{"id":"72bebac3-b8f6-4cd8-9f9c-ae41b0f06e8d","arxiv_id":"1908.04497","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"In deficiency-one reaction networks with reactant-determined power-law kinetics, two nonterminal complexes whose kinetic order vectors differ in exactly one species force that species to have the same concentration in every positive steady state.","lead":"This paper extends a classic theorem about absolute concentration robustness in chemical reaction networks from standard mass-action kinetics to a broader power-law kinetic class. The extension applies to models of biology and geochemistry where reactions occur in crowded or fractal environments, and is demonstrated on a power-law version of a pre-industrial carbon cycle model.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified.","rationale":"The reader accepted the paper with high confidence, and my independent check agrees that the main theorem is sound. The reader's identified weakest assumption concerns the carbon-cycle illustration: the ACR result applies to the power-law approximation, not necessarily to the original nonlinear model. That is a real limitation of the application, but it is disclosed in the paper and does not affect the central theorem. Since I found no load-bearing concern with Theorem 1 itself, the appropriate verdict is unchanged.","tokens_in":10679,"tokens_out":9137,"duration_ms":101825,"concrete_test":"Generate a few thousand random deficiency-one PL-RDK networks satisfying the hypotheses of Theorem 1; for each network, solve the steady-state equations from many initial conditions and verify that the designated species concentration is identical across all positive equilibria. If a counterexample appears, Proposition 2 has a hidden assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is Theorem 1, and I checked the proof of Proposition 2 (Section 3). The key step is that the dimension bound dim(Ker YA_kappa) <= 1 + t combines with 1_C not in Ker A_kappa to force equality, so every vector in Ker YA_kappa is a linear combination of 1_C and terminal-supported kernel vectors. The comparison of nonterminal coefficients in Eq. (3.15) then gives Eq. (3.16), and the reduction to T-columns is valid because nonterminal complexes are necessarily reactant complexes. I found no gap in the extension; the theorem holds for PL-RDK exactly as stated. The carbon-cycle application is explicitly an analysis of the power-law approximation (Eq. 4.1), not the original Anderies ODEs (A.1), and the paper discloses this; it is a limitation of the illustration, not of the central theorem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the Shinar-Feinberg absolute concentration robustness (ACR) theorem from mass-action kinetics to power-law kinetics with reactant-determined interactions (PL-RDK). Theorem 1 states that for a deficiency-one CRN and a PL-RDK system admitting a positive equilibrium, if two nonterminal complexes have kinetic order vectors that differ only in species Xi, then the system has ACR in Xi. The proof is based on Proposition 2, which generalizes the original Shinar-Feinberg argument by replacing stoichiometric complexes with kinetic-order vectors encoded in the T-matrix. The paper applies the theorem to a GMA power-law approximation of the Anderies et al. pre-industrial carbon-cycle model, showing ACR in atmospheric carbon A2 when the human terrestrial carbon off-take coefficient α vanishes.","tokens_in":10790,"tokens_out":18998,"duration_ms":166745,"significance":"The main result is a genuine and useful generalization of a celebrated theorem. The proof is complete and transparent, relying only on standard CRNT results (the Structure Theorem of the Laplacian Kernel and the deficiency-bound Corollary 1). The carbon-cycle illustration demonstrates a plausible application, though the authors correctly limit their claim to the power-law approximation rather than the original nonlinear model. This is a solid contribution to chemical reaction network theory, of interest to researchers in CRNT and biochemical systems theory.","major_comments":[],"minor_comments":[{"comment":"The sentence \"Particulary, for the pre-industrial scenario where there are anthropogenic causes that reduce the capacity of terrestrial carbon pool to store carbon, the power law approximation leads to an ACR-possessing PL-RDK system\" is inconsistent with the derivation in Section 4 and Appendix A, where ACR is obtained when the human terrestrial carbon off-take coefficient α vanishes. Please rephrase to avoid implying that ACR arises in the presence of anthropogenic off-take.","section":"Section 1, last paragraph"},{"comment":"The sentence \"Observe that each vector bi, i = 0, 1,...,t , has its support entirely on terminal complexes\" is incorrect for i=0, since b_0 = 1_C has support on all complexes. The subsequent argument only requires that b_i for i=1,...,t have terminal support; please correct the indexing.","section":"Section 3, proof of Proposition 2, after Eq. (3.15)"},{"comment":"The phrase \"Therefore, c* and c** are positive equilibria ... if and only if Equations (3.11) and (3.12) hold\" is slightly misleading, since (3.11) holds for c* by construction of κ. Consider rephrasing to state that (3.11) is equivalent to c* being an equilibrium and (3.12) is equivalent to c** being an equilibrium.","section":"Section 3, proof of Proposition 2, after Eq. (3.12)"},{"comment":"The phrase \"which accounts for the which accounts for human activities\" is a typo; please remove the duplicated words.","section":"Section 5, item 2"},{"comment":"There are minor typographical errors: \"Particulary\" should be \"Particularly\" in the abstract and Section 1, and \"a phenomena\" should be \"a phenomenon.\"","section":"Abstract and Section 1"},{"comment":"The proof that \"T = ŷ∘Y_res is also an isomorphism\" is compressed; since Y_res may not be surjective onto R^m, it is more precise to say T is injective (hence the system is PL-RLK). Please elaborate or rephrase to avoid confusion.","section":"Section 3, Proposition 3"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is scientifically sound. The central theorem is correct and the proof is rigorous. The only substantive concern is the framing of the carbon-cycle scenario in the introduction, which appears to contradict the technical result (ACR arises when α=0, i.e., without human off-take). The indexing error in the proof of Proposition 2 and the various typos are easily fixed. These issues do not affect the theorem and are appropriate for a minor revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper really does extend Shinar-Feinberg to PL-RDK, and the proof checks out. The carbon-cycle application is a nice exercise on a power-law approximation, not a result about the original nonlinear model, and the paper says so.\n\nThe new thing is Theorem 1: for deficiency-one PL-RDK systems, if two nonterminal complexes have kinetic order vectors differing only in species Xi, all positive steady states agree on Xi. That is a clean generalization; mass-action is the special case. The proof largely retraces Shinar-Feinberg's argument, replacing stoichiometric complexes with kinetic-order columns T. The key step—dim(Ker YA_k) <= 1+t plus 1_C not in Ker A_k forcing the nonterminal coefficients to match—is correct. I checked Eqs. (3.15)-(3.17): the comparison of nonterminal coefficients uses the fact that nonterminal complexes are reactant complexes, which is valid. So the central claim is sound.\n\nThe carbon-cycle application works through a GMA approximation (4.1) of the Anderies model, using the authors' own total-CRN representation from [13]. The condition p1=p2 follows from setting the human off-take alpha=0, so ACR in atmospheric carbon holds for the approximated PL-RDK system. The paper does not show that the original ODEs (A.1), with logistic and exponential terms, also have ACR; it explicitly frames the result as about the approximation. That is a limitation of the illustration, not of the theorem. If the authors wanted a stronger claim they would need to show the approximation preserves ACR, which they don't. Also the steady-state existence is assumed; they don't discuss parameter ranges, but the explicit equilibrium set in Section 4 helps.\n\nMinor soft spots: Proposition 3 is a fairly direct corollary and doesn't add much. The references to [13] are appropriate since that is their previous work. I didn't see self-citation inflation; the cited results are structural and used as tools.\n\nBottom line: the theorem is a genuine, modest extension with a complete proof, and the application is honest about its scope. This deserves a serious referee. I would accept it for a mathematical biology or CRNT journal after minor revision, mainly for clarity around the approximation caveat.","headline":"A sound, honest extension of Shinar–Feinberg ACR to PL-RDK systems; the proof is correct, and the carbon-cycle application is clearly scoped to a power-law approximation.","tokens_in":11382,"tokens_out":1475,"would_cite":true,"duration_ms":15523,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37N25","80A30"],"pacs":[],"model":"deepseek-v4-flash","headline":"A structural robustness theorem now covers power-law reaction kinetics","keywords":["absolute concentration robustness","chemical reaction network","power law kinetics","reactant-determined interactions","deficiency-one","carbon cycle model","Shinar-Feinberg theorem"],"falsifier":"The theorem would be falsified by a deficiency-one PL-RDK system satisfying its hypotheses but possessing two positive steady states with different $X_i$ concentrations. In the carbon setting, one concrete check is to solve the original ODE system (A.1) with $\\alpha=0$ and verify whether every positive steady state satisfies $A_2=(k_2/k_1)^{1/(q_1-q_2)}$; a positive steady state with a different $A_2$ would show the approximation does not transfer ACR.","tokens_in":10428,"feed_emoji":"⚗️","tokens_out":8600,"duration_ms":80685,"temperature":0.7,"pith_summary":"This paper proves that absolute concentration robustness (ACR) — a species taking the same value at every positive steady state, independent of initial conditions — can be certified from network structure for a broad class of power-law kinetic systems, not only the mass-action systems covered by the original Shinar-Feinberg theorem. The relevant class is power-law kinetics with reactant-determined interactions (PL-RDK), where reactions that share a reactant complex share the same kinetic order vector. The main theorem states that in a deficiency-one reaction network admitting a positive equilibrium, if two nonterminal complexes have kinetic order vectors differing only in species $X_i$, then the system has ACR in $X_i$. This is a genuine extension because power-law kinetics models reactions in crowded intracellular or fractal-like media, where kinetic orders need not be the integer stoichiometric coefficients of mass action. The paper illustrates the result on a power-law approximation of a pre-industrial global carbon-cycle model, where the atmospheric carbon pool is absolutely robust when the human terrestrial carbon off-take term vanishes.","feed_headline":"A robustness theorem now covers power-law kinetics","feed_subtitle":"Non-integer kinetic orders still permit provable concentration robustness, with a carbon-cycle illustration.","key_machinery":"The load-bearing object is the T-matrix, whose columns are the kinetic order vectors of the reactant complexes; PL-RDK is exactly the condition that these columns are well defined because reactions with the same reactant complex have identical kinetic orders. The proof leans on the Structure Theorem of the Laplacian Kernel, which supplies a basis of the Laplacian kernel supported on terminal strong linkage classes, and on the deficiency-one bound $\\dim(\\ker YA_\\kappa)\\le 1+t$. These ingredients yield Proposition 2's log-linear relation between any two positive equilibria for nonterminal complexes. Theorem 1 follows because nonterminal complexes are necessarily reactant complexes, so the relevant columns of $T$ exist, and a single-coordinate difference in those columns forces equality of the $\\log$-coordinates.","core_discovery":"Theorem 1 is the paper's central claim: for a deficiency-one chemical reaction network $(S,\\mathcal{C},\\mathcal{R})$ equipped with PL-RDK kinetics and a positive equilibrium, if two nonterminal complexes $y,y'$ have kinetic order vectors that differ only in species $X_i$, then every positive steady state assigns the same concentration to $X_i$. The proof derives, for any two positive equilibria $c^*$ and $c^{**}$, the log-linear equation $(T_{\\cdot,y}-T_{\\cdot,y'})\\cdot \\log(c^{**}/c^*)=0$, where $T$ is the reactant-complex kinetic order matrix. When the two columns differ only in coordinate $i$, this equation leaves only $\\log c_i^{**}=\\log c_i^*$, so the species is absolutely robust. The paper also exhibits the concrete equilibrium set of the approximated carbon-cycle scenario, in which $A_2=(k_2/k_1)^{1/(q_1-q_2)}$ is fixed while $A_1$ adjusts to conserve total carbon.","pith_inferences":["The theorem itself is parameter-free, but the carbon conclusion is not: it relies on the power-law (GMA) approximation of the Anderies model and on setting the human terrestrial off-take coefficient to zero. A testable extension is to check whether the original logistic-exponential model also has a unique positive $A_2$ at equilibrium under $\\alpha=0$.","Proposition 2's log-linear relation holds for any pair of nonterminal complexes, not just pairs differing in one species; this suggests a possible general robustness criterion or quantitative constraints on ratios of species concentrations, which the paper does not develop.","Because the condition is a comparison of two columns of $T$, robustness certification could be automated: scan a PL-RDK system's T-matrix for single-coordinate differences among nonterminal reactant complexes before doing any dynamics, which would make ACR screening practical for large networks."],"forward_implications":["Any PL-RDK system that meets the structural hypotheses inherits the Shinar-Feinberg guarantee: the distinguished species has one fixed concentration across all positive steady states, regardless of rate constants or initial conditions.","The carbon-cycle application identifies a pre-industrial scenario in which the approximated atmospheric carbon pool $A_2$ is absolutely robust, with the explicit fixed value $(k_2/k_1)^{1/(q_1-q_2)}$.","Because mass-action kinetics is a special case of PL-RDK, the theorem contains the original Shinar-Feinberg ACR theorem rather than merely mimicking it.","The hypothesis is readable off the network and kinetic-order matrix, so practitioners can certify ACR in models with fractional kinetic orders without solving the steady-state equations."],"supporting_citations":[{"why":"supplies the original Shinar-Feinberg ACR theorem and the proof template that Proposition 2 adapts.","marker":"[26]"},{"why":"provides the deficiency index, the Structure Theorem of the Laplacian Kernel, and the dimension bound used in the proof.","marker":"[11]"},{"why":"gives the total CRN representation of GMA systems used to translate the carbon ODEs into a PL-RDK system.","marker":"[5]"},{"why":"contains the earlier derivation of the deficiency-one CRN and the power-law approximation of the carbon-cycle model adopted here.","marker":"[13]"},{"why":"is the Anderies et al. carbon-cycle model whose pre-industrial scenario supplies the application.","marker":"[1]"},{"why":"provides the T-matrix and positive-equilibrium framework for PL-RDK systems used in the statement and proof.","marker":"[27]"}],"fun_headline_variants":["Power-law kinetics now have a robustness theorem","Robustness theorem broadened to power-law reaction networks","Carbon cycle illustrates new robustness proof for kinetics","Concentration robustness proven beyond mass-action systems","PL-RDK systems gain absolute concentration robustness proof"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The theorem assumes the network has deficiency one and admits a positive equilibrium; for the carbon illustration, the added load-bearing assumptions are that the power-law approximation faithfully represents the original nonlinear dynamics and that the human terrestrial off-take coefficient is exactly zero.","fun_headline_variants_meta":{"raw":{"variants":["Power-law kinetics now have a robustness theorem","Robustness theorem broadened to power-law reaction networks","Carbon cycle illustrates new robustness proof for kinetics","Concentration robustness proven beyond mass-action systems","PL-RDK systems gain absolute concentration robustness proof"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000199,"raw_usage":{"total_tokens":1360,"prompt_tokens":923,"completion_tokens":437,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":539,"completion_tokens_details":{"reasoning_tokens":367}},"tokens_in":539,"tokens_out":437,"duration_ms":4597,"temperature":1.0,"reasoning_tokens":367,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:42:15.345765+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The theorem would be falsified by a deficiency-one PL-RDK system satisfying its hypotheses but possessing two positive steady states with different $X_i$ concentrations. In the carbon setting, one concrete check is to solve the original ODE system (A.1) with $\\alpha=0$ and verify whether every positive steady state satisfies $A_2=(k_2/k_1)^{1/(q_1-q_2)}$; a positive steady state with a different $A_2$ would show the approximation does not transfer ACR.","supporting_citations":[{"cited_title":"Science 327(5971), 1389–1391 (2010)","cited_arxiv_id":null,"evidence_quote":"supplies the original Shinar-Feinberg ACR theorem and the proof template that Proposition 2 adapts."},{"cited_title":"Notes of lectures given at the mathe- matics research center of the University of Wisconsin (1979)","cited_arxiv_id":null,"evidence_quote":"provides the deficiency index, the Structure Theorem of the Laplacian Kernel, and the dimension bound used in the proof."},{"cited_title":"Mathematical Biosciences 269, 135–52 (2015)","cited_arxiv_id":null,"evidence_quote":"gives the total CRN representation of GMA systems used to translate the carbon ODEs into a PL-RDK system."},{"cited_title":"Journal of Mathematical Chemistry 56(10), 2929–2962 (2018)","cited_arxiv_id":null,"evidence_quote":"contains the earlier derivation of the deficiency-one CRN and the power-law approximation of the carbon-cycle model adopted here."},{"cited_title":"Environmental Research Letters 8(4), 044–048 (2013)","cited_arxiv_id":null,"evidence_quote":"is the Anderies et al. carbon-cycle model whose pre-industrial scenario supplies the application."},{"cited_title":"Journal of Mathematical Chemistry 56(2), 358–394 (2018) 12","cited_arxiv_id":null,"evidence_quote":"provides the T-matrix and positive-equilibrium framework for PL-RDK systems used in the statement and proof."}],"review_version":1}