{"id":"99c6bcfe-7ab1-4e1d-ab41-4fc6cb9422e3","arxiv_id":"2508.20004","paper_version":2,"verdict":"REJECT","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"Claims that occupancy-induced curvature on a Boolean state manifold, defined as a zero-sum Laplacian, determines least-action proteoform transitions and predicts cysteine oxidation order in GAPDH.","lead":"A theory paper proposes that a protein's possible chemical states form a discrete geometric landscape whose curvature, created by which states are occupied, steers which transition happens next. The authors test it on cysteine oxidation states of GAPDH, claiming the geometry predicts which cysteine reacts first.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The simulation evidence for 'modal geometry governs' is circular: the Figure 2A regression predicts the action integral from curvature and anisotropy terms that are additively included in that same action integral.","rationale":"I read the paper as attempting to establish a universal law: conserved curvature (Axiom 4) dictates least-action transitions. For that to be credible, there must be some independent demonstration that curvature/anisotropy actually shape dynamics. The paper's only such demonstration is the Figure 2A regression, plus qualitative geodesic/Ricci-flow plots. I examined the action integral and found that curvature and anisotropy are not merely candidate predictors of 𝒜; they are terms in its definition. Regressing an outcome on its own components cannot test whether those components govern the outcome. The reader's weakest assumption targeted Axiom 4's zero-sum curvature as a re-description of occupancy; I agree that conservation is definitional (since 1ᵀL = 0), and the field equation is un-derived and dimensionally inconsistent. The GAPDH ranking of Cys152 as most reactive is a useful, testable heuristic and should be credited; however, the quantitative 76 vs 67 kJ/mol agreement depends on the free constant κ_modal, so it is a calibration match, not a prediction. My conclusion is that the central claim is unsupported as stated; the REJECT verdict stands.","tokens_in":16277,"tokens_out":6377,"duration_ms":74909,"concrete_test":"Generate 10^3 null trajectories under the unbiased Hamming-1 stochastic dynamics (or any rates independent of R and A), compute the same action integral with the same α coefficients, and fit the same GLM of 𝒜 against mean R and A. If significant β's appear — as they must because R² and A² enter 𝒜 — the Figure 2A result is a construction artifact. Complementary check: remove α_geom(⟨R²⟩ + ⟨‖A‖²⟩) from the action definition and refit; if the geometry coefficients disappear, the action-based support for 'geometry governs' vanishes.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the main text, the action integral is defined as 𝒜 = Σ_s[ΔS(s) + λ_R R(x_s)² + λ_A ‖A(x_s)‖²]; Methods give the same object as 𝒜 = Σ_t[α_mass Σ|Δρ| + α_geom(⟨R²⟩ + ⟨‖A‖²⟩) + α_entropy S_deg]. Figure 2A then reports a 'general linear modelling of the action integral against mean curvature (R) and anisotropy (A)' with β_R = −1059.7 and β_A = +2112.8. But R² and ‖A‖² are additive components of the very action being regressed. Significant coefficients are therefore guaranteed by construction for any trajectory set with nonzero variance in R or A; the sign is determined by how these terms covary with the other action components. The statement that 'geometry shaped proteoform dynamics' is thus a tautology of the estimator, not an empirical finding. This is the load-bearing demonstration for the universal claim; Axiom 4's 'conservation' is likewise definitional (1ᵀL_φ = 0), and the GAPDH 76 vs 67 kJ/mol agreement depends on the unstated coupling κ_modal, so removing the circular regression leaves no independent evidence for the central law.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes 'Modal Geometric Field (MGF) Theory' for proteoform dynamics. Four axioms define the proteoform state space as a Boolean lattice (Axiom 1), impose volume invariance (Axiom 2), restrict transitions to Hamming-1 moves (Axiom 3), and introduce a graph-Laplacian 'curvature' field that is claimed to be a conserved geometric quantity controlling transitions (Axiom 4). The central field equation couples an energy term, this curvature, and an entropy term. The manuscript reports simulations of the resulting action integral, claims that curvature and anisotropy significantly 'shape' the action, analyzes geodesics and commutator algebras, and presents a GAPDH case study in which PDE-derived bitwise weights predict Cys152 as the primary oxidation site and give an activation energy near an empirical value. The paper concludes that modal geometry governs proteoform dynamics universally and scale-invariantly.","tokens_in":16628,"tokens_out":7230,"duration_ms":82279,"significance":"If the central claims were correct, the paper would offer a new, universal geometric framework for proteoform dynamics, with concrete predictions such as the preferred GAPDH oxidation pathway 000 → 100 → 110 → 111. Strengths include the clear presentation of the Boolean/combinatorial state space, the availability of the Julia source code, and the attempt to connect a geometric model to a specific experimental system. However, the load-bearing evidence is not independent: the 'conservation of curvature' is a mathematical identity of the graph Laplacian, the regression showing that curvature predicts the action is circular because the action includes the same curvature and anisotropy terms as additive components, and the GAPDH activation-energy comparison relies on an unconstrained coupling constant. At present the manuscript does not establish the claimed geometric governance; the central law is asserted and then fitted rather than tested.","major_comments":[{"comment":"The field equation T_{i→j} = ΔE_{i→j}·exp[ρ(x_i)·ΔH1] − R(x_i) + ΔS is dimensionally inconsistent. ΔE is an energy, ΔS is an entropy (or an energy only if multiplied by T), and R(x_i) is defined in Axiom 4 as a graph Laplacian of a dimensionless occupancy/shape field, hence dimensionless. The three terms cannot be added in a physical equation. Moreover, the stationary condition ∑_{k=0}^R P(i→j)·T_{i→j}=0 is ill-defined: the summation index k does not appear in the summand, and T is not a probability. Because this equation is the central law of the paper, the manuscript does not provide a testable physical equation.","section":"The MGF Theory Field Equation (p. 5–6)"},{"comment":"The 'conservation of curvature' is a tautology. With R = Lρ for the graph Laplacian L, the identity 1^T L = 0 implies ∑_x R(x) = 0 for every vector ρ at every instant. The text's own derivation, ∑ R = 1^T L ρ = 0, makes this explicit. This is the zero-sum property of the Laplacian, not a dynamical conservation law, and it does not depend on Axiom 2 or on any symmetry. The statement that 'curvature cannot be created or destroyed, only transported' therefore carries no physical content beyond the definition of R. The Noether analogy and the abstract's claim that conserved curvature governs dynamics are not established.","section":"Conservation of Curvature (p. 10; Supplemental Axiom 4)"},{"comment":"The regression in Figure 2A is circular. The response is defined as 𝒜 = Σ_s[ΔS(s) + λ_R R(x_s)² + λ_A ||A(x_s)||²] in the main text and as 𝒜 = Σ_t[α_mass Σ|Δρ| + α_geom(⟨R²⟩ + ⟨||A||²⟩) + α_entropy S_deg] in Methods. The predictors in the regression are R and A, which appear additively inside the response. For any trajectory ensemble with nonzero variance in R or A, the general linear model must return nonzero coefficients; the sign of each coefficient is determined by covariance with the other additive terms. Figure 2A therefore cannot support the conclusion that 'geometry shaped proteoform dynamics'; it is a property of the estimator, not an empirical finding.","section":"Modelling MGF Theory / Methods: Action integral (p. 6, p. 14)"},{"comment":"The GAPDH activation-energy comparison is not a parameter-free prediction. The text states a symmetric barrier of ≈33 kJ/mol per cysteine bit, then defines ΔG_eff^‡(b) = κ_modal |K|(1 − ln w_E(b)) with |K| ≈ 33 kJ/mol. For Cys152 (w_E = 0.388) this gives roughly 64 kJ/mol, not the claimed 76 kJ/mol. To obtain 76 kJ/mol one must choose κ_modal ≈ 1.18, an unstated free parameter. Since κ_modal is not determined independently, the reported agreement with the empirical 67 kJ/mol is a fit, not a validation of the theory.","section":"Curvature, Energy, and Entropy / Numerical example (p. 12–13)"},{"comment":"The geometric-primitive calculation depends on several arbitrary choices: Gaussian kernel width σ = 4.0 Å, cutoff r_c = 6.5 Å, Cα-only representation, the specific construction of the substrate Laplacian, the choice of source terms, and rescaling of Forman-Ricci curvature. No sensitivity analysis is provided. This calculation is the only molecular-level input that breaks the iso-potential symmetry of the three cysteine bits, so the prediction that Cys152 is the geometric primitive (000 → 100) is load-bearing. Without a demonstration that the result is robust to these modeling choices, the prediction may be an artifact of the chosen parameters.","section":"The Geometric Primitive (p. 11–12)"}],"minor_comments":[{"comment":"Axiom 3 defines single-molecule occupancy as ρ(x)=1 at the occupied mode and 0 elsewhere, but the simulations evolve 100 molecules per run and use fractional occupancies. The relationship between the single-molecule axiom and the ensemble-level field equation needs clarification.","section":"Axiom 3 vs. simulations"},{"comment":"The underbrace 'Energy' covers only ΔE·exp[...], while −R(x_i) and +ΔS are also treated as energy-like terms. The notation suggests that R and ΔS are not energy contributions, which is inconsistent with the dimensional issue noted above.","section":"Field equation notation"},{"comment":"Two definitions of Φ(ρ) are given: Φ(ρ)=max_{x,y}|R(x)−R(y)| and Φ(ρ)=∑(ρ_x−1/|ℳ|)². These are not equivalent. The text calls them 'two equivalent forms' (p. 15), which is incorrect.","section":"Ricci Flow definitions"},{"comment":"The scale-invariance test consists of running the same code for different R values. This does not test universality across different PTM types or protein structures, nor does it establish that the same κ_modal applies across systems.","section":"Scale-invariance and universality"},{"comment":"The GAPDH empirical oxidation claim cites references [22–24], but the mapping from text citations to the reference list is inconsistent (e.g., references 24–27 are about unrelated topics). There are also typos: 'Erying' for Eyring, 'In' for ln, and 'communicative' for commutative.","section":"References and typos"}],"recommendation":"reject","confidential_remarks":"The paper is creative and the code is openly available, but the central evidence is definitional or circular. The 'conserved curvature' is a Laplacian identity, the regression evidence is a tautology, and the GAPDH agreement is a free-parameter fit. These are load-bearing issues that cannot be fixed by local revision; the manuscript would need a new, independent derivation and a genuinely predictive test to support the claims. I therefore recommend rejection in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper claims a universal law: conserved curvature governs proteoform dynamics by dictating least-action transitions. That does not hold up. The curvature conservation is tautological (sum R = 1^T L_phi sigma = 0 because the Laplacian rows sum to zero), and the headline simulation is circular—the action integral is defined with R^2 and ||A||^2 terms in it, then R and A are regressed against that same action, so the significant β coefficients are guaranteed by construction. The field equation is asserted without derivation and mixes units (energy minus a bare curvature). The GAPDH activation-energy agreement (76 vs 67 kJ/mol) depends on an unstated coupling constant κ_modal, so it is calibration, not prediction.\n\nWhat is actually new: the PDE-derived bitwise weights from the atomic structure of GAPDH, lifted into a Boolean hypercube to break the degeneracy of iso-potential cysteine oxidation steps. That is a testable, reproducible heuristic (code is on GitHub), and it correctly ranks Cys152 as the most oxidation-prone, in line with known data. That part I take seriously. The surrounding framework—Boolean enumeration, Hamming-1 restriction, probability conservation—is standard Markov-chain mechanics dressed up as geometry.\n\nThe soft spots are structural, not cosmetic. Axiom 4 is the load-bearing assumption, and it is just a projection of occupancy into a graph Laplacian; if curvature is a re-description of occupancy, the field equation reduces to a Markov chain with arbitrary edge weights, and the geometric governance claim evaporates. I see no other evidence tying modal geometry to dynamics.\n\nWho benefits: someone working on redox proteomics might pick up the bitwise-weight idea as a predictor of cysteine reactivity. But the MGF theory as presented should not be cited as established. A serious referee could still help separate that kernel from the overclaim, so I would not desk-reject out of hand—I'd send it to review, expecting major revision or a resubmission of the GAPDH method alone.\n\nRecommendation: reject the theoretical package, encourage the geometric primitive as a focused methods paper.\n\nBest,","headline":"Ambitious but unsupported: the universal geometric law is a tautology and the headline regression is circular, though the GAPDH bitwise PDE weights are a concrete, testable seed.","tokens_in":17093,"tokens_out":4680,"would_cite":false,"duration_ms":52942,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that conserved modal curvature, not raw energy alone, dictates which proteoform transitions happen and in what order, with least-action paths selected on a Boolean lattice.","keywords":["proteoform dynamics","modal manifold","curvature","cysteine oxidation","GAPDH","least-action geodesics","Ricci flow","post-translational modification"],"falsifier":"Watch single GAPDH molecules as they oxidize: the model predicts Cys152 is modified first and the fully oxidized 111 state is reached mainly through 000 to 100 to 110 to 111; if a different first cysteine or a different dominant route is observed, the theory's core claim fails.","tokens_in":16116,"feed_emoji":"🧬","tokens_out":8694,"duration_ms":95428,"temperature":0.7,"pith_summary":"The paper tries to establish a field-theoretic law for proteoform dynamics, the stepwise transitions between distinct molecular variants of a protein. It proposes four axioms: possible states form a Boolean lattice, total occupancy is conserved, only single-site changes occur, and real molecular occupancy and shape project into the lattice as a conserved curvature field. The central claim is that this curvature governs which transition happens next by making activation energy relative: it distinguishes otherwise equal moves and selects least-action geodesics. If correct, the theory would predict preferred oxidation orders, such as GAPDH oxidizing Cys152 first and reaching the fully oxidized 111 state along the path 000 to 100 to 110 to 111. A reader should care because the framework offers a geometric explanation for path dependence, hysteresis, and entropy in post-translational modification dynamics, where standard statistical mechanics treats the state space as flat.","feed_headline":"Conserved curvature chooses which cysteine oxidizes first","feed_subtitle":"A four-axiom modal-geometry theory predicts GAPDH oxidizes Cys152 first and reaches the 111 state via 000 to 100 to 110 to 111","key_machinery":"The central object is the modal manifold, the Boolean hypercube whose vertices are all possible proteoform states, equipped with a conserved scalar curvature field defined as a zero-sum Laplacian of occupancy and shape. This curvature is the mechanism that differentiates iso-potential transitions: in the MGF field equation, curvature acts as a geometric penalty, and least-action selection on the lattice chooses shortest Hamming paths such as 000 to 100 to 110 to 111. Supporting machinery includes the four axioms, PDE-derived bitwise weights from the atomic structure, and a bounded Ricci-flow degeneracy that quantifies oscillations between ordered and chaotic occupancy distributions.","core_discovery":"MGF Theory asserts that proteoform dynamics is governed by conserved modal curvature. On the modal manifold of all possible modification states, each mode is one configuration, e.g., the eight cysteine redox states of GAPDH from 000 to 111. Axioms one through three fix the lattice structure, volume conservation, and first-order Hamming-1 transitions. Axiom four defines curvature as a zero-sum Laplacian of occupancy and shape, so total curvature always vanishes. The MGF field equation adds a curvature penalty to the energy cost and entropy gain of a transition: concentrated occupancy creates wells that suppress transitions, while distributed occupancy flattens the landscape and facilitates mo","pith_inferences":["My extension: the total-curvature identity sum zero holds for any occupancy because the graph Laplacian has zero-sum rows; the theory's physical content therefore depends on whether the curvature field itself, rather than the occupancy it is derived from, changes transition rates.","My extension: the Poisson and Dirichlet-energy recipe can be treated as a general predictor of oxidation order; applying it to other multi-cysteine proteins with known experimental oxidation hierarchies would either generalize or localize the GAPDH result.","My extension: the predicted Ricci-flow oscillations imply proteoform populations should show wave-like spreading in time, so time-resolved measurements of modification-state distributions could look for periodic ordering and disordering of occupancy."],"forward_implications":["If the theory is right, activation-energy barriers in proteoform networks are not fixed: they shift as occupancy redistributes curvature, producing hysteresis and path-dependent histories.","For GAPDH specifically, the theory predicts Cys152 oxidizes first and full oxidation traverses 000 to 100 to 110 to 111, giving the hyperoxidized 111 mode a definite preferred route.","If curvature governs transitions, moves with equal energy cost are distinguished geometrically, so no extra energy scale is needed to explain why one modification site is favored over another.","The same axioms apply to any number of modification sites and any post-translational modification basis, so preferred geodesics should exist for tyrosine phosphorylation, lysine acetylation, and other proteoform systems."],"supporting_citations":[{"why":"Establishes the binary reduced/oxidized cysteine basis and prior nonlinear redox dynamics that the modal manifold is built on.","marker":"[7]"},{"why":"Supplies the binomial theorem that enumerates modal strata and geodesic counts on the Boolean lattice.","marker":"[12]"},{"why":"Defines the statistical-mechanics baseline with a flat state space that MGF Theory claims to extend geometrically.","marker":"[8]"},{"why":"Provides the general-relativity analogy for coupling curvature to energy.","marker":"[9]"},{"why":"Supplies the Ricci-flow formalism used to model order, chaos, and entropy.","marker":"[21]"},{"why":"Supplies the AlphaFold GAPDH structure from which atomic coordinates for the Poisson PDE calculation are taken.","marker":"[1]"},{"why":"Gives the Eyring equation used to convert the empirical rate constant into the 67 kJ/mol activation energy the theory compares against.","marker":"[31]"},{"why":"Provides experimental evidence that Cys152 is the most oxidation-prone GAPDH cysteine, the empirical match for the predicted geometric primitive.","marker":"[25]"},{"why":"Documents the fully oxidized 111 GAPDH mode that the predicted geodesic 000 to 100 to 110 to 111 is said to reach.","marker":"[26]"}],"fun_headline_variants":["Geometry dictates which cysteine oxidizes first","Curvature conservation shapes protein fates","A four-axiom theory predicts protein oxidation order","How modal geometry steers proteoform evolution","Conserved curvature: the hidden ruler of protein states"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The argument stands on the premise that the curvature computed from a molecule's occupancy and shape is a physical agent that changes transition rates, rather than merely a mathematical summary of the occupancy itself.","fun_headline_variants_meta":{"raw":{"variants":["Geometry dictates which cysteine oxidizes first","Curvature conservation shapes protein fates","A four-axiom theory predicts protein oxidation order","How modal geometry steers proteoform evolution","Conserved curvature: the hidden ruler of protein states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000173,"raw_usage":{"total_tokens":1117,"prompt_tokens":745,"completion_tokens":372,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":489,"completion_tokens_details":{"reasoning_tokens":302}},"tokens_in":489,"tokens_out":372,"duration_ms":4695,"temperature":1.0,"reasoning_tokens":302,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T15:14:05.138443+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Watch single GAPDH molecules as they oxidize: the model predicts Cys152 is modified first and the fully oxidized 111 state is reached mainly through 000 to 100 to 110 to 111; if a different first cysteine or a different dominant route is observed, the theory's core claim fails.","supporting_citations":[{"cited_title":"N., Chatzinikolaou, P","cited_arxiv_id":null,"evidence_quote":"Establishes the binary reduced/oxidized cysteine basis and prior nonlinear redox dynamics that the modal manifold is built on."},{"cited_title":"Statistical Mechanics","cited_arxiv_id":null,"evidence_quote":"Defines the statistical-mechanics baseline with a flat state space that MGF Theory claims to extend geometrically."},{"cited_title":"Die Grundlage der allgemeinen Relativitätstheorie","cited_arxiv_id":null,"evidence_quote":"Provides the general-relativity analogy for coupling curvature to energy."},{"cited_title":"The Activated Complex in Chemical Reactions","cited_arxiv_id":null,"evidence_quote":"Gives the Eyring equation used to convert the empirical rate constant into the 67 kJ/mol activation energy the theory compares against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides experimental evidence that Cys152 is the most oxidation-prone GAPDH cysteine, the empirical match for the predicted geometric primitive."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents the fully oxidized 111 GAPDH mode that the predicted geodesic 000 to 100 to 110 to 111 is said to reach."}],"review_version":1}