{"id":"4d95bf4c-54e9-43a5-8a98-18825aebb7a4","arxiv_id":"2502.04754","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Every bidirectional mass-action chemical system is the reduction of a closed, detailed-balanced system obtained by adding substances and reactions.","lead":"This paper proves that any reversible chemical reaction network, even with rates that violate detailed balance, can be embedded as an open slice of a larger closed network that obeys detailed balance. This gives a rigorous mathematical bridge between standard non-equilibrium biochemical models and thermodynamically consistent closed systems.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: Theorem 6.3 is internally consistent; the finite-α caveat is a modeling limitation, not a flaw in the completion construction.","rationale":"The reader's verdict is CONDITIONAL, driven mainly by the finite-α interpretation and by minor typos. My stress-test focused on the actual load-bearing steps of Theorem 6.3 and found them sound. The finite-α concern is a legitimate caveat for the physical interpretation of 'freezing' concentrations, but it is not a flaw in the algebraic reduction/completion theorem, which is the central claim. I therefore see no reason to alter the reader's conditional verdict: the typos and imprecisions still warrant a conditional recommendation, but no deeper correctness risk surfaced. The independent construction check proposed would further confirm the non-trivial Steps 2 and 3 of Proposition 6.2 on a concrete non-conservative network.","tokens_in":36075,"tokens_out":17852,"duration_ms":205366,"concrete_test":"Run an independent check of Proposition 6.2 on a non-conservative bidirectional network with M = {0}, e.g., reactions (1,-1,0), (0,1,-1), (1,-1,1) plus their reverses. Verify that the Step 2 matrix has a conservation law positive on every original substance and that the Step 3 construction reduces the cycle dimension to zero while preserving one-to-one projections and the absence of sources/sinks. If both hold, the central theorem needs no revision beyond the typo fixes.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the algebraic completion theorem: every bidirectional mass-action kinetic system is the reduction of a closed, detailed-balanced system. I checked the three steps of Proposition 6.2, which underpin Theorem 6.3. Step 1 removes sources/sinks by adding two substances per bad reaction. Step 2 mirrors the non-conserved part B: for each reaction it appends the coefficients (π_B R, -π_B R), and any vector (0, y, y) with y>0 is then a conservation law positive on every old B substance, so conservativeness is restored. Step 3 removes cycles by appending two row functionals W1, W2 per cycle; a direct computation confirms the claimed new conservation law (m_c, (1/4)m_e(i), (1/4)m_e(i)) annihilates every augmented reaction column, since W1 + W2 = ζ and m_c differs from m_e only by halving the i-th component. The theorem's reduction identity K[n_U] = K at n_U = 1 follows from the one-to-one projection property. The reader's finite-α concern is real for physical modeling of open systems, but the paper explicitly defines reduction via the α → ∞ limit and does not claim finite-rate equivalence. It is a limitation of the modeling interpretation, not a gap in the mathematical claim. The proof has only the noted typo (Cc ≠ {0} should be Cc = {0}) and minor wording issues, which do not affect the theorem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a reduction formalism for mass-action chemical reaction networks in which a subset of concentrations is frozen, and asks when such reduced systems inherit the detailed balance property from the original system. It proves that freezing at equilibrium values preserves detailed balance (Proposition 5.1), characterizes robust preservation in terms of cycle data (Theorem 5.4 and Propositions 5.6, 5.8, 5.10), and then proves the main completion theorem (Theorem 6.3): every bidirectional mass-action kinetic system is the reduction of a closed, conservative, detailed-balanced system with additional substances frozen at concentration 1. The paper also treats constrained completions and defines kinetic systems with fluxes.","tokens_in":36326,"tokens_out":16531,"duration_ms":180525,"significance":"If the main theorem is correct, the paper establishes a strong and useful embedding statement: arbitrary reversible open networks, even those that violate detailed balance, can be realized as fast-exchange reductions of thermodynamically consistent closed networks. This gives a rigorous framework for quantifying how far a biochemical model is from equilibrium. The proofs are self-contained, the constructions are explicit, and no fitted parameters or back-computed constants appear; the finite-α caveat in the reduction formalism is clearly stated as a modeling limitation rather than a mathematical claim. The central theorem is internally consistent, but one algebraic step in the proof of Proposition 6.2 needs correction as written.","major_comments":[{"comment":"The displayed conservation law in Step 3 is incorrect as written. For an old reaction column R, with W1 + W2 = ζ = e_i^T Re and m_c = m_e - (1/2)m_e(i)e_i, the contribution of the old substances is -1/2 m_e(i) ζ(R), while the two new substances contribute a ζ(R) if each has coefficient a. To cancel, one needs a = 1/2 m_e(i), not 1/4 m_e(i). Thus the vector should be (m_c, 1/2 m_e(i), 1/2 m_e(i)) rather than (m_c, 1/4 m_e(i), 1/4 m_e(i)). Since conservativeness of the completed network is load-bearing for Theorem 6.3, this correction is necessary; after the correction the argument goes through, choosing m_e positive as guaranteed by conservativeness.","section":"Section 6.1, Proposition 6.2, Step 3"},{"comment":"The proof asserts that the row ζ used in Step 3 of Proposition 6.2 has ζ(i) = 0 for every constrained reaction R_i, but Step 3 only selects a row with two nonzero entries on the cycle being eliminated; it does not control the entries on reactions outside that cycle. Since Ra-admissibility requires W1 and W2 to vanish on constrained reactions, an additional linear-algebra argument is needed to show that a suitable row-space vector (or a different splitting) can be chosen with this vanishing property. As written, Proposition 6.7 is not fully proved.","section":"Section 6.2, Proposition 6.7"}],"minor_comments":[{"comment":"The sentence \"hence Cc ≠ {0}\" contradicts Proposition 6.2 and should read \"hence Cc = {0}\".","section":"Theorem 6.3, proof"},{"comment":"In the final displayed computation, the phrase \"since c ∈ CV\" appears in a sum over the full reaction set R; the cycle membership that gives ∑ c(j)R_j(s) = 0 is the lifted cycle c ∈ C, so the notation should be corrected to avoid confusion.","section":"Theorem 5.4, proof"},{"comment":"The notation R* = [0,∞) and R+ = (0,∞) is nonstandard and should be flagged more prominently, since many readers will expect the opposite convention.","section":"Section 1.1"}],"recommendation":"major_revision","confidential_remarks":"The main completion theorem is plausible and the construction is essentially correct, but the proof of Proposition 6.2 contains a concrete algebraic error in the conservativeness argument that must be fixed before the paper can be accepted. The gap in Proposition 6.7 is secondary but should also be addressed. The finite-α concern raised in the stress-test note is a modeling limitation, not a defect in the mathematical claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline is simple: this paper proves a new universal embedding theorem for mass-action networks, and the proof is sound. Any bidirectional mass-action kinetic system can be realized as a reduction of a closed, detailed-balanced system obtained by adding substances and freezing them at concentration 1. That is a real result, and I don't see a gap in the construction.\n\nWhat is new is Theorem 6.3 together with the robust inheritance characterization in Section 5.2. The theorem says that every bidirectional kinetic system admits a closed completion; the earlier literature has related ideas, but not this universal construction. The paper also gives necessary and sufficient topological conditions for the reduced system to inherit detailed balance in a stable way: one-to-one reduced reactions, no zero-V reductions, and p(C)=C_V. That is a clean package, and the proofs are largely self-contained. The authors use Feinberg's book and the Wegscheider criterion as background, not as a substitute for the main argument. No data fitting, no back-computed constants; the circularity burden is minimal.\n\nThe soft spots are minor. The proof of Theorem 6.3 has the typo Cc != {0} where Proposition 6.2 proved Cc = {0}; that should be corrected. A few other small typos and wording slips appear in Section 5.2 and in the proof of Theorem 5.4, but none affect the substance. The stress-test note about the cycle-removal step in Proposition 6.2 overstating the dimension drop does not hold up on reading: when W2 = ζ - W1, on the cycle space one has W1 + W2 = ζ = 0, so the two added rows have rank one restricted to the cycles; the dimension drops by exactly one, as the paper says. The finite-α caveat is real for physical modeling, but the paper explicitly defines reduction through the α→∞ limit and does not claim finite-rate equivalence. That is an interpretational limitation, not a flaw in the completion construction.\n\nFor whom: this is for people working in chemical reaction network theory and mathematical biology, especially anyone who wants to justify out-of-equilibrium models as reductions of closed detailed-balanced systems. The flux-based measure in Section 7 is a nice extra.\n\nRecommendation: send it to peer review. It deserves a serious referee. The referee should ask for typo fixes, a brief comment on the finite-α regime, and perhaps a short discussion connecting the topological conditions to known circuit/cycle conditions. The core result is sound and new.","headline":"A genuinely new embedding theorem for mass-action networks with a sound proof; the main caveat is interpretational, not mathematical.","tokens_in":36860,"tokens_out":3404,"would_cite":true,"duration_ms":34348,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["34A34","37N25","80A30","92C42"],"pacs":[],"model":"deepseek-v4-flash","headline":"Every reversible chemical network can be embedded in a closed, detailed-balanced system built by adding new substances and reactions.","keywords":["chemical reaction networks","detailed balance","mass action kinetics","non-equilibrium systems","kinetic systems with fluxes","closed completion","reduction of kinetic systems","Wegscheider criterion"],"falsifier":"Take the four-cycle $(1) \\leftrightarrows (2) \\leftrightarrows (3) \\leftrightarrows (4) \\leftrightarrows (1)$ with rate constants whose product around the cycle is not $1$, declare all four reactions constrained, and attempt to build a closed admissible completion: Proposition 6.8 says none exists, and verifying that no such completion can be written down would test the theorem's boundary. A complementary check is to simulate the completion constructed in Theorem 6.3 with added concentrations held at $1$ and confirm that the original ODEs are recovered exactly.","tokens_in":35826,"feed_emoji":"⚖️","tokens_out":7119,"duration_ms":69890,"temperature":0.7,"pith_summary":"Many biochemical models violate detailed balance, the condition that each reaction is balanced by its reverse at equilibrium. This paper argues that such non-equilibrium systems are still compatible with a thermodynamically consistent starting point: any reversible mass-action network can be obtained by freezing some concentrations in a larger closed network that satisfies detailed balance. The central result, Theorem 6.3, constructs a closed completion for every bidirectional kinetic system, adding substances and reactions so that the completion has no cycles and hence satisfies the circuit condition automatically. The reduction of that completion, with the added substances held at concentration 1, reproduces the original kinetics exactly. This matters because it makes the failure of detailed balance an effective, rather than fundamental, property of open biochemical systems.","feed_headline":"Detailed balance can always be restored by adding substances","feed_subtitle":"A new theorem embeds any reversible mass-action system in a closed, thermodynamically consistent network.","key_machinery":"The load-bearing object is the reduction map $K[n_U]$, which extracts an effective rate for each projected reaction by summing original mass-action rates evaluated with the frozen concentrations $n_U$ held fixed, together with the complementary notion of a closed completion of the original network. Detailed balance is checked through the circuit condition: along every cycle the product of forward-to-reverse rate ratios must equal $1$. The completion construction makes all cycles disappear, so the circuit condition holds automatically, while setting the frozen concentrations of added substances to $1$ makes the reduction return the original rates. Proposition 5.1 supplies the companion statement that freezing concentrations at equilibrium values $e^{-E}$ preserves detailed balance in the reduced system.","core_discovery":"On its own terms, the paper establishes that every bidirectional mass-action kinetic system $\\Omega, R, K$ is admissible: there exists a closed kinetic system $\\Omega_c, R_c, K_c$, detailed-balanced, conservative, and with no sources or sinks, whose reduction by freezing the added substances at concentration $1$ is exactly the original system. The proof is constructive: sources and sinks are paired with dummy substances; substances outside all conservation laws are given mirror copies to make the network conservative; and cycles are removed one by one by adding reactions and two new substances per cycle. A network without cycles satisfies the circuit condition, also called the Wegscheider criterion, for any choice of rates, so the completed system is automatically detailed-balanced. The same analysis shows that if certain reactions are constrained, a closed completion exists provided the constrained reactions do not lie in cycles; if an entire cycle whose rates violate the circuit condition is constrained, no closed admissible completion exists, as stated in Proposition 6.8.","pith_inferences":["If finite exchange rates are physically relevant, a natural next step is to study the completed system for large but finite $\\alpha$; the exact embedding would then be the singular limit $\\alpha \\to \\infty$, and finite-$\\alpha$ corrections could be quantified.","The constructive proof suggests a quantitative measure of how far a network is from equilibrium: the minimal number of added substances and reactions needed for a closed completion, or the minimal number of reactions per cycle that must be modified. The paper does not pursue this measure.","Because the paper indicates that robust adaptation is impossible for closed chemical systems, a testable consequence is that any robustly adapting biochemical network must have a completion carrying nonzero external fluxes; this is an inference, not a claim made in the paper.","The completion with frozen concentrations fixed at $1$ effectively models the environment as infinite reservoirs, so an alternative reading is that every open mass-action network is the shadow of a conservative closed network, which could help in thermodynamic costing of biochemical functions."],"forward_implications":["Any open bidirectional mass-action model, however far from detailed balance, has a closed, thermodynamically consistent embedding; lack of detailed balance can therefore be treated as an effective phenomenon caused by frozen concentrations.","If a cycle with nonzero energy imbalance contains only constrained reactions, the network has no closed completion, so such networks cannot be embedded without altering the constrained reactions.","Reduced systems whose frozen concentrations sit at equilibrium values inherit detailed balance and, per stoichiometric compatibility class, a unique globally stable steady state, as follows from Propositions 5.1 and 3.8.","For kinetic systems with fluxes at equilibrium frozen values, the free energy obeys $\\partial_t F = -D_R + J_{\\mathrm{ext}}$ and solutions relax to $e^{-E}$, with finite integrated external fluxes, as stated in Theorem 7.1.","The cycle-free completion can be chosen so that every original reaction is a one-to-one reduction and every added reaction has nonzero projection onto the original substances, making the reduction faithful."],"supporting_citations":[{"why":"Feinberg's necessary-and-sufficient conditions for detailed balancing in mass action systems; Lemma 3.2 uses this equivalence to the circuit condition.","marker":"[7]"},{"why":"Foundations of chemical reaction network theory supplies the proof that detailed balance implies a unique globally stable steady state in each stoichiometric class, used in Proposition 3.8.","marker":"[8]"},{"why":"Wegscheider's criterion states the product condition on rate ratios along cycles, used throughout as the circuit condition for detailed balance.","marker":"[24]"},{"why":"Krein-Milman theorem is used in Lemma 2.9 to show that extreme rays of the cone of conservation laws form a basis, a step needed for the conservative-completion construction.","marker":"[21]"}],"fun_headline_variants":["Detailed balance always restorable by adding substances","Any reversible network embeds in a detailed-balanced one","Add substances to make any reaction network detailed-balanced","New proof: closed completions exist for all reversible systems","Constructive extension to thermodynamically consistent networks"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The theorem assumes the exchange with the environment is infinitely fast, so frozen concentrations are exactly constant; if that exchange is finite, the reduced system is only an approximation and the exact embedding statement no longer describes the open dynamics. It also assumes all reactions are reversible and obey mass-action kinetics.","fun_headline_variants_meta":{"raw":{"variants":["Detailed balance always restorable by adding substances","Any reversible network embeds in a detailed-balanced one","Add substances to make any reaction network detailed-balanced","New proof: closed completions exist for all reversible systems","Constructive extension to thermodynamically consistent networks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000208,"raw_usage":{"total_tokens":1348,"prompt_tokens":832,"completion_tokens":516,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":448,"completion_tokens_details":{"reasoning_tokens":442}},"tokens_in":448,"tokens_out":516,"duration_ms":5441,"temperature":1.0,"reasoning_tokens":442,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T21:36:55.500348+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the four-cycle $(1) \\leftrightarrows (2) \\leftrightarrows (3) \\leftrightarrows (4) \\leftrightarrows (1)$ with rate constants whose product around the cycle is not $1$, declare all four reactions constrained, and attempt to build a closed admissible completion: Proposition 6.8 says none exists, and verifying that no such completion can be written down would test the theorem's boundary. A complementary check is to simulate the completion constructed in Theorem 6.3 with added concentrations held at $1$ and confirm that the original ODEs are recovered exactly.","supporting_citations":[{"cited_title":"Feinberg, Necessary and sufficient conditions for detailed balancing in mass action systems of arbitrary complexity, Chemical Engineering Science, 44 (1989), pp","cited_arxiv_id":null,"evidence_quote":"Feinberg's necessary-and-sufficient conditions for detailed balancing in mass action systems; Lemma 3.2 uses this equivalence to the circuit condition."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Foundations of chemical reaction network theory supplies the proof that detailed balance implies a unique globally stable steady state in each stoichiometric class, used in Proposition 3.8."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Wegscheider's criterion states the product condition on rate ratios along cycles, used throughout as the circuit condition for detailed balance."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Krein-Milman theorem is used in Lemma 2.9 to show that extreme rays of the cone of conservation laws form a basis, a step needed for the conservative-completion construction."}],"review_version":1}