{"id":"f3ad367f-050a-40f1-b86b-5c56279b93fb","arxiv_id":"2507.22639","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A 1D active spin model for cell assemblies is solved exactly in the confluent limit and mapped to the KLS model in the fast-switching limit, yielding analytic cluster size distributions.","lead":"This paper proposes a one-dimensional lattice model of cells that move, switch direction, attract each other, and respond to contact with neighbors. It derives exact cluster size formulas in the fast-switching limit and shows how cluster size varies with activity and interaction strengths.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Q>>1 prediction <mc> ~ Q^{1/2} e^{J1/4} hinges on the ad hoc dimer-balance closure in Appendix C (condition d); this closure is not derived from the microscopic rates and is untested, so the headline scaling could fail.","rationale":"The reader's weakest_assumption correctly identifies Appendix C condition (d) as the least secure step. The free-energy minimization alone cannot determine the two length scales mc and mg; condition (d) is an externally imposed balance of dimer production and disintegration rates that is not derived from the master equation and is not independently validated. Since Eq. (26) and the Fig. 5(c) collapse follow from this closure, an error there would directly undermine the paper's main Q>>1 analytic claim. The Q<<1 KLS mapping, by contrast, is supported by the detailed-balance structure of the effective rates and by the reported MC agreement, so I do not see a comparably serious objection in that part of the paper. The suggested direct measurement of Eq. (d) would settle the concern without requiring a new theory. Because the reader already returned CONDITIONAL and this check would either confirm or refute the closure, no verdict change is needed beyond the existing conditional recommendation.","tokens_in":17306,"tokens_out":16366,"duration_ms":211084,"concrete_test":"Run MC in the Q>>1 regime (e.g., L=2000, rho=0.5, b=1, a=50, J1=1, J2=J3=0) and measure directly the steady-state gas-phase dimer production rate (emission events from dense-cluster edges that form a gas dimer before reabsorption) and the gas dimer disintegration rate; compare both sides of Appendix C Eq. (d) using the measured Gg(m), Gc(2), and <mg>. If the balance is violated beyond MC error, recompute <mc> from the measured rates and compare with Eq. (26); disagreement would confirm that the closure is load-bearing and the Q>>1 scaling needs an alternative derivation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Q>>1 central scaling <mc> ~ Q^{1/2} e^{J1/4} is fixed by an unvalidated closure, not by the free-energy minimization itself. Minimizing F gives exponential forms Gc(m)=A_c e^{-m/mc} and Gg(m)=A_g e^{-m/mg}; conditions (a)-(c) only impose total particle number, cluster number, and density balance, leaving mc and mg underdetermined. Appendix C condition (d) closes the problem by equating a phenomenological gas-phase dimer production rate to a dimer disintegration rate. The production rate assumes every particle emitted from a cluster forms a dimer before reaching the next cluster, uses a geometric-mean escape rate k_out = b(a-δ)/(b+a-δ), and the disintegration rate is evaluated with Gc(2) from the very exponential form being fitted. None of these rate expressions is derived from the microscopic transition rates, and the paper provides no direct check of Eq. (d) against measured currents. If this balance is inaccurate, Lagrange multipliers shift and the predicted scaling and collapse in Fig. 5(c) would fail. The Q<<1 KLS results are supported by MC and are not the main risk.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a 1D lattice-gas model of cell assemblies with self-propulsion, polarity switching, short-range attraction, and contact inhibition of locomotion (CIL). In the fully packed state, translation is arrested and the model reduces to an equilibrium spin model solved exactly by transfer matrix methods. In the presence of vacancies and in the absence of CIL/alignment interactions, the authors map the fast-switching limit Q<<1 to the Katz-Lebowitz-Spohn (KLS) model and derive closed-form expressions for the cluster size distribution and average cluster size, validating them with Monte Carlo simulations at Q=0.01. For Q>>1, an effective free-energy minimization is used to predict exponentially distributed clusters with average size scaling as Q^{1/2} e^{J1/4}, and a data collapse in terms of a scaled cluster size is reported. The paper closes with parameter estimates from MDCK experiments and a discussion of non-monotonic cluster-size behavior in contour plots.","tokens_in":17593,"tokens_out":17702,"duration_ms":198159,"significance":"If the central results hold, the paper provides a useful analytic reference for clustering in quasi-1D cell assays: an exact (in the fast-switching limit) cluster size distribution for the non-CIL case, and a compact scaling prediction for the CIL-dominated regime. The transfer-matrix solution of the fully packed state and the KLS-based expressions for cluster statistics are clear strengths, and the Monte Carlo comparisons at Q=0.01 support the KLS mapping. The paper also gives a falsifiable prediction for the average cluster size and a plausible route to estimate interaction parameters from experiments. However, the Q>>1 scaling rests on an ad hoc closure that is not derived from the microscopic rates, and the range of validity of the scaling is not established; these issues affect the central large-Q claim.","major_comments":[{"comment":"The Q>>1 derivation is closed by a dimer production/disintegration balance that is not derived from the microscopic transition rates in Table I. Conditions (a)-(c) fix only total particle number, cluster number, and density balance, leaving mc and mg undetermined; condition (d) then fixes them by assuming that every particle emitted from a cluster forms a dimer before reaching the next cluster, using the geometric-mean escape rate k_out = b(a-δ)/(b+a-δ), and evaluating the disintegration rate with G_c(2) from the exponential form being fitted. None of these rate expressions follows from the master equation, and no simulation test of condition (d) is reported. Since Eq. (26) and the collapse in Fig. 5(c) are consequences of this closure, the central Q>>1 scaling prediction is not yet established. I ask the authors to derive condition (d) from the microscopic transition rates or to test it directly by measuring dimer production and disintegration currents in the simulations, and to assess the sensitivity of Eq. (26) to alternative closures.","section":"Appendix C, condition (d)"},{"comment":"The stated range of validity preceding Eq. (26), namely 'Q >> e^{J0}, e^{J1}', is asserted rather than derived. As written, this condition does not guarantee that the leading term Q^{1/2} e^{J1/4} dominates the subleading terms in the full expression for <mc> in Appendix C, particularly for large J1 where the full expression contains e^{J1}-type contributions. The authors should expand the full expression explicitly, identify the subleading terms, and state the precise asymptotic condition under which Eq. (26) holds.","section":"Equation (26) and Appendix C"}],"minor_comments":[{"comment":"The text calls the reduction to KLS an 'exact mapping', but the model is only approximated by KLS in the limit Q<<1; please rephrase to 'asymptotically exact as Q→0' to avoid overstatement.","section":"Abstract and Section III.B.1"},{"comment":"The captions of Figs. 4 and 5 do not list the parameter values used for each curve (e.g., the values of Q and Jo for each data set, and the full parameter list for the collapse in Fig. 5(c)). Without these values, the comparison with Eq. (26) and the claimed universal collapse cannot be independently assessed; please add a table or explicit parameter lists.","section":"Figures 4 and 5"},{"comment":"The expression for the average energy per site in Eq. (16) has unbalanced parentheses and ambiguous notation; please define all symbols explicitly and ensure the equation is typeset correctly.","section":"Equation (16)"},{"comment":"There are several typographical errors, including 'aseemblies' in Section I, 'avarage' in Appendix C, and garbled layouts in the displayed formulas of Appendix C; a careful proofread is needed.","section":"General text"},{"comment":"The non-monotonic 're-entrant-like' behavior is discussed for ρ=0.8 only, while the abstract and scaling results emphasize the role of density; the dependence of the contour behavior on ρ is not characterized, so the generality of the phase behavior remains unclear.","section":"Section III.C.2"}],"recommendation":"major_revision","confidential_remarks":"The Q>>1 result essentially reproduces the central scaling of the authors' previous paper (Ref. [21]), and the genuinely new contributions are the KLS mapping with its closed-form cluster statistics and the inclusion of the attractive interaction. The editor may wish to weigh the incremental novelty of the Q>>1 section against the close overlap with Ref. [21]. The main technical risk is the unvalidated closure in Appendix C; if the authors can derive it from the microscopic rates or validate it against simulations, the paper would be considerably stronger."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nQuick take: this is a solid, modest paper with one genuinely new analytic result and an over-reliance on an approximate closure for the headline Q>>1 scaling. It deserves a serious referee, but the referee should push on the Appendix C closure.\n\nWhat is actually new: the Q<<1 mapping to KLS and the exact exponential cluster size distribution P(m) = (e^D - 1) e^{-Dm}, with D given by Eq. (18). The derivation is clean, the transfer matrix is standard, and the MC data at Q=0.01 support it. The fully packed solution (Eqs. 7–9) is also exact and matches simulation. The numerical phase diagrams showing non-monotonic average cluster size as a function of J1 and δ are new and experimentally relevant for 1D micropatterned assays. The paper is honest that the Q>>1 result is the same as Ref. [21] with attraction subdominant.\n\nSoft spots: the Q>>1 scaling <mc> ~ sqrt(ρ/(1-ρ)) e^{J1/4} Q^{1/2} comes from a free energy minimization that needs an extra condition, condition (d) in Appendix C, balancing a phenomenological dimer production rate against a dimer disintegration rate. This closure is not derived from the microscopic rates — it assumes every emitted particle forms a dimer, uses a geometric-mean escape rate, and evaluates Gc(2) from the very exponential form being fitted. That is a real gap. That said, the scaling is checked against Monte Carlo for several parameter sets (Figs. 4 and 5), so the result looks right even if the derivation is not self-contained. I would call it an incomplete derivation rather than a wrong result.\n\nMinor issues: no code or data deposit, no error bars on MC points, and the physical interpretation of J1/J2 in Section III.C.2 is muddled — the Hamiltonian penalizes inward-pointing pairs, but the text says J1 promotes alignment; that needs rewording.\n\nOverall: this is a reasonable extension of the authors' prior work with a new exact limit that will be useful to people working on 1D cell assays and lattice gas models. I'd send it to review, mainly to force a proper justification or direct test of condition (d) and a clarification of the CIL sign issue. Not a desk reject, but likely major revision.","headline":"A solid but modest paper: the new KLS exact limit is worth knowing, while the Q>>1 scaling rests on an ad hoc closure the referee should push on.","tokens_in":18071,"tokens_out":3789,"would_cite":true,"duration_ms":43521,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["82C22","82B20","82B26"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper derives an exact exponential cluster-size law for fast-switching cell assemblies on 1D substrates and an approximate universal scaling law for the high-activity regime.","keywords":["active matter","cell assemblies","lattice gas model","cluster size distribution","contact inhibition of locomotion","Katz-Lebowitz-Spohn model","Peclet number","quasi-1D substrate"],"falsifier":"Run the same Monte Carlo dynamics at $Q\\gg 1$ with fixed switching rate $b$ and measure the average dense-cluster size over a range of $Q$ and $J_1$; if $\\langle m_c\\rangle/(Q^{1/2}e^{J_1/4})$ drifts systematically instead of staying flat, the assumed gas-phase dimer balance is wrong. Alternatively, count dimer creation and breakup events in the gas phase directly and compare their rates with condition (d) of Appendix C.","tokens_in":17106,"feed_emoji":"🧫","tokens_out":12627,"duration_ms":129331,"temperature":0.7,"pith_summary":"The paper builds a one-dimensional lattice model of cells on a micropatterned stripe, where each cell carries a left or right polarity, hops in the direction it points, switches polarity at a rate $b$, and feels short-range attraction plus contact-inhibition-of-locomotion (CIL) and alignment interactions with neighbours. Its central proposal is that the statistics of cell clusters in this active, out-of-equilibrium system are controlled by one dimensionless ratio, the Péclet number $Q = a/b$, where $a$ is the bare hopping rate. In the fast-switching limit $Q \\ll 1$ and without CIL or alignment, the model maps exactly onto the Katz-Lebowitz-Spohn lattice gas, giving the closed-form cluster-size distribution $P(m) = (e^D - 1)e^{-Dm}$ and average cluster size $\\langle m\\rangle = e^D/(e^D - 1)$. In the opposite limit $Q \\gg 1$, minimizing an effective free energy predicts an exponential cluster-size distribution whose mean grows as $Q^{1/2}e^{J_1/4}$, with $J_1$ the CIL strength, together with an approximate universal rescaling of the full distribution. The payoff is that measured cluster histograms from quasi-1D cell assays become, in principle, direct estimates of the underlying cell-cell interaction parameters.","feed_headline":"Stripe-confined cell clusters obey an exact exponential law","feed_subtitle":"Fast-switching cells follow P(m)=(e^D-1)e^{-Dm}; at high activity, cluster size grows as sqrt(Q) e^{J1/4}.","key_machinery":"The argument runs on two calculational devices plus one controlling parameter. First, the KLS mapping: in the fast-switching limit $Q\\ll 1$ and with CIL and alignment interactions switched off, the polarity degrees of freedom equilibrate, leaving an occupation Hamiltonian $H_{\\rm KLS} = -J_o\\sum_i n_i n_{i+1}$ with $J_o = \\ln((a+\\delta)/(a-\\delta))$; rewriting occupations as Ising spins $s_i = 2n_i - 1$ lets the transfer matrix produce the exact partition function and cluster probabilities. Second, for $Q\\gg 1$ an effective Helmholtz free-energy functional counts dense clusters $G_c(m)$ of size $m$, combines the configurational entropy of arranging them with cluster energies $E_a(m) = -(m-1)J_o$ and the CIL energy $E_p(m)$, and fixes particle number and cluster number by Lagrange multipliers; minimizing $F$ yields the exponential cluster-size distributions and their means. Third, the dimensionless Péclet number $Q = a/b$ selects which device applies and sets the scaling $\\langle m_c\\rangle \\propto Q^{1/2}$, so the model's regime of validity is written directly into its main predictions.","core_discovery":"At the paper's heart is the claim that a minimal driven lattice gas, not a hydrodynamic description, captures the emergent clustering of confined cell sheets. For a confluent monolayer the model is an equilibrium spin chain solved exactly by transfer matrices: mean polarization vanishes, polarization correlations decay exponentially with a length set by $\\Delta J = J_1 - J_2 + 4J_3$, and a polarizing field produces the magnetization response of Eq. (9). With vacancies, the fast-switching limit without CIL or alignment is an exact KLS model, and the transfer-matrix solution yields the exponential cluster-size distribution $P(m) = (e^D - 1)e^{-Dm}$, with decay constant $D$ given by Eq. (18); the average cluster size follows as $\\langle m\\rangle = e^D/(e^D - 1)$, reducing at density $\\rho = 1/2$ to $\\langle m\\rangle = 1 + e^{J_o/2}$. In the high-activity regime $Q \\gg 1$, minimizing the effective Helmholtz free energy over cluster numbers gives an exponential dense-cluster distribution $P_c(m) = A_c e^{-m/m_c}$ whose mean behaves at leading order as $\\langle m_c\\rangle = (\\rho/(1-\\rho))^{1/2} e^{J_1/4} Q^{1/2}$, independent of the attractive strength $J_o$; rescaling cluster sizes by $\\omega = m((1-\\rho)/(Q\\rho))^{1/2} e^{-J_1/4}$ collapses distributions from different parameters onto one curve in the approximate sense stated in the paper. The same free-energy framework underlies the reported non-monotonic, re-entrant dependence of average cluster size on CIL strength, hopping rate, and attraction.","pith_inferences":["Because the low-$Q$ formula is exact, it invites an inverse analysis the paper does not develop: fitting experimental $P(m)$ histograms to $P(m)=(e^D-1)e^{-Dm}$ yields a value of $J_o$, and repeating the fit at several densities tests the model itself rather than assuming it.","The high-$Q$ prediction that $\\langle m_c\\rangle$ is, to leading order, independent of the attractive strength suggests that mutations altering cell adhesion may leave cluster growth unchanged when CIL dominates; a crossover experiment varying adhesion at fixed CIL would probe this boundary.","The paper's own caveat that strict exclusion and unscaled switching and hopping suppress motility-induced phase separation in one dimension suggests a concrete next test: relaxing particle crossing or rescaling rates with system size could make macroscopic phase separation appear, and the scaling law derived here would then mark the crossover."],"forward_implications":["In the fast-switching regime, the closed form $P(m) = (e^D - 1)e^{-Dm}$ makes the effective attraction $J_o$ directly inferable from measured cluster-size histograms.","At high activity, dense-cluster size grows as $Q^{1/2}e^{J_1/4}$ and only weakly depends on attraction, so in this regime CIL strength is the dominant control knob for cluster growth.","Rescaling cluster sizes by $\\omega = m((1-\\rho)/(Q\\rho))^{1/2}e^{-J_1/4}$ should collapse cluster-size distributions across densities, activities, and CIL strengths onto a single exponential master curve.","For confluent packing, the model predicts no spontaneous polar order and an exponentially decaying polarity correlation function, with correlation length controlled by $\\Delta J$; this is a measurable prediction for dense monolayers."],"supporting_citations":[{"why":"Supplies the quasi-1D micropatterned-substrate experiments and the cell speeds, densities, and polarity-switching rates used to estimate Q and the CIL parameters.","marker":"[20]"},{"why":"Supplies the effective free-energy minimization and the Q^{1/2} e^{J1/4} cluster-size law that this paper extends to include attractive interactions.","marker":"[21]"},{"why":"Supplies the persistent-exclusion-process entropy-minimization argument used for the high-activity cluster-size distribution.","marker":"[17]"},{"why":"Supplies the KLS model, its exact solution, and the Ising representation through which the low-activity cluster statistics are computed.","marker":"[30-32]"}],"fun_headline_variants":["Exact exponential law governs cell clusters on stripes","Active spin model yields exact cluster-size law","Stripe geometry: cell clustering follows exact exponential law","Mapping to KLS model gives exact cluster law for cells"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"In the high-activity regime the paper assumes, rather than derives, a steady-state balance between the creation and breakup of two-cell clusters in the sparse gas phase (condition (d) in Appendix C); the predicted $Q^{1/2}$ scaling would fail if that balance is inaccurate.","fun_headline_variants_meta":{"raw":{"variants":["Exact exponential law governs cell clusters on stripes","Active spin model yields exact cluster-size law","Stripe geometry: cell clustering follows exact exponential law","Mapping to KLS model gives exact cluster law for cells"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000916,"raw_usage":{"total_tokens":4054,"prompt_tokens":1191,"completion_tokens":2863,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":807,"completion_tokens_details":{"reasoning_tokens":2802}},"tokens_in":807,"tokens_out":2863,"duration_ms":24791,"temperature":1.0,"reasoning_tokens":2802,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T11:26:33.183140+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same Monte Carlo dynamics at $Q\\gg 1$ with fixed switching rate $b$ and measure the average dense-cluster size over a range of $Q$ and $J_1$; if $\\langle m_c\\rangle/(Q^{1/2}e^{J_1/4})$ drifts systematically instead of staying flat, the assumed gas-phase dimer balance is wrong. Alternatively, count dimer creation and breakup events in the gas phase directly and compare their rates with condition (d) of Appendix C.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the quasi-1D micropatterned-substrate experiments and the cell speeds, densities, and polarity-switching rates used to estimate Q and the CIL parameters."},{"cited_title":"Szabo and et al., Phys","cited_arxiv_id":null,"evidence_quote":"Supplies the effective free-energy minimization and the Q^{1/2} e^{J1/4} cluster-size law that this paper extends to include attractive interactions."},{"cited_title":"Scarpa and et al., Biol","cited_arxiv_id":null,"evidence_quote":"Supplies the persistent-exclusion-process entropy-minimization argument used for the high-activity cluster-size distribution."}],"review_version":1}