{"id":"620970ef-bccd-4144-9aa1-9eecc7ecd049","arxiv_id":"2507.19919","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A new analytical framework gives density-dependent diffusivity and conductivity for a 2D lattice model of cellular aggregates with intermittent dipolar attractions, matching simulations up to moderate activity.","lead":"This paper derives equations for how fast cells in a growing bacterial colony spread (diffusivity) and how they respond to an external push (conductivity), using a lattice model of pili-driven attraction. The authors show that both transport coefficients fall as cell-cell attraction gets stronger, which they connect to the slowdown of colony spreading during aggregation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central 2D result depends on treating each row as an independent 1D exclusion model, but vertical hopping makes row occupancy non-conserved; Eqs. (30)-(31) are not yet established for the actual 2D system.","rationale":"The paper's central claim is that Eqs. (30)-(31) give the density-dependent bulk diffusivity and conductivity for the 2D pili-driven aggregation model, and the applications to transport slowdown depend on these being the true 2D coefficients. The 1D zero-range derivation appears internally consistent: with the factorized steady state and gradient property, the UMTM computation of G via Eq. (25) and the MFT identification of D and chi are standard. The empirical kernel v(m) with fitted s1 and s2 is a parameterization rather than a first-principles derivation, but since v(m) is reported to be independent of rho and p, it does not by itself invalidate the predicted density dependence. The load-bearing step is the reduction from 2D to independent rows. Row occupancy is not conserved in 2D, and the exact EM-to-UMTM map used in the derivation presupposes a closed 1D system with fixed particle number. The appendix's SSEP calculation demonstrates separability of x and y Laplacians at p=0 only; the extrapolation to p>0 is an assumption. The Fig. 5 agreement at p <= 0.7 is genuine evidence and should be credited, but it is an indirect test: a suitably chosen 1D chain model with the same G(rho) would also match those data. The modified-1D-chains test would isolate the decoupling assumption. Because this concern is exactly the reader's weakest assumption and does not overturn the moderate-p simulation support, the appropriate verdict remains CONDITIONAL, and no change from the reader's verdict is needed.","tokens_in":24600,"tokens_out":7826,"duration_ms":107022,"concrete_test":"Run the sinusoidal-relaxation protocol of Sec. IV on a modified 2D EM in which vertical hopping is disabled, producing a stack of independent 1D chains with the same total density and interaction parameter p. The Sec. III reduction predicts that the diffusivity D_1Dchains(rho,p) measured this way equals the original D_2D(rho,p) shown in Fig. 5(c); if the two curves differ beyond error bars for any rho at p <= 0.7, the row-decoupling assumption is falsified and Eqs. (30)-(31) do not describe the coupled 2D system.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The derivation of Eqs. (30)-(31) reduces the 2D exclusion model to independent 1D rows. In Sec. III the authors state that 'we can derive hydrodynamics for any row or column of the 2D system which effectively will represent the hydrodynamics of the entire system', justified by independence of the x and y components of the current. However, rows in the 2D EM are open systems: particles execute vertical hops with p-dependent rates, so row occupancy N1(t) fluctuates, and the EM-to-UMTM mapping of Sec. III A fixes the UMTM ring length to the instantaneous row particle number. That mapping is exact for a closed 1D ring, as in Ref. [36]; for a row coupled to its upper and lower neighbors, the UMTM lattice length is time-dependent and the 1D master-equation derivation of G, D, and chi does not directly apply. The appendix establishes x/y separability only for the p=0 SSEP limit; for p>0 the argument rests on absence of cross-linking between horizontal and vertical pili, which prevents one pilus binding in both directions but does not by itself decouple vertical particle exchange from horizontal gap statistics. Thus Eqs. (30)-(31) may characterize an auxiliary 1D chain model rather than the coupled 2D colony. The simulation agreement in Fig. 5 supports the reduction empirically at moderate p, and the p=0.9 mismatch is acknowledged, but the reduction itself has not been tested independently of the transport-coefficient comparison.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops a fluctuating hydrodynamics description for a two-dimensional lattice model of cellular aggregation mediated by intermittent dipolar (pili) forces. The authors reduce the 2D exclusion model to independent 1D rows/columns, map each row to an unbounded mass transfer model (UMTM), and assume a factorized steady state to derive analytic expressions for the bulk diffusivity D(ρ) and conductivity χ(ρ) [Eqs. (30)-(31)]. They then compare these expressions with Monte Carlo simulations of the full 2D model, obtaining D from the exponential relaxation of a sinusoidal density perturbation and χ from both the variance of the fluctuating current and the mean current under a small external bias. Agreement is good for p up to 0.9, and both coefficients decrease with increasing cell-cell interaction strength p, which the authors interpret as transport slowdown during colony formation.","tokens_in":24923,"tokens_out":15760,"duration_ms":168967,"significance":"If the derivation holds, this is a valuable contribution: analytic, density-dependent transport coefficients for an interacting active lattice gas are rare, and the paper's closed forms explicitly connect D and χ to the microscopic parameters (p, Np, l0, T0). The numerical verification is unusually complete—two independent measurements of χ and one of D, all consistent with the theory across densities. The main caveat is that the central 2D-to-1D reduction is not rigorously derived for p>0; the appendix proves separability only for the p=0 SSEP limit. Thus the significance is conditional on the validity, or further justification, of that reduction.","major_comments":[{"comment":"The reduction of the 2D model to independent 1D rows/columns is not established for p>0. Appendix A derives x/y independence of the current only for the p=0 SSEP limit [Eqs. (38)-(41)]. For p>0, the argument based on isotropy and absence of cross-linking between horizontal and vertical pili does not imply that vertical particle exchange decouples from horizontal gap statistics: a vertical hop changes the occupancy of a row and therefore changes the inter-particle gaps m_i that control the interaction weight v(m_i). In the UMTM construction of Sec. III A, the ring length is the instantaneous row particle number N1, which fluctuates because rows are open systems; the 1D master-equation derivation of D and χ assumes a closed ring with fixed particle number. Consequently Eqs. (30)-(31) are derived for an auxiliary 1D chain, and their validity for the coupled 2D system rests on the simulation agreement in Fig. 5, not on a derivation. I recommend either a controlled derivation of the decoupling (e.g., showing that the row dynamics in the 2D model converge to the 1D UMTM in the hydrodynamic limit) or an independent numerical test of the reduction, such as comparing row-resolved transport coefficients with full 2D results, or measuring correlations between horizontal currents and vertical occupancy fluctuations.","section":"Sec. III (before III A) and Appendix A"},{"comment":"The assertion that the microscopic jump weight v(m) is independent of ρ and p is stated without supporting data: 'We further check (not shown) that v(m) does not depend on density ρ and cell-cell interaction rate p.' Since v(m) enters every transport coefficient via Eqs. (30)-(31), a hidden p-dependence would mean that part of the p-dependence of D and χ is fitted rather than predicted. Please either show the v(m) data for multiple ρ and p values, or provide a microscopic argument for this independence.","section":"Sec. III B"},{"comment":"The rates in Appendix B, Eq. (42), are inconsistent with Eq. (7) in the main text. Eq. (7) assigns a total loss rate (q a_i + p a_i v(m_i))/2 (with each directional rate /4), leading to D = (1/4)∂G/∂ρ and χ = (1/4)G. Appendix B uses q a_i for the q-loss and p a_i v(m_i)/2 for the p-loss, yielding D = (1/2)∂G/∂ρ and χ = (1/2)G in Eq. (50). The main-text factors are the ones that agree with the p=0 SSEP limit (D=1/4) and with simulation; the appendix should be corrected to match the model definition, so the two derivations do not give different answers.","section":"Appendix B vs. Sec. III C"},{"comment":"The second-moment equation and its steady-state reduction contain apparent typographical errors: Eq. (20) as printed does not follow from the rates in Eq. (7), and Eq. (21) repeats the term '-2⟨m_i g_i⟩ + ⟨g_i⟩' twice. The final relation Eq. (22) is correct, but the intermediate equations should be fixed for the derivation to be reproducible.","section":"Sec. III E, Eqs. (20)-(21)"}],"minor_comments":[{"comment":"The notation N1 is used without definition; please define it as the number of particles in the selected row.","section":"Sec. III A"},{"comment":"The statement 'correlation vanishes beyond distance r = 0 for any p' is confusing; it should read 'for r > 0'.","section":"Fig. 4(c) caption"},{"comment":"The statement 'we can now write ⟨BiCj⟩ = ⟨Bi⟩⟨Cj⟩ if i≠j' is justified only by the observed absence of long-range mass correlations; this is an assumption rather than a proven property of the UMTM for p>0, and should be flagged as such in the text.","section":"Sec. III E"},{"comment":"The footnote [54] 'Please see the appendices for detail' should cite the specific appendix sections (VI A and VI B) rather than a bare reference.","section":"Ref. [54]"},{"comment":"The definition of MCT as 'each particle, on average, has one opportunity to jump' should be clarified: with the 1/4 hopping weights, the mean number of horizontal jumps per MCT is 1/2 per particle, which is relevant for the relationship between simulation time and the hydrodynamic time τ.","section":"Sec. II"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong candidate for publication after revision. The main issue is the justification of the 2D-to-1D reduction; the appendix inconsistency and the unsupported p/ρ-independence of v(m) also need attention. I do not see grounds for rejection, as the simulation evidence suggests the formulas are correct for moderate p, but the central claim is currently not fully supported by the derivation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a sensible, incremental step beyond the authors' 1D work, and the main results are the explicit formulas for D(rho) and chi(rho), Eqs. (30)-(31). What the paper does well: it works through the mass-dependent jump weights, proposes a factorized steady state in the UMTM, expresses the transport coefficients in terms of that steady state, and tests D and chi against simulations for p up to 0.9. The agreement is genuinely good for an interacting active lattice gas, and the paper is honest about the high-p mismatch and about the numerical evaluation of P(m).\n\nThe soft spot is the 2D-to-1D reduction. The authors argue that x and y currents are independent, so one row (or column) represents the whole system. That is not the same as showing a row is a closed 1D system. In the 2D exclusion model, vertical hops change row occupancy, so the UMTM ring length is time-dependent; the appendix proves x/y separability only in the p=0 SSEP limit. For p>0 the argument rests on the absence of cross-linking between horizontal and vertical pili, which is plausible but not a derivation. So Eqs. (30)-(31) strictly describe an auxiliary 1D chain model. The simulation agreement then gives empirical support for the reduction, but it is support from the same model, and the D and chi comparison is indirect. I would like to see a direct test: measure row occupancy fluctuations and compare against the closed-ring assumption, or probe the 1D row prediction with a perturbation that is not uniform in the transverse direction.\n\nThe two fitting constants s1 and s2 are a minor weakness. They are calibrated to microscopic bound-pair counts rather than to the transport coefficients, so not circular, but they temper the claim of a first-principles derivation.\n\nOverall, the central claim holds up at moderate activity and the paper is worth a serious referee. The reduction issue can likely be fixed with additional numerical diagnostics, so my recommendation is to send it to review, with the authors expected to address the open-row problem.","headline":"Solid incremental extension with a real soft spot: the row-reduction to 1D is an assumption that simulation support makes plausible but not rigorous; worth refereeing.","tokens_in":25418,"tokens_out":2706,"would_cite":false,"duration_ms":33391,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["82C22","82C31","82C70","60K35"],"pacs":[],"model":"deepseek-v4-flash","headline":"Two analytical formulas fix the density-dependent bulk diffusivity and conductivity of pili-driven cellular aggregates, and predict that transport slows as colonies form.","keywords":["two-dimensional cellular aggregates","pili-mediated aggregation","fluctuating hydrodynamics","bulk diffusivity","conductivity","exclusion process","zero-range process","transport coefficients"],"falsifier":"In the 2D agent-based model, measure the bulk diffusivity from the decay rate of a small sinusoidal density perturbation at $p = 0.9$ while increasing system size $L$; the paper already reports a mismatch at that activity, so if the discrepancy does not shrink as $L$ grows, the 1D row-decoupling and factorised steady-state assumptions are the point of failure.","tokens_in":24361,"feed_emoji":"🦠","tokens_out":10241,"duration_ms":100790,"temperature":0.7,"pith_summary":"Starting from a two-dimensional lattice model of Neisseria gonorrhoeae aggregation, in which hard-core cells hop by substrate-driven diffusion and by short-lived attractive pili pairs, the paper derives closed-form expressions for two macroscopic transport coefficients: bulk diffusivity $D(\\rho)$ and conductivity $\\chi(\\rho)$. These coefficients are the proportionality factors of the hydrodynamic cell current in a density gradient and under an external force, respectively. The formulas show that both coefficients decrease as the cell-cell interaction rate $p$ increases, predicting that transport slows down as colonies form. Monte Carlo simulations of the 2D exclusion model match the analytical curves well for moderate activity, with deviations at the highest activity $p = 0.9$. If the derivation holds, the density dependence of transport in such cellular aggregates is fixed by the microscopic pili parameters rather than left to phenomenological modelling.","feed_headline":"Two formulas pin down how cell transport slows during aggregation","feed_subtitle":"Bulk diffusivity and conductivity emerge from pili dynamics and match 2D simulations.","key_machinery":"The load-bearing object is the local coarse-grained observable $$G(\\rho) = q(1-\\rho) + p\\rho \\sum_{m=0}^\\infty m\\, v(m)\\, P(m|\\rho),$$ from which the transport coefficients follow as $D(\\rho) = -\\frac{1}{4}\\partial G/\\partial \\rho$ and $\\chi(\\rho) = \\frac{1}{4}\\rho G(\\rho)$. All microscopic detail enters through the jump weight $v(m) = \\frac{s_1}{2}(1 + s_2 m/l_0)\\exp(-m/l_0)$, the probability that two exponentially distributed pili facing each other across a gap $m$ form a bound pair, with $s_1$ and $s_2$ fitted to simulation. To average $v(m)$ analytically, the exclusion model is mapped to an unbounded mass transfer model (a zero-range process) in which site masses are the gaps between particles and mass-mass correlations vanish beyond nearest neighbours, giving a factorised steady state $P(m) = f(m) z^m / F(z)$. The mapping converts an interacting 2D exclusion problem into a tractable product-measure calculation, and the reverse mapping returns $D$ and $\\chi$ to the original cell density $\\rho$.","core_discovery":"The paper's central result is that the fluctuating hydrodynamics of the pili-driven aggregation model is governed by the two transport coefficients $$D(\\rho) = \\frac{1}{4}\\left[q - p \\frac{\\partial}{\\partial \\rho}\\left(\\rho \\sum_{m=0}^\\infty m\\, v(m)\\, P(m|\\rho)\\right)\\right]$$ and $$\\chi(\\rho) = \\frac{1}{4}\\left[q\\rho(1-\\rho) + p\\$rho^{2}$ \\sum_{m=0}^\\infty m\\, v(m)\\, P(m|\\rho)\\right],$$ where $v(m)$ is the probability that two pili facing each other across a gap $m$ form a bound pair and $P(m|\\rho)$ is the steady-state mass distribution of an equivalent unbounded model. The authors derive the observable $G(\\rho)$ analytically, avoiding the numerical averages of their earlier 1D study, and use macroscopic fluctuation theory to show that $D$ and $\\chi$ enter the hydrodynamic current as $-D\\nabla\\rho$ and $\\chi F$. Both coefficients decrease with increasing $p$, which the authors interpret as transport slowdown during colony formation; the equilibrium limit $p = 0$ recovers $D = 1/4$ and $\\chi = \\rho(1-\\rho)/4$ of the symmetric exclusion process.","pith_inferences":["Beyond the paper: if vertical-horizontal current coupling becomes measurable at high $p$ in larger simulations, the effective 2D conductivity could fall below the 1D-reduction prediction, since the decoupling assumption is exactly what the paper identifies as strained there.","Beyond the paper: the fitted parameters $s_1$ and $s_2$, which absorb pili lifetime and number, could be estimated directly from pili statistics in experiments; then Eqs. (30)-(31) become parameter-free predictions for colony-scale transport.","Beyond the paper: one could test a density-dependent Einstein-type relation between $D(\\rho)$ and $\\chi(\\rho)$; the paper does not claim it, but its formulas supply both coefficients from the same observable $G(\\rho)$."],"forward_implications":["If Eqs. (30) and (31) are correct, the hydrodynamic current in a colony is quantitatively fixed once the density and the microscopic parameters ($p$, $N_p$, $l_0$, $T_0$) are known.","Both $D(\\rho)$ and $\\chi(\\rho)$ fall as the cell-cell interaction rate $p$ increases, so the model predicts that colony formation itself slows further transport, a feedback that could stabilise aggregates.","At $p = 0$ the formulas reduce to the known symmetric exclusion process values $D = 1/4$ and $\\chi = \\rho(1-\\rho)/4$, connecting the active model to its equilibrium baseline.","Conductivity computed from current fluctuations in the unbiased system and from the mean current under a small bias agree with each other and with the analytical formula, supporting the macroscopic fluctuation theory description.","The same fluctuating-hydrodynamics route should apply to other contractile-force aggregates, such as tumour spheroids and organoids."],"supporting_citations":[{"why":"The earlier 1D semi-analytical study of the same pili model that this work extends to 2D and makes fully analytical.","marker":"[36]"},{"why":"Macroscopic fluctuation theory; supplies the fluctuating hydrodynamic equation and the identification of conductivity from current fluctuations.","marker":"[19]"},{"why":"Zero-range process and unbounded mass transfer model; provides the EM-to-UMTM mapping, the gradient structure, and the factorised steady states.","marker":"[40]"},{"why":"Symmetric exclusion process; supplies the equilibrium p = 0 baseline and the current-fluctuation method used to measure conductivity.","marker":"[52]"},{"why":"Experimental study of pili-induced clustering of N. gonorrhoeae; motivates the model and gives the cluster-size distribution with the condensation hump.","marker":"[2]"},{"why":"Zero-range process steady states; used for the factorised mass distribution f(m) and the site-reservoir expression P(m) = f(m) z^m / F(z).","marker":"[66]"},{"why":"Duality in single-file diffusion; used for the reverse mapping from UMTM back to EM transport coefficients.","marker":"[67]"},{"why":"Gradient-type exclusion processes; supports writing the dynamics with a discrete gradient structure that justifies the hydrodynamic limit.","marker":"[53]"}],"fun_headline_variants":["Cell transport slows as aggregates form, new formulas show","Density-dependent transport equations predict slowdown in colony formation","Why cell migration slows during aggregation: two key coefficients","Pili-driven aggregation: formulas for diffusivity and conductivity","Exact transport coefficients show why cells slow during aggregation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes that each row and column of the two-dimensional lattice is an independent one-dimensional exclusion process, so that horizontal and vertical currents never couple; if that decoupling fails, the two formulas do not describe the real 2D colony.","fun_headline_variants_meta":{"raw":{"variants":["Cell transport slows as aggregates form, new formulas show","Density-dependent transport equations predict slowdown in colony formation","Why cell migration slows during aggregation: two key coefficients","Pili-driven aggregation: formulas for diffusivity and conductivity","Exact transport coefficients show why cells slow during aggregation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001002,"raw_usage":{"total_tokens":4243,"prompt_tokens":954,"completion_tokens":3289,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":570,"completion_tokens_details":{"reasoning_tokens":3212}},"tokens_in":570,"tokens_out":3289,"duration_ms":24558,"temperature":1.0,"reasoning_tokens":3212,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T13:51:37.957493+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In the 2D agent-based model, measure the bulk diffusivity from the decay rate of a small sinusoidal density perturbation at $p = 0.9$ while increasing system size $L$; the paper already reports a mismatch at that activity, so if the discrepancy does not shrink as $L$ grows, the 1D row-decoupling and factorised steady-state assumptions are the point of failure.","supporting_citations":[{"cited_title":"Derrida, Non-equilibrium steady states: fluctuations and large deviations of the density and of the current, J","cited_arxiv_id":null,"evidence_quote":"The earlier 1D semi-analytical study of the same pili model that this work extends to 2D and makes fully analytical."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Macroscopic fluctuation theory; supplies the fluctuating hydrodynamic equation and the identification of conductivity from current fluctuations."},{"cited_title":"Henrichsen, Twitching motility, Annual Review of Mi- crobiology 37, 81 (1983), pMID: 6139059","cited_arxiv_id":null,"evidence_quote":"Zero-range process and unbounded mass transfer model; provides the EM-to-UMTM mapping, the gradient structure, and the factorised steady states."},{"cited_title":"Chakraborti, T","cited_arxiv_id":null,"evidence_quote":"Symmetric exclusion process; supplies the equilibrium p = 0 baseline and the current-fluctuation method used to measure conductivity."},{"cited_title":"Taktikos, Y","cited_arxiv_id":null,"evidence_quote":"Experimental study of pili-induced clustering of N. gonorrhoeae; motivates the model and gives the cluster-size distribution with the condensation hump."},{"cited_title":"Chakraborty and P","cited_arxiv_id":null,"evidence_quote":"Zero-range process steady states; used for the factorised mass distribution f(m) and the site-reservoir expression P(m) = f(m) z^m / F(z)."},{"cited_title":"Godr` eche, Dynamics of condensation in zero-range processes, J","cited_arxiv_id":null,"evidence_quote":"Duality in single-file diffusion; used for the reverse mapping from UMTM back to EM transport coefficients."},{"cited_title":"Bodineau and B","cited_arxiv_id":null,"evidence_quote":"Gradient-type exclusion processes; supports writing the dynamics with a discrete gradient structure that justifies the hydrodynamic limit."}],"review_version":1}