E 103, 042114 (2021). p. 7
Where the paper says it

Esposito, Local detailed balance across scales: From diffusions to jump processes and beyond, Phys. E 103, 042114 (2021).

[28] J. A. Owen, T. R. Gingrich, and J. p. 7

Where the paper says it

E 103, 042114 (2021). Gingrich, and J.

M. Horowitz, Universal Thermodynamic Bounds on Nonequilibrium Response with Biochemical Applications, Phys. p. 7
Where the paper says it

Gingrich, and J. Horowitz, Universal Thermodynamic Bounds on Nonequilibrium Response with Biochemical Applications, Phys.

Rev. X 10, 011066 (2020). p. 7
Where the paper says it

Horowitz, Universal Thermodynamic Bounds on Nonequilibrium Response with Biochemical Applications, Phys. X 10, 011066 (2020).

[29] G. Fernandes Martins and J. p. 7

Where the paper says it

X 10, 011066 (2020). Fernandes Martins and J.

M. Horowitz, Topologically constrained fluctuations and thermodynamics regulate nonequilibrium response, Phys. p. 7
Where the paper says it

Fernandes Martins and J. Horowitz, Topologically constrained fluctuations and thermodynamics regulate nonequilibrium response, Phys.

Rev. E 108, 044113 (2023). p. 7
Where the paper says it

Horowitz, Topologically constrained fluctuations and thermodynamics regulate nonequilibrium response, Phys. E 108, 044113 (2023).

[30] T. Aslyamov and M. p. 7

Where the paper says it

E 108, 044113 (2023). Aslyamov and M.

Esposito, Nonequilibrium Response for Markov Jump Processes: Exact Results and Tight Bounds, Phys. p. 7
Where the paper says it

Aslyamov and M. Esposito, Nonequilibrium Response for Markov Jump Processes: Exact Results and Tight Bounds, Phys.

Rev. Lett. 132, 037101 (2024). p. 7
Where the paper says it

Esposito, Nonequilibrium Response for Markov Jump Processes: Exact Results and Tight Bounds, Phys. 132, 037101 (2024).

[31] See Supplementary Information (SI) at [URL] for more details. p. 7

Where the paper says it

132, 037101 (2024). [31] See Supplementary Information (SI) at [URL] for more details.

[32] E. T. Jaynes, The Minimum Entropy Production Principle, Annual Review of Physical Chemistry 31, 579 (1980). p. 7

Where the paper says it

[31] See Supplementary Information (SI) at [URL] for more details. Jaynes, The Minimum Entropy Production Principle, Annual Review of Physical Chemistry 31, 579 (1980).

[33] S. Pressé, K. p. 7

Where the paper says it

Jaynes, The Minimum Entropy Production Principle, Annual Review of Physical Chemistry 31, 579 (1980). Pressé, K.

Ghosh, J. Lee, and K. p. 7
Where the paper says it

Pressé, K. Lee, and K.

A. Dill, Principles of maximum entropy and maximum caliber in statistical physics, Rev. p. 7
Where the paper says it

Lee, and K. Dill, Principles of maximum entropy and maximum caliber in statistical physics, Rev.

Mod. Phys. 85, 1115 (2013). p. 7
Where the paper says it

Dill, Principles of maximum entropy and maximum caliber in statistical physics, Rev. 85, 1115 (2013).

[34] J. D. Mallory, A. B. Kolomeisky, and O. A. Igoshin, Kinetic control of stationary flux ratios for a wide range of biochemical processes, Proceedings of the National Academy of Sciences 117, 8884 (2020). p. 7

Where the paper says it

Kolomeisky, and O. Igoshin, Kinetic control of stationary flux ratios for a wide range of biochemical processes, Proceedings of the National Academy of Sciences 117, 8884 (2020).

[35] J. A. Wagoner and K. p. 7

Where the paper says it

Igoshin, Kinetic control of stationary flux ratios for a wide range of biochemical processes, Proceedings of the National Academy of Sciences 117, 8884 (2020). Wagoner and K.

A. Dill, Mechanisms for achieving high speed and efficiency in biomolecular machines, PNAS 116, 5902 (2019). p. 7
Where the paper says it

Wagoner and K. Dill, Mechanisms for achieving high speed and efficiency in biomolecular machines, PNAS 116, 5902 (2019).

[36] A. I. Brown and D. p. 7

Where the paper says it

Dill, Mechanisms for achieving high speed and efficiency in biomolecular machines, PNAS 116, 5902 (2019). Brown and D.

A. Sivak, Theory of Nonequilibrium Free Energy Transduction by Molecular Machines, Chem. p. 7
Where the paper says it

Brown and D. Sivak, Theory of Nonequilibrium Free Energy Transduction by Molecular Machines, Chem.

Rev. 120, 434 (2020). p. 7
Where the paper says it

Sivak, Theory of Nonequilibrium Free Energy Transduction by Molecular Machines, Chem. 120, 434 (2020).

[37] H. Kacser and J. p. 7

Where the paper says it

120, 434 (2020). Kacser and J.

A. Burns, The control of flux, Biochem Soc Trans 23, 341 (1995). p. 7
Where the paper says it

Kacser and J. Burns, The control of flux, Biochem Soc Trans 23, 341 (1995).

[38] R. Heinrich and T. p. 7

Where the paper says it

Burns, The control of flux, Biochem Soc Trans 23, 341 (1995). Heinrich and T.

A. Rapoport, A Linear SteadyState Treatment of Enzymatic Chains, European Journal of Biochemistry 42, 89 (1974), eprint: https://febs. p. 7
Where the paper says it

Heinrich and T. Rapoport, A Linear SteadyState Treatment of Enzymatic Chains, European

onlinelibrary.wiley.com/doi/pdf/10.1111/j.1432-

[39] A. Murugan, D. p. 7

Where the paper says it

Sivak, Flow of Energy and Information in Molecular Machines, Annual Review of Physical Chemistry 76, 379 (2025). Murugan, and S.

A. Huse, and S. p. 8
Where the paper says it

Murugan, D. Huse, and S.

Leibler, Speed, dissipation, and error in kinetic proofreading, Proceedings of the National Academy of Sciences 109, 12034 (2012). p. 8
Where the paper says it

Huse, and S. Leibler, Speed, dissipation, and error in kinetic proofreading, Proceedings of the National Academy of Sciences 109, 12034 (2012) .

[40] P. Sartori and S. p. 8

Where the paper says it

Leibler, Speed, dissipation, and error in kinetic proofreading, Proceedings of the National Academy of Sciences 109, 12034 (2012) . Sartori and S.

Pigolotti, Thermodynamics of Error Correction, Phys. p. 8
Where the paper says it

Sartori and S. Pigolotti, Thermodynamics of Error Correction, Phys.

Rev. X 5, 041039 (2015). p. 8
Where the paper says it

Pigolotti, Thermodynamics of Error Correction, Phys. X 5, 041039 (2015) .

[41] We define “anti-proofreading” by the net reversal of the proofreading cycle, which operates via a different physical mechanism than the state-occupancy or speed-driven antiproofreading regimes discussed previously by Murugan et al. p. 8

Where the paper says it

X 5, 041039 (2015) . [41] We define “anti-proofreading” by the net reversal of the proofreading cycle, which operates via a different physical mechanism than the state-occupancy or speed-driven antiproofreading regimes discussed previously by Murugan et al.

[?].

[42] R. Rao and L. Peliti, Thermodynamics of accuracy in kinetic proofreading: dissipation and efficiency trade-offs, J. p. 8

Where the paper says it

[41] We define “anti-proofreading” by the net reversal of the proofreading cycle, which operates via a different physical mechanism than the state-occupancy or speed-driven antiproofreading regimes discussed previously by Murugan et al. Peliti, Thermodynamics of accuracy in kinetic proofreading: dissipation and efficiency trade-offs, J.

Stat. Mech. 2015, P06001 (2015). p. 8
Where the paper says it

Peliti, Thermodynamics of accuracy in kinetic proofreading: dissipation and efficiency trade-offs, J. 2015, P06001 (2015) .

[43] D. Chiuchiu, S. p. 8

Where the paper says it

2015, P06001 (2015) . Chiuchiu, S.

Mondal, and S. p. 8
Where the paper says it

Chiuchiu, S. Mondal, and S.

Pigolotti, Pareto optimal fronts of kinetic proofreading, New J. p. 8
Where the paper says it

Mondal, and S. Pigolotti, Pareto optimal fronts of kinetic proofreading, New J.

Phys. 25, 043007 (2605). p. 0
Where the paper says it

In some situations, molecular machines are driven by differences across two or more equilibrium reservoirs and the flows are constrained by the Local Detailed Balance (LDB) condition [25–27]. This fixes the cycle affinity of any arXiv:2605.

[44] A. Goetz, J. Barrios, R. p. 8

Where the paper says it

25, 043007 (2023) . Barrios, R.

Madsen, and P. p. 8
Where the paper says it

Barrios, R. Madsen, and P.

Dixit, Non-equilibrium strategies for ligand specificity in signaling networks, eLife 14, 10. p. 8
Where the paper says it

Madsen, and P. Dixit, Non-equilibrium strategies for ligand specificity in signaling networks, eLife 14 , 10.

7554/eLife. p. 8
Where the paper says it

Dixit, Non-equilibrium strategies for ligand specificity in signaling networks, eLife 14 , 10. 7554/eLife.

107524.

[45] Y.-J. Yang, C. D. Kocher, and K. p. 8

Where the paper says it

7554/eLife. Kocher, and K.

A. Dill, Nonequilibrium Theory for Adaptive Systems in Varying Environments (2025), arXiv:2506. p. 8
Where the paper says it

Kocher, and K. Dill, Nonequilibrium Theory for Adaptive Systems in Varying Environments (2025), arXiv:2506.

19018 [cond-mat]. p. 8
Where the paper says it

Dill, Nonequilibrium Theory for Adaptive Systems in Varying Environments (2025), arXiv:2506. 19018 [cond-mat].

[46] R. Chetrite and H. p. 8

Where the paper says it

19018 [cond-mat]. Chetrite and H.

Touchette, Nonequilibrium Microcanonical and Canonical Ensembles and Their Equivalence, Phys. p. 8
Where the paper says it

Chetrite and H. Touchette, Nonequilibrium Microcanonical and Canonical Ensembles and Their Equivalence, Phys.

Rev. Lett. 111, 120601 (2013). p. 8
Where the paper says it

Touchette, Nonequilibrium Microcanonical and Canonical Ensembles and Their Equivalence, Phys. 111, 120601 (2013) .

[47] H. Touchette, The large deviation approach to statistical mechanics, Physics Reports 478, 1 (2009). p. 8

Where the paper says it

111, 120601 (2013) . Touchette, The large deviation approach to statistical mechanics, Physics Reports 478, 1 (2009) .

Supplementary Information

The Supplementary Information for “Nonequilibrium Theory for Molecular Machine Design” is available at DOI: 10.5281/zenodo.20057619.

Contents

The contents provide a roadmap for the paper, detailing CFT review, coordinate construction, and specific examples for F1-ATPase, kinetic proofreading, and enzyme inhibition.

I. A Brief Review of Caliber Force Theory (CFT)

This section recapitulates the foundations of CFT by detailing the construction of conjugate coordinates via Maximum Caliber [13]. It details the derivation of the Response-Inverse-Matrix (RIM) relation [14]. p. 9

Where the paper says it

Caliber Force Theory (CFT) is designed to mirror this exact conjugate structure for nonequilibrium (NEQ) flows within a network state space. In this section, we recapitulate the foundations of CFT, detailing the construction of conjugate coordinates via Maximum Caliber [13], the derivation of the Response-Inverse-Matrix (RIM) relation [14], and the explicit proofs for the Arrhenius energetic perturbation rules utilized in the main text.

**A.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Traffic Control for Molecular Motors: The AI can design faster molecular motors by treating dynamics as a flow network and optimizing speed through traffic control, specifically by routing flows via kinetic barriers to correct futile backstepping, as shown in Example 1.

  2. Decoupling Performance Metrics in Proofreaders: The AI system can optimize the speed, accuracy, and energy cost of kinetic proofreading networks by recognizing that these trade-offs are boundary constraints and not absolute within the operational interior.

  3. Topological Routing for Enzyme Inhibition: The AI can design better enzyme inhibitors by distinguishing between competitive/uncompetitive inhibitors (topological dead-ends optimized solely by node energies) and noncompetitive inhibitors (leaky loops requiring kinetic barrier routing for optimization).

  4. Inverse Design of Transition Rates: The system can map desired functional performance metrics like speed, error, and cost directly back to the required microscopic transition rate configurations using the derived coordinate transformation (S56–S60).

  5. Predictive Modeling of Kinetic Capacity: The AI can predict the physical performance bounds for proofreading networks by calculating constraints like kinetic capacity which dictates that productive incorporation (the first term) and futile proofreading (the second term) must compete for a finite kinetic bandwidth.

Abstract

Modeling the dynamical flows on networks of biomolecular machines often entails computing node populations and edge fluxes with Master Equations and correlating machine performance with entropy production. But this alone is not sufficient for design, optimization and evolution because it doesn't treat cost-benefit tradeoffs, or small-system misflows (backsteps, futile cycles, ineffective actions), or differential properties for flow design. Here we develop CFT Design, based on the recently developed Caliber Force Theory (CFT). We apply it to: designing faster molecular motors through ``traffic control''; optimizing speed, energy, and accuracy in kinetic proofreaders; and designing better enzyme inhibitors. CFT Design provides a general framework for optimizing nonequilibrium flow networks.

Sources

Related papers