Computing and Bounding Equilibrium Concentrations in Athermic Chemical Systems

David Soloveichik; Hamidreza Akef; Minki Hhan

REVIEW 4 major objections 3 minor 19 references

Computing and Bounding Equilibrium Concentrations in Athermic Chemical Systems

T0 review · 4 major / 3 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 For any stable on-target polymer set, an iterative leveling algorithm assigns concentration exponents so that the configuration $c^{\bar{\mu}(P)}$ is the exact equilibrium, with every off-target polymer exponentially suppressed.

desk verdict The leveling algorithm is a genuine contribution, but Theorem 5.4 overreaches with a false universal-c claim and the application section rests on unproven worst-case reactions. read the letter →

arxiv 2507.12699 v1 pith:4CK3FPNO submitted 2025-07-17 cs.DS q-bio.MN

Hamidreza Akef , Minki Hhan , David Soloveichik This is my paper · ORCID

classification cs.DSq-bio.MN MSC 92E20

keywords equilibriumconcentrationsathermicsystemsdetailedbalancethermodynamicbindingnetworksmonomer-polymermodelleaksuppressionDNAstranddisplacementiterativelevelingalgorithm

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

The paper asks whether equilibrium concentrations of assembled molecular complexes can be set by design rather than found by numerical simulation, and answers it in the athermic case where all interactions are enthalpy-neutral. It proves that from a chosen high-concentration set $S$, an iterative leveling algorithm extends the assignment to every polymer: at each step the newly introduced polymers receive the concentration exponent that balances the canonical reaction with the smallest imbalance-to-novelty ratio $k(\alpha)/l(\alpha)$. When $S$ is stable, meaning every canonical reaction that creates something new satisfies $k(\alpha)/l(\alpha)>1$, the resulting configuration is the minimum of the entropic free energy under mass conservation, and all off-target polymers have concentration exponents above 1, hence concentrations below the on-target baseline $c$. This turns combinatorial entropy-loss arguments in the Thermodynamic Binding Networks (TBN) model into explicit real-valued concentration bounds, and the paper demonstrates the bounds on a DNA AND gate and on a parameterized translator cascade, where it predicts exponentially decreasing leak once waste concentrations are tuned.

What carries the argument

The load-bearing mechanism is the level-by-level construction over canonical reactions. A canonical reaction has all reactants in the on-target set; its $i$th-level imbalance $k_i(\alpha)=\bar{\mu}(M_1)-\bar{\mu}(\hat{M}_2)$ measures the exponent surplus on the known side, and its novelty $l_i(\alpha)$ counts how many new off-target polymers the reaction introduces. The ratio $k_i(\alpha)/l_i(\alpha)$ is the effective price of creating new species: high imbalance makes products expensive, while high novelty makes them entropically cheap. Algorithm 1 sets $\mu_i$ to the minimum such ratio, assigns that same exponent to every polymer first appearing in a minimizing reaction, and repeats until all polymers are levelized. Stability of $S$ is exactly the condition that the first such ratio is strictly greater than 1, forcing the exponent floor $\mu_1>1$. The Hilbert-basis argument reduces the infinite set of canonical reactions to a finite generating set, making the minimum well-defined and the algorithm terminating, and Lemma 5.3 converts balance on canonical reactions into the true minimum of the entropic free energy.

What would settle it

Compute the finite Hilbert basis of canonical reactions for a proposed stable set $S$ and test $k(\alpha)/l(\alpha)>1$ on every basis element with $l(\alpha)>0$; any ratio $\le 1$ refutes the conclusion that all off-target exponents exceed 1. For the translator cascade, solve the detailed-balance equations at the specified fuel and waste concentrations and check whether the leak output is below $c^{1/4+N/4}$ and whether more than half the signal reaches the output layer; failure of either inequality would falsify the worst-case-reaction claim.

Watch

Extended reading notes

Core claim

The central claim is Theorem 5.4: if $S$ is stable with concentration exponents $\mu$, then Algorithm 1 produces extended exponents $\bar{\mu}$ such that for any $0<c<1$ there exist monomer concentrations $x_0$ with the configuration $x_P = c^{\bar{\mu}(P)}$ minimizing $g(x)=\sum_{P\in\Psi} x_P(\log x_P-1)$ subject to $A\cdot x=x_0$, and every off-target polymer satisfies $\bar{\mu}(P)\ge \mu_1>1$. The proof shows every canonical reaction is balanced in that configuration: looking at the highest level among a reaction's products forces the level-defining minimum to contradict itself, and the opposite direction reduces to this case by summing the levelizing reactions of the products. A lemma shows that balancing canonical reactions is sufficient for a global free-energy minimum, because arbitrary reactions decompose into canonical ones. As a corollary, a TBN-stability-closed set with uniform on-target exponents has off-target concentrations bounded by $c^{\min_\alpha(e(\alpha)/l(\alpha))+1}$, where $e(\alpha)$ is entropy loss and $l(\alpha)$ is novelty of a canonical reaction.

Load-bearing premise

The construction rests on $S$ being stable: every canonical reaction that introduces at least one off-target polymer must have $k(\alpha)/l(\alpha)>1$, and for the TBN applications this condition is checked only informally by arguing that particular reactions are worst-case.

Editorial extensions

If this is right

For any stable on-target set, equilibrium concentrations are obtained by the leveling recursion instead of by numerically minimizing the free energy, and each early stop of the algorithm yields a valid upper bound on every not-yet-levelized off-target polymer.
In TBN systems, any TBN-stability-closed set with uniform $\mu=1$ gives explicit leak bounds of the form $c^{\min(e/l)+1}$, so combinatorial entropy-loss statements become quantitative concentration guarantees.
For the DNA AND gate, the analysis gives leak $\le c^{1.5}$ without inputs and $\le c^{1.33}$ with one input present, so leak is polynomially suppressed as concentrations shrink.
For two translators in cascade, uniform fuel concentrations only tighten the leak bound toward $c^2$ as redundancy $N$ grows; setting waste at concentration $c$ and fuel at $2c$ instead makes the bound $c^{1/4+N/4}$, giving exponential leak suppression in $N$ while preserving a constant fraction of signal reaching the output.
All of these consequences hold for equilibrium concentrations under athermic, enthalpy-neutral conditions, the regime of saturated strongly bonded DNA domains.

Reading between the lines

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

The ratio-minimization view suggests a computational test for stability: enumerate the finite Hilbert basis of canonical reactions and check $k(\alpha)/l(\alpha)>1$ on that generating set, turning the informal worst-case arguments used for the applications into a verification algorithm.
The same leveling recursion can be read as a min-ratio path problem on the reaction cone, so efficient combinatorial optimization algorithms may yield the worst-case ratios and hence the concentration exponents without enumerating all reactions.
If the framework were extended to per-polymer free energies $\Delta G_P$, the imbalance would likely become $\bar{\mu}(M_1)-\bar{\mu}(\hat{M}_2)$ shifted by free-energy terms, giving a natural route toward toehold-mediated, non-athermic DNA systems.
The translator-cascade analysis implies a design principle: to make leak decay with a redundancy parameter, keep the novelty $l(\alpha)$ of the worst leak reaction constant while growing its imbalance, which is achieved precisely by nonuniform on-target concentration exponents.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Dear colleague,

The leveling algorithm is the real thing. The idea of assigning concentration exponents level by level, using the ratio of imbalance to novelty, is new and produces explicit upper bounds on off-target concentrations that the log-linear approach cannot easily provide. Corollary 7.2, connecting TBN entropy loss to concentration exponents, is clean and useful. And the translator cascade analysis—showing that increasing redundancy N alone does not arbitrarily suppress equilibrium leak—is a genuinely valuable corrective to the 'leakless' label. The Hilbert basis appendix is a thoughtful way around the infinite set of canonical reactions.

That said, the main theorem as stated is false. Theorem 5.4 says that for any 01/2. So the universal quantifier over c must be replaced by a small-c condition or an explicit bound on c. The paper actually cares about the dilute regime, so this is fixable, but it is a real error in the central existence claim.

The proof of Theorem 5.4 has another gap in Case 2. The reduction sums levelizing reactions β_P for every P in M2, but on-target products in M2 have no levelizing reaction, so the sum is not defined for canonical reactions whose products include on-target polymers. I suspect the case can be patched, but the proof as written is incomplete.

Section 8's worst-case reaction claims are explicitly informal. The leak bounds for the AND gate and the translator cascade depend on the claim that the identified reactions minimize e/l or k/l, and the authors admit they have no proof. For the translator, the qualitative point that novelty grows with N is likely robust, but the exact exponential bound should be treated as conditional on those unproven claims.

This paper deserves a serious referee. The framework is novel, the connection between TBN combinatorics and real-valued concentrations is valuable, and the flaws are localized to the theorem statement, one proof case, and the informal application claims. I would send it to peer review, requiring the c-quantifier fix and the Case 2 patch; the application section should either contain proofs or be explicitly labeled as conjectural. If you work on DNA leak, cite the leveling framework and the redundancy result, but verify the worst-case claims before relying on the bounds.

Referee Report

4 major / 3 minor

Summary. The paper studies equilibrium concentrations in an athermic polymer-monomer model, where polymers are multisets of monomers and the free energy is purely entropic. Given a set S of on-target polymers with prescribed concentration exponents mu, the authors define canonical reactions, their imbalance k(alpha) and novelty l(alpha), and a stability condition k(alpha)/l(alpha) > 1. Algorithm 1 iteratively assigns extended exponents bar-mu to all polymers by levelizing canonical reactions. Theorem 5.4 claims that, for any 0 < c < 1, there exist monomer concentrations x0 in (0,1) such that setting each polymer P to c^{bar-mu(P)} minimizes the entropic free energy subject to mass conservation, with all off-target exponents strictly above 1. The paper then derives a bound based on TBN-stability (Corollary 7.2) and applies it to an AND gate and to translator cascades, arguing that leak concentrations scale polynomially in c and, in a modified parameter regime, exponentially in the redundancy parameter.

Significance. If the main theorem is correct, the levelizing construction gives a genuinely useful bridge between combinatorial TBN entropy-loss arguments and real-valued equilibrium concentrations, and the paper's explicit algorithm and Hilbert-basis treatment (Appendix A) are valuable contributions. The monotonicity lemma (Lemma 6.3) and the bound framework (Section 6) are also interesting and potentially reusable. However, the central existence theorem is currently false as stated because of the unrestricted quantifier over c, and the proof contains a genuine gap in Case 2. The application results in Section 8 are explicitly conditional on unproven worst-case reaction identifications. These issues materially affect the paper's main claims, so the significance is real but conditional on a substantial revision.

major comments (4)

[Theorem 5.4 and Definition 2.1] The statement 'for any 0 < c < 1' is false because mass conservation forces x0 = A·x, and the entries of x0 must lie in (0,1). A concrete counterexample is S = {A} with A = {a}, an off-target polymer B = {a,a}, and uniform mu(A) = 1. S is stable and Algorithm 1 assigns bar-mu(B) = 2. For c = 0.9, the configuration x_A = 0.9, x_B = 0.81 lies in (0,1)^Psi and satisfies detailed balance, but x0(a) = x_A + 2 x_B = 2.52, contradicting x0 in (0,1)^Psi0. The theorem needs a small-c quantifier or an explicit upper bound on c; this is not a presentation issue but a load-bearing error in the main existence result.
[Theorem 5.4, proof, Case 2] The reduction in Case 2 is undefined for canonical reactions with on-target products. The proof says to choose, for each polymer P in M2, a levelizing reaction beta_P that includes P as a product, and then sums M2[P] copies of beta_P. But if P belongs to the on-target set S, there is no levelizing reaction for P, because levelizing reactions are canonical reactions with l(alpha) != 0 that attain the minimum ratio, whereas on-target polymers are level 0 and are not assigned by any levelizing reaction. Thus the sum over P in M2 cannot be formed when M2 intersects S. The proof needs a separate argument for on-target products before the contradiction can be concluded.
[Lemma 5.3] The proof of Lemma 5.3 is only a sketch and does not rigorously establish that every reaction is a combination of canonical reactions. The displayed derivation replaces alpha by a reaction involving M1 + M1' and M2 + M2', but it is not shown that repeating this procedure terminates in a reaction whose reactants all lie in S, nor that the final combined reaction is equivalent to the original one in the sense needed for detailed balance. Since Lemma 5.3 is the bridge between balancing canonical reactions and minimizing g(x) subject to mass conservation, this gap is load-bearing and requires a complete proof.
[Section 8 and Corollary 7.2] The application bounds are not established because the worst-case canonical reactions are asserted without proof. The paper explicitly states in Section 8 that the authors 'claim to have identified the worst-case canonical reactions' and in the Discussion that 'the argument is informal.' Consequently the concrete leak bounds c^{1.5}, c^{1.33}, c^{4/3}, and c^{1/4+N/4} are conditional on unverified stability and worst-case assumptions. Either a proof of the worst-case identification must be supplied, or the statements must be reworded as conjectures or heuristic bounds.

minor comments (3)

[Remark 3.2] The footnote in Remark 3.2 already acknowledges that mole-fraction concentrations require 'the regime of less polymer than solvent,' and this should be reconciled with the unrestricted 'any 0 < c < 1' quantifier in Theorem 5.4 and Corollary 7.2.
[Algorithm 1, line 12] The sentence 'Append all polymers P in M2 that are not in hat(M2) to S_i and assigns bar-mu(P) = mu_i' has a subject-verb agreement issue and should be rephrased, for example as 'assign bar-mu(P) = mu_i to each such P.'
[Definitions 5.1 and 5.2] The notation hat(M2) is reused in Definition 5.1 for the level-dependent intersection with the union of previously levelized sets, while the same symbol is used in Definition 3.3 for M2 intersect S. The dependency on i should be made explicit, such as hat(M2)_i, to avoid confusion.

Circularity Check

0 steps flagged · score 1.0 of 10

No material circularity: the main theorem is a constructive derivation from the stated stability assumption; the paper's own caveats about informal worst-case claims are soundness gaps, not circular steps.

full rationale

The derivation chain is self-contained rather than circular. Algorithm 1 and Theorem 5.4 take as inputs the on-target set S, concentration exponents mu, and the canonical reactions, and then construct extended exponents mu-bar and prove that the resulting configuration is balanced and hence minimizes the entropic free energy. The off-target bound mu-bar(P) > 1 is not assumed: it follows from the stability premise (k(alpha)/l(alpha) > 1 for all canonical reactions) via the definition of mu_1 and the monotonicity lemma (Lemma 6.3), so the conclusion is a derived consequence, not a restatement of the input. The paper's references to prior TBN work, including work co-authored by the present authors, are used in the application sections to supply entropy-loss facts; they are not the justification for the main existence theorem, and no uniqueness theorem or ansatz is imported from those citations into the core construction. The manuscript explicitly flags its own limitation in Section 8 ('we claim to have identified the worst-case canonical reactions... without proof') and in Section 9 ('While we believe the canonical reactions highlighted in Section 8 are indeed worst-case, the argument is informal'). This is an acknowledged soundness gap in the application analysis, not circularity. Likewise, any issue with the universal quantifier over c in Theorem 5.4's claim that x0 lies in (0,1) would be a correctness defect of the stated theorem, not a circular reduction: the theorem's assertion is not equivalent to its assumptions by construction. Overall, no load-bearing step in the paper reduces to its own inputs. The self-citations exist but are not load-bearing, and the central derivation is independent of them, so the appropriate circularity score is low.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced; the levels, imbalance, and novelty are analytical constructs. The main free choices are the base concentration c, the on-target exponents μ, and application-specific parameters such as y; these are explicit design inputs, not hidden fitted values.

free parameters (3)

base concentration c = user-chosen, 0<c<1
The paper fixes c as the mole-fraction concentration for on-target polymers with exponent 1; the theorem holds for any c<1, but in practice c must be small enough for monomer concentrations to remain below 1.
on-target concentration exponents μ(P) = user-chosen in (0,1]
The desired concentrations of on-target polymers are set via μ; the algorithm extends μ to off-target polymers. In the translator example, μ(F_i)=1+log_c 2 and μ(W_i)=1 are chosen by hand.
leak bound parameter y = y≤c/4
In Section 8, the analysis assumes the output concentration y is at most c/4 to derive k(β)≥n/4; this is a design constraint, not fitted to data.

assumptions (4)

domain assumption Athermicity: Σ_P x_P log Ω_P is constant across all configurations satisfying mass conservation (Section 2).
This reduces the free energy to pure entropy (Eq 3). It is justified by the strongly bonded, saturated regime of DNA domains, consistent with TBN configurations.
standard math Free energy minimization equals detailed balance (Theorem B.1, Appendix B).
Borrowed from chemical reaction network theory (Horn-Jackson, Feinberg); the paper provides a sketch.
standard math Hilbert basis existence and computability (Appendix A).
Used to reduce the infinite set of canonical reactions to a finite basis so that the minima in Definition 5.2 are attained and Algorithm 1 terminates.
domain assumption Dilute-solution mole fraction regime with c<1 (Remark 3.2).
The free energy form (Eq 2) from Dirks et al. assumes polymer concentrations are mole fractions less than solvent; the paper assumes 0<c<1 throughout.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Computing and Bounding Equilibrium Concentrations in Athermic Chemical Systems." pith.science (2026). https://pith.science/paper/4CK3FPNO

@misc{pith2026250712699,
  author       = {Pith},
  title        = {Pith review of: Computing and Bounding Equilibrium Concentrations in Athermic Chemical Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4CK3FPNO}},
  note         = {Machine review of arXiv:2507.12699}
}

read the original abstract

Computing equilibrium concentrations of molecular complexes is generally analytically intractable and requires numerical approaches. In this work we focus on the polymer-monomer level, where indivisible molecules (monomers) combine to form complexes (polymers). Rather than employing free-energy parameters for each polymer, we focus on the athermic setting where all interactions preserve enthalpy. This setting aligns with the strongly bonded (domain-based) regime in DNA nanotechnology when strands can bind in different ways, but always with maximum overall bonding -- and is consistent with the saturated configurations in the Thermodynamic Binding Networks (TBNs) model. Within this context, we develop an iterative algorithm for assigning polymer concentrations to satisfy detailed-balance, where on-target (desired) polymers are in high concentrations and off-target (undesired) polymers are in low. Even if not directly executed, our algorithm provides effective insights into upper bounds on concentration of off-target polymers, connecting combinatorial arguments about discrete configurations such as those in the TBN model to real-valued concentrations. We conclude with an application of our method to decreasing leak in DNA logic and signal propagation. Our results offer a new framework for design and verification of equilibrium concentrations when configurations are distinguished by entropic forces.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 19 canonical work pages

[1]

Programming substrate-independent kinetic barriers with thermodynamic binding networks

Keenan Breik, Cameron Chalk, David Doty, David Haley, and David Soloveichik. Programming substrate-independent kinetic barriers with thermodynamic binding networks. IEEE/ACM Trans. Comput. Biol. Bioinformatics , 18(1):283--295, 2021. https://doi.org/10.1109/TCBB.2019.2959310 doi:10.1109/TCBB.2019.2959310

work page arXiv 2021
[2]

Computing properties of stable configurations of thermodynamic binding networks

Keenan Breik, Chris Thachuk, Marijn Heule, and David Soloveichik. Computing properties of stable configurations of thermodynamic binding networks. Theor. Comput. Sci. , 785:17--29, 2019. https://doi.org/10.1016/j.tcs.2018.10.027 doi:10.1016/j.tcs.2018.10.027

work page doi:10.1016/j.tcs.2018.10.027 2019
[3]

Bruns, B

W. Bruns, B. Ichim, C. S\"oger, and U. von der Ohe. Normaliz. A lgorithms for rational cones and affine monoids. Available at https://www.normaliz.uni-osnabrueck.de

work page
[4]

How far is complex balancing from detailed balancing? Bulletin of mathematical biology , 73(4):811--828, 2011

Alicia Dickenstein and Mercedes P \'e rez Mill \'a n. How far is complex balancing from detailed balancing? Bulletin of mathematical biology , 73(4):811--828, 2011

work page 2011
[5]

Thermodynamic analysis of interacting nucleic acid strands

Robert M Dirks, Justin S Bois, Joseph M Schaeffer, Erik Winfree, and Niles A Pierce. Thermodynamic analysis of interacting nucleic acid strands. SIAM review , 49(1):65--88, 2007

work page 2007
[6]

Rogers, David Soloveichik, Chris Thachuk, and Damien Woods

David Doty, Trent A. Rogers, David Soloveichik, Chris Thachuk, and Damien Woods. Thermodynamic binding networks. In DNA Computing and Molecular Programming, 23rd International Conference, DNA 23 , pages 249--266, 2017. https://doi.org/10.1007/978-3-319-66799-7_16 doi:10.1007/978-3-319-66799-7_16

work page doi:10.1007/978-3-319-66799-7_16 2017
[7]

Necessary and sufficient conditions for detailed balancing in mass action systems of arbitrary complexity

Martin Feinberg. Necessary and sufficient conditions for detailed balancing in mass action systems of arbitrary complexity. Chemical Engineering Science , 44(9):1819--1827, 1989

work page 1989
[8]

Foundations of chemical reaction network theory

Martin Feinberg. Foundations of chemical reaction network theory . Springer, 2019

work page 2019

Show all 19 references

[9]

COPASI ---a complex pathway simulator

Stefan Hoops, Sven Sahle, Ralph Gauges, Christine Lee, J \"u rgen Pahle, Natalia Simus, Mudita Singhal, Liang Xu, Pedro Mendes, and Ursula Kummer. COPASI ---a complex pathway simulator. Bioinformatics , 22(24):3067--3074, 2006

2006
[10]

General mass action kinetics

Fritz Horn and Roy Jackson. General mass action kinetics. Archive for rational mechanics and analysis , 47:81--116, 1972

1972
[11]

A deficiency-based approach to parametrizing positive equilibria of biochemical reaction systems

Matthew D Johnston, Stefan M \"u ller, and Casian Pantea. A deficiency-based approach to parametrizing positive equilibria of biochemical reaction systems. Bulletin of Mathematical Biology , 81:1143--1172, 2019

2019
[12]

A. M. M. Leal. Reaktoro: An open-source unified framework for modeling chemically reactive systems, 2015. URL: https://reaktoro.org

2015
[13]

Energy-based modeling in BioNetGen

John AP Sekar, Justin S Hogg, and James R Faeder. Energy-based modeling in BioNetGen . In 2016 IEEE international conference on bioinformatics and biomedicine ( BIBM ) , pages 1460--1467. IEEE , 2016

2016
[14]

Concentrat.io , 2025

Elie Soloveichik, Leo Orshansky, Cameron Chalk, and Boya Wang. Concentrat.io , 2025. Accessed 25 June 2025. URL: https://concentrat.io/

2025
[15]

Leakless DNA strand displacement systems

Chris Thachuk, Erik Winfree, and David Soloveichik. Leakless DNA strand displacement systems. In Andrew Phillips and Peng Yin, editors, DNA Computing and Molecular Programming , pages 133--153, Cham, 2015. Springer International Publishing

2015
[16]

Molecular computation at equilibrium via programmable entropy

Boya Wang, Cameron Chalk, David Doty, and David Soloveichik. Molecular computation at equilibrium via programmable entropy. bioRxiv , 2025. https://doi.org/10.1101/2024.09.13.612990 doi:10.1101/2024.09.13.612990

2025 doi
[17]

Ellington, Erik Winfree, and David Soloveichik

Boya Wang, Chris Thachuk, Andrew D. Ellington, Erik Winfree, and David Soloveichik. Effective design principles for leakless strand displacement systems. Proc. Natl. Acad. Sci. USA , 115(52):E12182–E12191, 2018. https://doi.org/10.1073/pnas.1806859115 doi:10.1073/pnas.1806859115

2018 doi
[18]

Speed and correctness guarantees for programmable enthalpy-neutral DNA reactions

Boya Wang, Chris Thachuk, and David Soloveichik. Speed and correctness guarantees for programmable enthalpy-neutral DNA reactions. ACS Synthetic Biology , 12(4):993--1006, 2023

2023
[19]

NUPACK: analysis and design of nucleic acid systems

Joseph N Zadeh, Conrad D Steenberg, Justin S Bois, Brian R Wolfe, Marshall B Pierce, Asif R Khan, Robert M Dirks, and Niles A Pierce. NUPACK: analysis and design of nucleic acid systems. Journal of computational chemistry , 32(1):170--173, 2011

2011

Pith tools

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

[1] [1]

Programming substrate-independent kinetic barriers with thermodynamic binding networks

Keenan Breik, Cameron Chalk, David Doty, David Haley, and David Soloveichik. Programming substrate-independent kinetic barriers with thermodynamic binding networks. IEEE/ACM Trans. Comput. Biol. Bioinformatics , 18(1):283--295, 2021. https://doi.org/10.1109/TCBB.2019.2959310 doi:10.1109/TCBB.2019.2959310

work page arXiv 2021

[2] [2]

Computing properties of stable configurations of thermodynamic binding networks

Keenan Breik, Chris Thachuk, Marijn Heule, and David Soloveichik. Computing properties of stable configurations of thermodynamic binding networks. Theor. Comput. Sci. , 785:17--29, 2019. https://doi.org/10.1016/j.tcs.2018.10.027 doi:10.1016/j.tcs.2018.10.027

work page doi:10.1016/j.tcs.2018.10.027 2019

[3] [3]

Bruns, B

W. Bruns, B. Ichim, C. S\"oger, and U. von der Ohe. Normaliz. A lgorithms for rational cones and affine monoids. Available at https://www.normaliz.uni-osnabrueck.de

work page

[4] [4]

How far is complex balancing from detailed balancing? Bulletin of mathematical biology , 73(4):811--828, 2011

Alicia Dickenstein and Mercedes P \'e rez Mill \'a n. How far is complex balancing from detailed balancing? Bulletin of mathematical biology , 73(4):811--828, 2011

work page 2011

[5] [5]

Thermodynamic analysis of interacting nucleic acid strands

Robert M Dirks, Justin S Bois, Joseph M Schaeffer, Erik Winfree, and Niles A Pierce. Thermodynamic analysis of interacting nucleic acid strands. SIAM review , 49(1):65--88, 2007

work page 2007

[6] [6]

Rogers, David Soloveichik, Chris Thachuk, and Damien Woods

David Doty, Trent A. Rogers, David Soloveichik, Chris Thachuk, and Damien Woods. Thermodynamic binding networks. In DNA Computing and Molecular Programming, 23rd International Conference, DNA 23 , pages 249--266, 2017. https://doi.org/10.1007/978-3-319-66799-7_16 doi:10.1007/978-3-319-66799-7_16

work page doi:10.1007/978-3-319-66799-7_16 2017

[7] [7]

Necessary and sufficient conditions for detailed balancing in mass action systems of arbitrary complexity

Martin Feinberg. Necessary and sufficient conditions for detailed balancing in mass action systems of arbitrary complexity. Chemical Engineering Science , 44(9):1819--1827, 1989

work page 1989

[8] [8]

Foundations of chemical reaction network theory

Martin Feinberg. Foundations of chemical reaction network theory . Springer, 2019

work page 2019