{"id":"c682ddab-c60a-4e73-bb0b-801da41959e7","arxiv_id":"2603.06478","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The spatial Muller's ratchet particle system converges to an infinite reaction-diffusion PDE system whose mutation ratios, Fisher-KPP spreading speed and no-surfing behaviour are now rigorously established.","lead":"This paper proves that a spatial particle model of Muller's ratchet — organisms that move, reproduce and die on a line while randomly accumulating harmful mutations — converges, at the right scale, to an infinite system of reaction-diffusion equations. It then derives rigorous bounds on mutation ratios and the invasion speed, and shows that in this deterministic limit deleterious mutations do not surf the expansion wave.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central convergence theorem depends on the uniform L^p moment bound imported from companion paper [56]; if that bound fails, tightness and Theorem 2.1 collapse.","rationale":"The reader identified the uniform moment bound (2.27)/(Theorem 3.3) as the weakest assumption, and my independent review agrees. The entire hydrodynamic-limit proof—tightness in D([0,∞), M(R)^{N0}), tightness in L^{4 deg q_-}, and the Green's-function martingale estimates—is scaffolded on this imported estimate. I also examined the internal machinery: the tightness criterion, Aldous' criterion, the uniqueness argument via Grönwall, and the Feynman–Kac analysis of the PDE limit. I found no internal contradiction or unsupported step beyond the companion-paper dependence. The paper is transparent about importing the moment bound, and the bound is plausible under Assumption 3, but the manuscript cannot stand alone. Since the reader already conditioned acceptance on verification of [56], no verdict adjustment is needed.","tokens_in":86863,"tokens_out":8218,"duration_ms":79438,"concrete_test":"Obtain companion paper [56] and verify Theorem 2.3 (especially the correlation-function proof of (2.27)): check that the constant C_r(t) is independent of N and does not require an a priori bound on ||η^N||. If that proof is sound, Theorem 2.1 is supported; if not, the hydrodynamic limit is unproved. Alternatively, as a numerical sanity check, simulate the single-type version with q+(u)=r(Bu+1), q-(u)=r(Bu+1)u, initial density 1, and compute sup_{t≤1} E[||u^N(t)||^p] for p=8 for N=100,200,400; boundedness in N would support the bound, though not settle it.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 2.1's proof rests on two results imported from the companion paper [56]: existence of the S_N-valued Markov process (Theorem 3.2) and the uniform moment bound (2.27)/(Theorem 3.3). Every tightness estimate in Lemmas 4.4–4.7, the L^{4 deg q_-} tightness criterion in Lemma 6.4/6.6, and the characterization of the limit in Lemma 6.8 invoke Theorem 3.3. In particular, Lemma 6.4(i) requires uniform bounds on E[||u^N||^{8 deg q_-}], and Lemma 4.6 uses higher moments to control F_k(u^N). If the local ℓ^1 moments blow up as N→∞ (e.g. because birth events create arbitrarily many new types before death regulation kicks in), tightness and hence convergence to (2.11) fail. The paper explicitly states this estimate is proved in [56] and does not reproduce the argument; thus the present manuscript cannot be assessed in isolation. No internal inconsistency is apparent, and the estimate is plausible under deg q+ < deg q- (death rate dominates at high density), but it is the single most load-bearing unverified premise.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves a functional law of large numbers for a spatial model of Muller's ratchet: an interacting particle system with infinitely many mutation types, random-walk migration, and density-dependent birth/death rates. Under Assumptions 1–4, the approximate density process is shown to converge in distribution, in the J1 topology on D([0,∞), M(R)^{N0}), to a unique continuous mild solution of the infinite reaction–diffusion system (2.11) with reaction term (2.12). Under monostable/Fisher–KPP assumptions, the authors also derive quantitative bounds on the ratios u_k(T,x)/u_0(T,x), the spreading speed into empty habitat, a high-probability analogue for the particle system (Theorem 2.8), and a tracer-dynamics result indicating that, in the deterministic PDE limit, deleterious mutations do not surf population waves. The proof introduces a tightness criterion for interacting particle systems in L^p([0,T]×R, λ̂; ℓ1) spaces and uses a Green's function representation plus Feynman–Kac estimates. The central convergence theorem is conditional on two results imported from the companion paper [56]: existence of the S_N-valued Markov process (Theorem 3.2) and, in particular, the uniform local ℓ1 moment bound (Theorem 3.3 / eq. (2.27)). The manuscript states that this moment bound is proved in [56] and does not reproduce the argument.","tokens_in":87149,"tokens_out":6047,"duration_ms":65145,"significance":"If the imported moment bound is correct, the paper is a significant contribution. It confirms non-rigorous conjectures of Foutel-Rodier and Etheridge, rigorously replaces a nonlinear interacting particle system with infinitely many types by a deterministic PDE description, and does so without a priori bounds on the local number of particles. The PDE results in Theorems 2.5–2.9 are quantitative and parameter-free in the sense that the constants α_k(Q_min/Q_max)^k, the spreading speed c*, and the Poisson-type limit (2.23)–(2.24) are derived from the model parameters rather than fitted. The new L^p tightness criterion (Lemma 6.4) is a methodological contribution that may be useful beyond this model. The authors are also appropriately careful in stating that the gene-surfing question is answered only for deterministic waves in the PDE limit, not for the stochastic particle system. However, the main convergence theorem cannot be assessed in isolation: its proof invokes a high-order uniform moment bound from a companion paper, and every tightness and characterization estimate in Sections 4–6 relies on it. The significance is therefore conditional on the validity and availability of that","major_comments":[{"comment":"The proof of Theorem 2.1 is not self-contained at its most load-bearing point. Estimate (2.27) — stated here as Theorem 3.3 and imported from the companion paper [56, Thm 2.3] — is invoked in Lemma 4.4, Lemma 4.6, Lemma 4.7, Lemma 6.4(i), Lemma 6.6, and Lemma 6.8. In particular, Lemma 6.4 requires uniform bounds on E[||u^N(t,x)||^{8 deg q-}_{ℓ1}], and Lemma 4.6 needs higher moments of the local ℓ1 density. The manuscript explicitly says the proof is in [56] and does not reproduce it. If this uniform local moment bound fails for the required powers as N→∞, the tightness in L^{4 deg q-}([0,T]×R, λ̂; ℓ1) and hence the characterization of the limiting density in Proposition 6.2 and Theorem 2.1 collapse. This is not an internal inconsistency, and the bound is plausible under Assumption 3 (deg q+ < deg q-), but as submitted the central claim is conditional on an unverified import. I request ei","section":"§2.6 / Theorem 3.3 / eq. (2.27)"},{"comment":"The existence and martingale properties of the S_N-valued process (η^N(t))_{t≥0} are also imported from the companion paper [56, Theorem 2.2] and stated as Theorem 3.2. This is a second external premise of Theorem 2.1, since the generator calculation and all subsequent martingale arguments require the process to exist as a strong Markov process in the weighted ℓ1-type state space S_N. The manuscript states the result clearly, but does not prove it. I regard this as less concerning than the moment bound, because the existence construction is described in outline and is presumably contained in [56]; nevertheless, the submitted paper alone does not establish that the process used in the proof of Theorem 2.1 is well defined. The authors should state explicitly the status of [56] (published, accepted, or preprint under review) and, if necessary, include the relevant existence proof or a detai","section":"§3.1, Theorem 3.2"}],"minor_comments":[{"comment":"The displayed bound contains a redundant supremum over x∈L^{-1}Z of a quantity independent of x: it should read sup_{x∈L^{-1}N Z} E[||u^N(t,x)||^p_{ℓ1}] ≲_{p,T} ||f||^p_{L∞(R;ℓ1)} + 1. As written, the right-hand side has the same value for every x.","section":"§2.6, eq. (2.27)"},{"comment":"The proof of the summability/finiteness of Φ(T) uses the strict inequality 0 < s_k < c for all k≥J. Assumption 2 only gives s_k ≥ 0 and lim s_k = 0; if some s_k = 0, the displayed inequality is not strictly positive on the left. This is easily patched by replacing '0 < s_k' with 's_k ≤ c' and treating zero separately, but the current statement is formally not implied by the assumptions.","section":"Lemma 8.5, eq. (8.14)"},{"comment":"The paper is long and the notation is heavy; a table of notation for the many constants (C_T, C_{q+,q-,f}, C^{(i)}_T, etc.) would improve readability. The frequent reuse of C^{(i)} names with different meanings across lemmas is occasionally confusing, though not mathematically harmful.","section":"General presentation"},{"comment":"The title question 'Can deleterious mutations surf deterministic population waves?' is answered only in the PDE limit, and the authors are explicit about this in the text. It may be worth adding a sentence in the abstract or introduction emphasizing that the stochastic surfing question remains open, so that readers do not over-interpret the title.","section":"§2.4 / Theorem 2.9"}],"recommendation":"major_revision","confidential_remarks":"This is a strong paper, and the PDE analysis in Sections 7–8 is careful and convincing. The obstacle to acceptance in its current form is the reliance on Theorem 3.3 from the companion paper for every tightness estimate in the main convergence theorem. If the editor can verify that [56] is available and correct, I would view the central result as defensible and likely publishable after a modest revision. I encourage the editors to request that the authors either include the proof of the moment bound in an appendix or make the dependence on [56] explicit and verifiable; otherwise the main theorem cannot be assessed from this manuscript alone."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real result, not a conjecture dressed up as a theorem. The paper proves weak convergence of the spatial Muller's ratchet to an infinite PDE system, confirms the Foutel-Rodier-Etheridge generator calculation, and then extracts quantitative mutation-ratio bounds, a Fisher-KPP spreading speed, and a no-surfing theorem for the deterministic limit. The proof is serious: Green's function representation, a new tightness criterion in L^r(ell1), uniqueness of mild solutions in Bochner spaces, and a Feynman-Kac analysis for the long-time behaviour. There are no fitted parameters, and the authors are explicit about what they do not prove (stochastic surfing, pushed waves).\n\nWhere I'd be careful: the floor of the argument is not in this manuscript. Both the existence of the S_N-valued Markov process and the uniform L^p moment bound (2.27), stated here as Theorems 3.2 and 3.3, come from the companion paper [56]. Every tightness estimate in Section 4 and the L^p tightness criterion in Section 6 lean on that moment bound; if local ell1 moments blow up as N grows, Theorem 2.1 collapses. That is a real dependency, and a referee should verify [56] before trusting the main theorem. Nothing in the text suggests the bound is false -- the death-rate-dominates-birth assumption makes it plausible -- but it is unproven in this paper.\n\nThe final sections on the speed and tracer results (8.3-8.4) were not fully readable from the supplied text, and the numerical figures ship no code. Those are minor; the central argument appears intact.\n\nThis paper deserves a serious referee. I'd send it out, with a request that the referee read the companion paper and confirm Theorems 3.2 and 3.3. If those hold, the results stand. I'd cite it if I worked in this area, and I'd put it on the reading-group list.","headline":"A careful and likely correct hydrodynamic limit for the spatial Muller's ratchet, with the caveat that the load-bearing moment bound is imported from the companion paper and must be checked.","tokens_in":87665,"tokens_out":1864,"would_cite":true,"duration_ms":18776,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60K35","35K57","60F17","92D15"],"pacs":[],"model":"deepseek-v4-flash","headline":"A functional law of large numbers converts the spatial Muller's ratchet into a unique infinite reaction-diffusion PDE system with explicit mutation-load bounds.","keywords":["Muller's ratchet","hydrodynamic limit","reaction-diffusion PDE","interacting particle systems","infinite-type systems","gene surfing","spreading speed","tightness in Bochner spaces"],"falsifier":"Compute or simulate the local moment E[||u^N(t,x)||^p_{ℓ1}] for the same particle model but with birth-degree at least death-degree (deg q+ ≥ deg q−): if it is unbounded in N, the principal tightness input fails. Alternatively, simulate the PDE (2.11)–(2.12) with only mutants present initially and no wild-type: Theorem 2.9 predicts the total labelled density decays to zero uniformly, so a PDE simulation showing a persistent travelling wave would contradict it.","tokens_in":86624,"feed_emoji":"🧬","tokens_out":5418,"duration_ms":51129,"temperature":0.7,"pith_summary":"The paper establishes that, as the population-size parameter grows, the random spatial Muller's ratchet particle system—where individuals carry arbitrarily many deleterious mutations and reproduce and die at density-dependent rates—converges in distribution to the solution of an infinite system of reaction-diffusion PDEs. The limit is unique, and the convergence comes with quantitative control: under monostable or Fisher-KPP conditions, the ratio of the density of k-mutation carriers to wild-type carriers is bounded between two explicit sequences, and the population invades empty habitat at the explicit speed sqrt(2m((1-mu)q+(0)-q-(0))). This confirms earlier non-rigorous generator calculations and turns a stochastic many-species problem into a deterministic PDE problem. It also answers the motivating biological question: in the PDE limit, deleterious mutations do not surf the expansion wave; the mutant fraction at the front descends from the wild-type rather than from pre-existing mutants.","feed_headline":"Deleterious mutations cannot surf deterministic waves","feed_subtitle":"Large-N Muller's ratchet converges to a PDE whose mutant front rides at wild-type speed.","key_machinery":"The argument runs through three devices. (1) A Green's function representation writes the local density as a random-walk semigroup plus a martingale and a finite-variation term; this yields space-time equicontinuity estimates and control of large-type tails. (2) A new tightness criterion for interacting particle systems in Bochner L^p spaces (spaces of functions taking values in ℓ1, based on a compactness characterisation of Díaz–Mayoral) controls the entire sequence of densities in L^{4 deg q−}([0,T]×R, λ̂; ℓ1), giving almost-everywhere convergence of densities and hence passage through the nonlinearity. (3) A Feynman-Kac formula for the PDE system turns the ratio bounds into comparisons wi","core_discovery":"Theorem 2.1: under space renormalization m_N/L_N^2 -> m, polynomial birth/death rates with deg q+ < deg q-, and initial data converging to f, the rescaled occupancy process converges in distribution on D([0,∞), M(R)^{N0}) with J1 topology to a unique continuous M(R)^{N0}-valued process whose densities form the unique non-negative mild solution of partial_t u_k = (m/2) Δ u_k + F_k(u), with F_k(u) = q+(||u||_ℓ1)(s_k(1−μ)u_k + 1_{k≥1} s_{k−1} μ u_{k−1}) − q−(||u||_ℓ1) u_k and u(0)=f. Corollaries: under monostability, u_k/u_0 is uniformly sandwiched between sequences converging to α_k(Qmin/Qmax)^k and α_k(Qmax/Qmin)^k; under Fisher-KPP dynamics, the spreading speed is the explicit formula above;","pith_inferences":["Editorial: the L^p tightness criterion is not specific to this model and should apply to other spatial birth-death processes with infinitely many types or unbounded local density, as long as analogous local moment bounds are available.","Editorial: the no-surfing conclusion is established for the deterministic PDE; in the stochastic system there may be a noise-driven surfing mechanism at the wave edge that disappears in the N→∞ limit, and this could be tested with a sharp large-deviation analysis at the moving front.","Editorial: the ratio sandwich depends on q+ only through Qmin/Qmax on [0,1], suggesting some universality of the mutation-load profile across a wide class of density-dependent growth laws; a testable prediction is that the front spectrum matches the equilibrium profile when q+ is constant.","Editorial: the weak-selection, low-mutation limit of the equilibrium profile α_k approaches the Poisson distribution (μ/s)^k/k!, so forward-time simulations or barcode range-expansion data could look for this signature as a fingerprint of the PDE limit."],"forward_implications":["If the claims hold, the stochastic spatial Muller's ratchet is asymptotically described by a deterministic infinite PDE, so questions about densities, mutation load, and speed can be posed and answered in the PDE rather than in the particle system.","Theorem 2.5 implies that at large times, the fraction of the population carrying k mutations stays within a compact interval, and when q+ is constant on [0,1] the ratio converges uniformly to the equilibrium profile α_k.","Under Fisher-KPP conditions, Theorem 2.7 gives the exact spreading speed and shows that every mutation class travels at the same speed as the wild-type class.","Theorem 2.8 transfers the PDE ratio bounds back to the finite-N particle system with high probability, for large enough N and after sufficiently long time.","Theorem 2.9 shows that in the PDE limit, descendants of the initially mutant subpopulation die out uniformly; deleterious mutations do not surf deterministic population waves."],"fun_headline_variants":["Spatial ratchet: functional LLN yields PDE limit","Mutation load in space: PDE convergence proven","Muller's ratchet in space: weak to PDE","Particle system to PDE: ratchet bounds","Spatial ratchet: new tightness, PDE limit"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that all high moments of the local rescaled population density are uniformly bounded as N grows—sup_N sup_{t,x} E[||u^N(t,x)||^p_{ℓ1}] < ∞—a bound imported from the authors' companion article; every tightness and equicontinuity estimate in the convergence proof invokes it, and if local ℓ1 moments blow up as N→∞, the law of large numbers collapses.","fun_headline_variants_meta":{"raw":{"variants":["Spatial ratchet: functional LLN yields PDE limit","Mutation load in space: PDE convergence proven","Muller's ratchet in space: weak to PDE","Particle system to PDE: ratchet bounds","Spatial ratchet: new tightness, PDE limit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000237,"raw_usage":{"total_tokens":1390,"prompt_tokens":836,"completion_tokens":554,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":580,"completion_tokens_details":{"reasoning_tokens":476}},"tokens_in":580,"tokens_out":554,"duration_ms":6073,"temperature":1.0,"reasoning_tokens":476,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T18:45:05.434074+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute or simulate the local moment E[||u^N(t,x)||^p_{ℓ1}] for the same particle model but with birth-degree at least death-degree (deg q+ ≥ deg q−): if it is unbounded in N, the principal tightness input fails. Alternatively, simulate the PDE (2.11)–(2.12) with only mutants present initially and no wild-type: Theorem 2.9 predicts the total labelled density decays to zero uniformly, so a PDE simulation showing a persistent travelling wave would contradict it.","supporting_citations":[],"review_version":1}