{"id":"067bb314-90dd-4e80-8cca-317ca8f7d0f2","arxiv_id":"1908.05214","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A bulk-surface reaction-diffusion model with oligomerization and positive feedback exhibits a threshold: above a feedback strength, uniform protein distributions destabilize into a single-patch steady-state.","lead":"This paper builds a mathematical model of proteins diffusing in the cell interior and clustering on the cell surface, and shows that when large clusters strongly recruit new protein from the interior, the surface pattern collapses into a single dense patch. The result is a theoretical explanation of how positive feedback plus mass conservation can produce a localized membrane-protein cluster without a pre-existing spatial cue.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Single-patch threshold is asserted from a few small-perturbation simulations; nonlinear selection of exactly one patch is the least secured condition.","rationale":"The paper is a competent bulk-surface Turing analysis. Mass conservation (Prop. 2.1) and the small-kb stability theorem (Thm. 3.1) are real independent contributions, and the numerical phase diagrams are plausible. However, the headline claim is not just that high kb destabilizes homogeneity; it is that the system forms a single-patch steady state. That second part requires a global or nonlinear statement about the attractor. The manuscript offers no bifurcation analysis, no basin-of-attraction study, and no code; the authors themselves flag the lack of a formal explanation and the reliance on numerical simulation for N=3. Because Regions 2 and 3 are bistable, there is no a priori reason that all unstable directions lead to the same one-patch state; mode competition or multi-patch states are possible. The proposed test, continuation plus multi-start Newton, directly checks existence, stability, and uniqueness of the one-patch attractor. If it passes, the conditional verdict should stand; if it fails, the central claim must be weakened. This is consistent with the reader's CONDITIONAL verdict, so no verdict change is needed.","tokens_in":25064,"tokens_out":15900,"duration_ms":156596,"concrete_test":"For the N=2 reduced system (2.23)-(2.24) on the unit sphere, discretize a1 and a2 in spherical harmonics up to degree L=40 and use pseudo-arclength continuation in kb along the branch that bifurcates from the homogeneous steady state at the Region-1 boundary for the Figure 5 parameter set. Then check (i) the branch consists of one-patch steady-states and is linearly stable for all kb above the bifurcation, and (ii) Newton solves from at least 100 independent random initial conditions at amplitudes 10^-10 and 10^-2, for kb in Regions 1, 2, and 3, converge to the same one-patch branch modulo rotation, with no stable two-patch or time-periodic attractors. If the continuation finds a disconnected or unstable one-patch branch, or the multi-start search finds other attractors, the abstract's 'drives the formation of a single-patch steady-state' claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's main claim is that high AN-dependent recruitment kb drives formation of a single-patch spatially heterogeneous steady-state. Linear-stability analysis can at most show that the homogeneous state is unstable for selected eigenmodes and parameters; it says nothing about which non-homogeneous state is approached. The paper's evidence for the single-patch attractor is numerical: four parameter points in Regions 1, 2, and 3 for N=2 (Fig. 5), one analogous N=3 case, and time traces starting from a single random perturbation of amplitude ε=10^-10 (Fig. 6, S5). Section 4.3 states the authors avoided solving the reduced system (2.23)-(2.24) and instead solved the bulk-surface system with \\tilde D=10^8. The Discussion explicitly says 'we lack a formal explanation for the emergence and robustness of the single patch steady-state' and calls the existence a 'hypothesis'. Since the system is mass-conserving and bistable in Regions 2 and 3, multiple stable patterns (e.g., two-patch or multi-patch states) could coexist; a single small random perturbation cannot rule them out. No code or data are provided, so the reported one-seed observations cannot be independently checked. Thus the central threshold-to-single-patch claim is not yet secured.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a bulk-surface reaction-diffusion model for membrane-protein clustering, in which cytosolic ligands bind to membrane receptors, form oligomers via Smoluchowski-type mass-action kinetics, and the largest oligomers recruit further cytosolic ligands through a linear feedback term in the boundary flux. The authors non-dimensionalize the model, prove mass conservation, reduce it formally to a surface-only system with a nonlocal functional when cytosolic diffusion is infinite, and derive a linear stability framework. For N=2 they prove a sufficient condition (Theorem 3.1) guaranteeing a unique homogeneous steady state and no diffusion-driven instability when the feedback strength k_b is sufficiently small. The central claim of the paper, stated in the abstract, is a threshold phenomenon: sufficiently high k_b destabilizes the uniform state and drives the formation of a single-patch spatially heterogeneous steady state. This claim is supported mainly by numerical simulations for N=2 and N=3 at a small number of parameter points, using one random perturbation per point, and the authors explicitly state in Section 5 that they lack a formal explanation for the emergence and robustness of the single-patch state.","tokens_in":25360,"tokens_out":5076,"duration_ms":51774,"significance":"If the threshold-to-single-patch claim were fully established, the model would provide a plausible mechanism for spontaneous membrane-protein clustering driven by oligomerization feedback, with relevance to cell polarization and amyloid aggregation. The paper has genuine strengths: the bulk-surface formulation is novel in combining Smoluchowski aggregation with a geometric PDE setup; the mass-conservation proof (Proposition 2.1) is clean; and Theorem 3.1 is a rigorous, if specialized, small-feedback stability guarantee for N=2. The linear stability framework for the reduced nonlocal system is clearly presented and the authors are candid about the limitations of the single-patch evidence. However, the headline result is currently a simulation-based observation for a few selected parameter points with single-seed initial conditions, and the paper itself admits that no formal explanation is available; as a consequence, the central claim is not yet secured at the level suggested by the abstract.","major_comments":[{"comment":"The central claim that high k_b drives the formation of a single-patch steady state is not established by the evidence presented. Linear stability analysis (Section 3.2.2) can only show that the homogeneous state is unstable for certain eigenmodes; it says nothing about which nonhomogeneous state is selected. The numerical evidence consists of one random perturbation (epsilon = 10^-10) for each of four parameter points in Regions 0-3 for N=2 (Fig. 5), one analogous N=3 case (Fig. S4 and Fig. S5), and time traces from a single simulation each (Fig. 6). Since the system is mass-conserving and bistable in Regions 2 and 3, multiple stable patterned states could coexist, and a single small random perturbation cannot rule out multi-patch or other attractors. The Discussion (Section 5) explicitly states \"we lack a formal explanation for the emergence and robustness of the single patch steady-state\" and labels the existence of the single patch a hypothesis. Because the abstract presents the single-patch threshold as the main result, this gap is load-bearing rather than cosmetic. The authors should either add systematic multi-seed simulations with varied perturbation amplitudes, numerical continuation/bifurcation analysis of nonhomogeneous steady states, or a reduced 1D analysis that can rigorously address the selection of a single patch, or revise the abstract and central claims to describe the result as numerical evidence for a conjectured threshold.","section":"§4.3, Fig. 5, Fig. 6, §5"},{"comment":"The numerical simulations approximate the reduced nonlocal surface system (2.23)-(2.24), which is derived in the limit D_u -> infinity, by solving the bulk-surface system (2.16)-(2.21) with the finite value \\tilde D = 10^8. The error between \\tilde D = 10^8 and the asymptotic limit \\tilde D = infinity is not quantified, and no convergence study in \\tilde D is reported. Since the single-patch pattern is obtained from this approximate numerical system, it is important to demonstrate that the observed pattern is not an artifact of the finite diffusion value and that the selected steady state is stable with respect to increasing \\tilde D. Without such a check, the connection between the numerical pattern and the reduced model, which is the object of the analytical stability study, remains incomplete.","section":"§2.5 and §4.3"},{"comment":"The existence and uniqueness theory for the model is not established. Section 3.4 states that the results of Sharma and Morgan, which apply to a similar system, \"appear to be too restrictive to cover the nonlinearities arising here\" and that the authors \"expect\" similar results can be shown. Since the main biological claim is supported by numerical simulations of the PDE system, a rigorous assurance that solutions exist, are unique, and depend continuously on data for the parameter regimes studied would materially strengthen the paper. At minimum, the authors should clearly mark that the well-posedness of (2.2)-(2.7) is an open question, rather than presenting the numerical solutions as unproblematic.","section":"§3.4"}],"minor_comments":[{"comment":"Equation (2.4) reads d_t a_2 = D_2 Delta a_1 + ...; the Laplacian should act on a_2, not a_1. The corresponding dimensionless equation (2.19) is correct, so this appears to be a typographical error.","section":"Eq. (2.4)"},{"comment":"The caption states \"for the eigenmodel = 1\"; this should read \"for the eigenmode index l = 1\".","section":"Fig. 3 caption"},{"comment":"The sentence \"the instabilities emerge in the bistability region (Regions 2 and 3) and also in the single steady-state regions Regions ( 0 and 1)\" is garbled; Regions 0 and 1 are the single-steady-state regions, so the parenthetical should be removed or rewritten for clarity.","section":"§4.2"},{"comment":"The text contains \"has been ofter related to a single-patch steady-state pattern\"; \"ofter\" should be \"often\".","section":"§5"},{"comment":"No code or data are provided in the manuscript or supplementary material. Given that the central claim relies on numerical simulations, a code/data supplement would significantly aid reproducibility and allow independent checking of the reported single-seed observations.","section":"Reproducibility"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope and contains a useful modeling framework plus one clean rigorous result (Theorem 3.1), but the abstract's main claim goes beyond what the current evidence supports. The authors are honest in the Discussion about the missing formal explanation, and the gap seems fixable within the scope of a revision: additional multi-seed simulations, a convergence check in \\tilde D, and/or a nonhomogeneous steady-state continuation study would strengthen the single-patch claim considerably. I would also encourage the editor to request a data/code availability statement. My recommendation of major_revision reflects that the central claim is defensible but currently under-supported, not that the work is fundamentally flawed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper is worth a serious referee, but the abstract's main claim runs ahead of the evidence. The genuinely new thing is the model: reversible Smoluchowski oligomerization on the membrane, coupled to a bulk-diffusing ligand, with recruitment driven by the largest oligomer. That combination is not in the cited literature. The authors also give a clean sufficient condition for stability in the N=2 case (Theorem 3.1): if the recruitment rate kb is small enough, no diffusion-driven instability occurs. That is rigorous and useful.\n\nWhere the paper runs into trouble is the threshold-to-single-patch claim. Linear stability analysis can only tell you the homogeneous state is unstable for certain eigenmodes; it does not tell you which non-homogeneous state is approached. The evidence that the system settles into a single-patch state is numerical: four parameter points for N=2, one for N=3, each with a single small random perturbation. The authors themselves state in the Discussion that they lack a formal explanation for the emergence and robustness of the single-patch steady-state. So the abstract's phrase \"our main result is a threshold phenomenon\" is too strong. What is actually established is that instability occurs and, in the few simulations shown, one patch forms.\n\nThere are smaller soft spots. The feedback term is taken to be exactly linear in u and aN; the authors acknowledge this assumes no saturation or steric effects. That is load-bearing: if recruitment saturates, the 'rich get richer' resource competition that yields a single patch may not hold. They also avoid solving the reduced nonlocal system, instead simulating the full bulk-surface system with Du=10^8, which is a reasonable approximation but adds a layer of uncertainty. The existence theory for the model is not proven; Section 3.4 points to a related result but says the growth conditions are too restrictive. And there are a few typos, e.g. Eq. (2.4) has D2∆a1 where it should likely be D2∆a2.\n\nNone of this kills the paper. The model is a good starting point, the N=2 theorem is correct, and the numerics are suggestive. The honest statement of limitations in the Discussion works in the authors' favor. But the framing needs to change: the 'main result' should be the instability and the numerical observation of single-patch behavior, not a claimed threshold theorem.\n\nWho is this for? People working on bulk-surface reaction-diffusion, cell polarity, or protein aggregation will find it a useful reference and a springboard for more rigorous work. I would send it to peer review with the expectation that the authors tone down the abstract and, ideally, add more simulations (multiple seeds, parameter variations) to support the single-patch claim. The reference list is appropriate.\n\nMy recommendation: accept peer review, conditional on revision. The core is sound; the overclaim is the main issue.","headline":"Solid modeling contribution with a clean N=2 theorem, but the single-patch threshold claim is numerical and overstated in the abstract.","tokens_in":25861,"tokens_out":4146,"would_cite":false,"duration_ms":40398,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B36","35K57","92C15","92C37","35B35"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that a sufficiently high rate of ligand recruitment by the largest membrane oligomers destabilizes uniform protein distributions and drives a single-patch steady state on the cell surface.","keywords":["plasma membrane","membrane-protein clustering","bulk-surface reaction-diffusion","diffusion-driven instability","positive feedback","single-patch steady state","mass conservation","oligomerization"],"falsifier":"Track the membrane distribution of monomers and the largest oligomer while increasing the recruitment rate $k_b$, for example by raising cytosolic ligand or stabilizing the largest oligomers; the model predicts a transition from a uniform surface to a single dominant patch, so observing persistent uniformity or multiple stable patches at high recruitment would contradict it. A model-side check is to compute the dispersion relation $h(l)$ for $N=2$ at parameters above the bound in Theorem 3.1 and show that $h(l)\\le 0$ for every eigenmode.","tokens_in":24900,"feed_emoji":"🧬","tokens_out":11656,"duration_ms":111108,"temperature":0.7,"pith_summary":"This paper builds a bulk-surface reaction-diffusion model in which a ligand diffuses through the cytosol, binds to the plasma membrane as a monomer, oligomerizes through mass-action steps, and is recruited back by the largest oligomer in a positive feedback loop. Its central claim is a threshold phenomenon: when the recruitment rate is low, the uniform surface distribution is stable, but when it is high enough the uniform state loses stability and the system converges to a single-patch, spatially heterogeneous steady state on the membrane. This matters because membrane-protein clustering is implicated in adhesion, signaling, and amyloid aggregation, and the model shows that clustering can arise without pre-patterns or cooperative binding, from diffusion, oligomerization, and feedback alone. For $N=2$, the low-recruitment stability statement is proved as a theorem, while the high-recruitment instability and the $N=3$ behavior are demonstrated numerically.","feed_headline":"Membrane proteins form one cluster past a recruitment threshold","feed_subtitle":"The result ties a measurable recruitment rate to spatial clustering on the cell surface.","key_machinery":"The carrying mechanism is the recruitment flux $f(u,a_1,a_N)=(k_0+k_b a_N)u-k_d a_1$, a linear positive feedback that makes the largest oligomer $a_N$ act as a catalyst for new monomer entry. The analytical workhorse is the reduced surface-only system obtained by taking the cytosolic diffusion $D_u\\to\\infty$, in which the bulk concentration becomes the nonlocal functional $\\mathcal{U}[a_1,\\ldots,a_N](t)=|\\Omega|^{-1}\\left(M_0-\\sum_{j=1}^N j\\int_\\Gamma a_j\\,ds\\right)$; this enforces total mass conservation and lets the linearized problem be diagonalized into eigenmodes of the Laplace-Beltrami operator, turning pattern formation into a dispersion-relation calculation.","core_discovery":"The authors study a closed cell in which a bulk species $u$ diffuses and exchanges with membrane species $a_1,\\dots,a_N$ through the flux $f(u,a_1,a_N)=(k_0+k_b a_N)u-k_d a_1$, so the largest oligomer feeds back on its own production. After reducing the system to the membrane using a nonlocal mass-conservation constraint, they linearize homogeneous steady states against eigenfunctions of the Laplace-Beltrami operator and analyze the dispersion relation $h(l)$. The steady-state polynomial has a coefficient $k_0|\\Gamma|N-M_0 k_b$ that determines whether the homogeneous state is unique or bistable. The main result is a threshold in the recruitment rate $k_b$: for $N=2$, Theorem 3.1 gives an explicit small-$k_b$ bound under which the unique steady state is stable against both homogeneous and non-homogeneous perturbations, while numerical simulations show that larger $k_b$ turns the uniform state unstable and produces a single high-concentration patch, with analogous behavior for $N=3$.","pith_inferences":["If the linear recruitment term were replaced by a saturating response, the single-patch regime would likely persist but the threshold would shift and might gain an upper bound; this extension is left open in the paper.","The single-patch outcome resembles winner-takes-all competition for a conserved molecular pool, suggesting that cells could toggle between uniform and patchy states by tuning recruitment rather than changing membrane geometry.","An experimental system that raises the effective recruitment rate, for example by stabilizing the largest oligomers or raising cytosolic ligand, should be pushed from a uniform membrane into one dominant aggregate; this is a testable prediction the paper does not itself demonstrate."],"forward_implications":["When the recruitment rate is small enough, or zero, the uniform state is stable and no spatial pattern forms.","Above the threshold, the model produces a single dominant patch rather than multiple coexisting clusters, with larger oligomers concentrated in a tighter and higher peak than monomers.","The lowest non-trivial membrane eigenmode dominates the instability, and higher modes have shrinking unstable regions, which is consistent with the single-patch selection.","The dimensional area of the patch grows roughly linearly with cell radius while its percentage of the membrane area falls like $1/R$, and larger $N$ gives larger patches.","Pattern formation does not require a Hill-type cooperative term: mass-action oligomerization, linear feedback, and mass conservation are enough."],"supporting_citations":[{"why":"Supplies the bulk-to-surface reduction when cytosolic diffusion is fast and the diffusion-driven linear stability method the analysis is built on.","marker":"[33, 34]"},{"why":"Defines the diffusion-driven-instability concept that justifies reading a positive dispersion relation as pattern formation.","marker":"[46]"},{"why":"Provides the mass-action kinetics of reversible aggregation used to write the oligomerization reactions.","marker":"[42]"},{"why":"Gives the earlier single-patch steady-state result in membrane signalling that the paper compares its pattern to.","marker":"[61]"},{"why":"Supports the biologically observed separation of time scales between cytosolic and membrane diffusion used in the reduction.","marker":"[45]"}],"fun_headline_variants":["Single cluster forms after recruitment threshold","Recruitment above critical rate yields one membrane patch","Membrane protein clustering goes single-patch past threshold","One patch wins when recruitment exceeds critical value","Recruitment threshold decides uniform vs. single-cluster state"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The threshold result rests on the assumption that the biggest cluster can pull in new building blocks in exact proportion to its own concentration, with no saturation or crowding limit; if real recruitment saturates, the single-patch outcome is not assured.","fun_headline_variants_meta":{"raw":{"variants":["Single cluster forms after recruitment threshold","Recruitment above critical rate yields one membrane patch","Membrane protein clustering goes single-patch past threshold","One patch wins when recruitment exceeds critical value","Recruitment threshold decides uniform vs. single-cluster state"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000655,"raw_usage":{"total_tokens":3027,"prompt_tokens":998,"completion_tokens":2029,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":614,"completion_tokens_details":{"reasoning_tokens":1959}},"tokens_in":614,"tokens_out":2029,"duration_ms":16242,"temperature":1.0,"reasoning_tokens":1959,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:19:59.742747+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Track the membrane distribution of monomers and the largest oligomer while increasing the recruitment rate $k_b$, for example by raising cytosolic ligand or stabilizing the largest oligomers; the model predicts a transition from a uniform surface to a single dominant patch, so observing persistent uniformity or multiple stable patches at high recruitment would contradict it. A model-side check is to compute the dispersion relation $h(l)$ for $N=2$ at parameters above the bound in Theorem 3.1 and show that $h(l)\\le 0$ for every eigenmode.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the diffusion-driven-instability concept that justifies reading a positive dispersion relation as pattern formation."},{"cited_title":"Bentz and S","cited_arxiv_id":null,"evidence_quote":"Provides the mass-action kinetics of reversible aggregation used to write the oligomerization reactions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the earlier single-patch steady-state result in membrane signalling that the paper compares its pattern to."},{"cited_title":"Postma, L","cited_arxiv_id":null,"evidence_quote":"Supports the biologically observed separation of time scales between cytosolic and membrane diffusion used in the reduction."}],"review_version":1}