{"id":"0b7ac7c1-5860-4fd0-864a-a61e1c718da5","arxiv_id":"1908.02055","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A lattice model with length-dependent vesicle transport predicts that an initial neurite length advantage amplifies into higher growth-cone vesicle pools, consistent with live imaging showing more anterograde vesicle movement in growing neurites.","lead":"Using live imaging of rat neurons and a lattice model of vesicle transport, the authors find that a neurite which is already longer receives more vesicle traffic, which they propose reinforces its growth and helps it become the axon. The paper combines experiment and simulation to argue that vesicle transport itself may drive neuron polarization, not just local growth machinery.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The reported asymmetry is computed with a non-conservative boundary scheme: Eqs. (2)/(8) omit the pool occupancy factors used in the pool ODEs (3)-(4), so the longer-neurite advantage in Λ_N may be a mass-leak artifact.","rationale":"Both the reader and I see the same paper, but the reader's weakest_assumption focuses on the linear potential scaling. I think the more decisive issue is the discrete scheme's mass balance. If the code implements (8) literally, the simulation is not of the model whose ODEs are written, and the asymmetry in Figure 8 cannot be attributed to vesicle transport. If the code instead implements the intended conservative fluxes, then the printed equations are wrong and the paper must at least be corrected and re-run. Either way, the central claim is not currently supported by the numerical evidence. The potential-scaling concern is real but secondary: it is a modeling choice that could be probed by sensitivity analysis, whereas the non-conservation is an internal mathematical inconsistency in the evidence itself. The proof gap in Lemma 5.1 and the 'purely estimated' diffusion coefficient are also secondary; they affect robustness but do not by themselves invalidate the qualitative result. I would keep the reader's CONDITIONAL verdict: the issues are concrete and fixable, and the model may still produce the reported asymmetry once the boundary fluxes are corrected. I do not see grounds to move to ACCEPT or REJECT without re-running the corrected scheme.","tokens_in":17496,"tokens_out":10683,"duration_ms":115011,"concrete_test":"Re-run Experiment 5.2 (Ω1=[0,1], Ω2=[0,0.3], Table 1 parameters, 400 grid points per domain, τ=10^-5) with the boundary fluxes corrected to match (3)-(4): ∂t a_N = ... − Ch β_a a_N(1−Λ_N/Λmax_N); ∂t r_N = ... + Ch α_r(1−ρ_N) Λ_N/Λmax_N; ∂t a_1 = ... + Ch α_a(1−ρ_1) Λ_som/Λmax_som; ∂t r_1 = ... − Ch β_r r_1(1−Λ_som/Λmax_som). Monitor total mass at every step. If the corrected scheme no longer yields Λ_N,1 > Λ_N,2 at T=100, or if the mass drift in the printed scheme exceeds the observed pool asymmetry, the central claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the two-neurite simulations in Section 5. Those simulations use boundary equations (8), whose fluxes do not match the pool ODEs (3)-(4) from which mass conservation is claimed. In (2)/(8), the anterograde influx into cell 1 is written +Ch α_a(1−ρ_1), without the factor Λ_som/Λmax_som that appears in the soma ODE (4); the anterograde outflux at the tip is −Ch β_a a_N, without (1−Λ_N/Λmax_N) from (3); and the retrograde influx at the tip is +Ch α_r(1−ρ_N), without Λ_N/Λmax_N. Equivalently, the rates αq(Λj), βq(Λj) defined just above (2) are not the rates used in (2)/(8). The numerical implementation is therefore not the model described in the text, and total mass is not conserved. Since both neurites are discretized with 400 points (Section 5), the long domain has larger cell size h; every missing occupancy factor multiplies a boundary term proportional to Ch, so the spurious sources and sinks are about 3.3 times stronger in the longer neurite. Thus the reported Λ_N,1 > Λ_N,2 in Figure 8 could be produced by the mass leak rather than by the intended vesicle-transport feedback. The authors' statement that they multiply the in- and outflux terms by h only rescales units; it does not restore the omitted occupancy factors.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper combines live-imaging experiments on Vamp2-GFP vesicle transport in cultured hippocampal neurons with a two-species lattice model of anterograde/retrograde vesicle transport along neurites, coupled to a soma pool and growth-cone pools. The central modelling claim is that an initial difference in neurite length creates a self-reinforcing asymmetry in growth-cone vesicle pools, so that the longer neurite acquires a higher growth potential; this is presented as a possible mechanism for axon specification. The authors also derive a formal cross-diffusion PDE limit and state an existence/regularity caveat. The experiments show a significant increase in anterograde vesicle transport during neurite extension.","tokens_in":17815,"tokens_out":8950,"duration_ms":91103,"significance":"If the central claim survives scrutiny, the model would provide a mechanism linking vesicle transport to axon selection and makes a falsifiable prediction about length thresholds. The experimental data are carefully quantified and the modelling framework is novel in coupling size-exclusion transport to finite-capacity pools. However, the numerical support is currently undermined by a boundary-flux inconsistency and by structural choices that favour the longer neurite; the qualitative conclusion should be re-established with a mass-conservative scheme and neutral parameter normalization.","major_comments":[{"comment":"The discrete boundary fluxes are inconsistent with the pool ODEs and with the PDE boundary conditions (11). Specifically, the anterograde influx into cell 1 in (2) is written as +Ch α_a(1−ρ_1) without the factor Λ_som/Λmax_som that appears in (4); the anterograde outflux at cell N is −Ch β_a a_N without (1−Λ_N/Λmax_N) from (3); and analogous factors are missing in the r_1 and r_N equations. The same omissions appear in the scaled scheme (8) and in the boundary update (14). Because both domains are discretized with 400 points, the missing factors multiply terms proportional to Ch, so the longer neurite has proportionally larger spurious sources and sinks; the mass imbalance therefore contaminates the comparison in Section 5.2 and Figure 8. The statement in Section 5 that multiplying in- and outflux terms by h ensures mass conservation does not address this inconsistency.","section":"Section 3, Eqs. (2)-(4); Section 3.1, Eq. (8); Section 5, Eq. (14)"},{"comment":"The potentials V_a(x)=1.75x and V_r(x)=−1.5x give the longer neurite a larger total potential drop (1.75 for Ω1=[0,1] versus 0.525 for Ω2=[0,0.3]) and, with the stated per-length maximal density, a larger total vesicle capacity; with uniform initial concentration 0.1, the longer neurite also contains more initial vesicles. The reported symmetry breaking may therefore reflect an initial structural advantage rather than a dynamic feedback. The authors should either normalize the potentials per neurite length, hold total capacity and initial vesicle number equal, or justify the chosen scaling from the biology.","section":"Section 5.1, Table 1"},{"comment":"The proof of the upper bound ρ≤1 is circular: inequality (17) states (1−ρ^{k+1}_i) ≥ (1−4τHV_max)(1−ρ^{k+1}_i), which is a tautology and gives no bound in terms of the previous time step. The CFL condition may still be correct, but the lemma as written is not proven; since the simulations rely on this scheme, the box constraints should be verified numerically or the proof corrected.","section":"Section 5, Lemma 5.1, Eq. (17)"}],"minor_comments":[{"comment":"The heading \"Exterimental Methods\" contains a typo; it should read \"Experimental Methods\".","section":"Section 2.2"},{"comment":"The bullet text says anterograde vesicles enter with rate α_a(Λ_som)(1−ρ_N), but equation (2) and the intended coupling require (1−ρ_1); similarly, retrograde vesicles should enter from the pool with rate α_r(Λ_N)(1−ρ_N), not (1−ρ_1).","section":"Section 3, bullet list"},{"comment":"The Ω2 row lists [0,3] while the text and simulations use [0,0.3] and [0,0.9]; also the β entries repeat β_a,2, which appears to be a typographical error.","section":"Table 1"},{"comment":"The text refers to a rapid change \"at the tip of the longer neurite (Λ_N,2)\", but in the simulations neurite 1 is the longer domain; the labels for Λ_N,1 and Λ_N,2 appear to be swapped in this discussion.","section":"Section 5.3, Figure 9"},{"comment":"In the r_1 equation, the notation \"Vr,x1\" should be \"Vr,1\" for consistency with the other potential terms.","section":"Section 3, Eq. (2)"}],"recommendation":"major_revision","confidential_remarks":"The boundary-flux inconsistency in (2)/(8) is a serious numerical-model mismatch; before inviting a revision, the editor may wish to ask the authors for corrected simulations showing whether the qualitative asymmetry survives mass-conservative boundary terms. The experimental section is otherwise solid."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague — this one is worth a look, but it's not as solid as it presents. The new thing here is simple: a two-species exclusion process on a lattice, with both neurites fed from the same soma pool, and a numerical observation that a longer neurite ends up with a higher growth-cone pool concentration. That's a plausible mechanism for axon selection, and the live-imaging data (Vamp2-GFP vesicle counts during extension vs. retraction) give at least some empirical motivation. The paper is clearly written and doesn't oversell the biology too much.\n\nBut the central numerical result rests on an inconsistency in the discrete equations. The boundary conditions in (2) and (8) write the fluxes as, e.g., +Ch α_a(1-ρ1) and -Ch β_a a_N, without the occupancy factors Λ_som/Λ_max_som and 1-Λ_N/Λ_max_N that appear in the pool ODEs (3)-(4) and in the macroscopic boundary conditions (11). The rates α_q(Λ), β_q(Λ) are defined just above (2) but then not used. So as written, the lattice and the pools don't exchange the same amount of mass. The code likely does exactly this, since the numerical section says they multiply by h but never mentions restoring the omitted factors. Mass is not conserved, and the leak is not neutral: the missing factors are order-one (e.g., Λ_som/Λ_max_som ≈ 0.7), and because the two domains are discretized with the same number of points, the boundary structure differs between them. It's entirely plausible that the asymmetry in Figure 8 is an artifact of this leak. That's a load-bearing flaw, because the paper's main claim is that this asymmetry is a vesicle-transport feedback. I'd want to see the corrected scheme, a mass-conservation test, and a sensitivity analysis before believing it.\n\nOther concerns: the potentials V_a=1.75x and V_r=-1.5x are the same slope on both neurites, so the longer neurite gets a bigger total potential drop and can hold more vesicles — that structural advantage may be doing the work. The diffusion coefficient ε=0.05 is admitted to be 'purely estimated,' and there's no sensitivity analysis. Lemma 5.1's proof uses an inequality that's tautological as written. The experiments are descriptive and don't test the pool-concentration prediction.\n\nNone of these secondary issues kill the paper on their own. The core idea — length-dependent symmetry breaking through shared-pool transport — is interesting and testable. But the mass-nonconservation issue has to be fixed before the simulation result means anything.\n\nMy recommendation: send it to peer review, yes. It's an important problem and the flaw is technical rather than fatal to the concept, but referees should ask for a corrected, mass-conserving implementation and proper sensitivity checks. I wouldn't cite it in its current form.","headline":"A simple two-neurite vesicle-transport model that predicts length-dependent axon selection, but the discrete boundary conditions don't conserve mass and may produce the central asymmetry as an artifact.","tokens_in":18431,"tokens_out":8789,"would_cite":false,"duration_ms":88537,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["92C37","35Q92"],"pacs":[],"model":"deepseek-v4-flash","headline":"Vesicle transport alone, the model predicts, turns a length lead into a growth-cone advantage.","keywords":["neurite growth","vesicle transport","neuronal polarization","symmetry breaking","lattice-based model","asymmetric exclusion process","cross-diffusion","growth cone vesicle pool"],"falsifier":"Run the same two-neurite simulation with the anterograde and retrograde potentials scaled by each neurite's own length, for example $V_a(x)=1.75x/L$, and with equal pool capacities per neurite; if the longer neurite no longer maintains a higher growth-cone pool concentration, the predicted length advantage is an artifact of giving longer domains larger total potential drops rather than a property of the transport dynamics.","tokens_in":17188,"feed_emoji":"🧠","tokens_out":8015,"duration_ms":82219,"temperature":0.7,"pith_summary":"The paper combines live imaging of Vamp2-labelled vesicles in unpolarized hippocampal neurons with a lattice model of bidirectional vesicle transport to ask whether transport alone can explain why one neurite becomes the axon. In the model, two neurites share a finite soma pool and exchange anterograde and retrograde vesicles with growth-cone pools at their tips. Live imaging found 37% more anterograde than retrograde vesicle movements in extending neurites, motivating a transport-based account. When the neurites have very different lengths, the longer one accumulates a higher vesicle concentration in its growth-cone pool, while nearly equal lengths show almost no asymmetry. The authors read the growth-cone pool as growth potential: more vesicles at the tip means more membrane material for extension, so an initial length advantage becomes self-reinforcing.","feed_headline":"Longer neurites win the growth-cone vesicle race","feed_subtitle":"In an otherwise symmetric simulation, only a big length gap breaks symmetry and feeds the longer neurite.","key_machinery":"The load-bearing object is a discrete lattice model of two vesicle species with size exclusion. Each neurite is a one-dimensional lattice of cells connected to a soma pool at one end and a growth-cone pool at the other; anterograde vesicles move toward the tip under potential $V_a(x)=1.75x$ and retrograde vesicles move back under $V_r(x)=-1.5x$, with jumps allowed only into cells that are not full. The rate of leaving the soma or a pool is proportional to occupancy, and the rate of entering is proportional to free space, so the shared finite soma couples the two neurites and makes their transport competitive. A formal continuum limit gives a cross-diffusion system for anterograde and retrograde densities, with boundary fluxes carrying exactly the soma and pool exchange terms. The mechanism carrying the argument is that a longer neurite presents a larger total potential drop and more vesicle capacity, pulling more anterograde vesicles out of the soma and leaving fewer for the shorter neurite.","core_discovery":"The central claim is that bidirectional vesicle transport through a common soma pool converts an initial difference in neurite length into a difference in growth potential, favoring the longer neurite. In simulations with otherwise symmetric initial data, the growth-cone pool concentration $\\Lambda_{N,1}$ rises above $\\Lambda_{N,2}$ when the domains are $\\Omega_1=[0,1]$ and $\\Omega_2=[0,0.3]$, but the asymmetry is nearly absent when $\\Omega_2=[0,0.9]$. The authors interpret the growth-cone vesicle concentration as the neurite's growth potential because vesicles fused at the membrane supply the surface area for extension. They also report a sudden drop in the shorter neurite's pool concentration after a metastable plateau between $t=26$ and $t=36$ in the strongly asymmetric run, and they compare this threshold-like event to behavior seen in polarizing neurons. Their conclusion is that length-dependent vesicle supply, mediated by bidirectional transport, can serve as a mechanical explanation for axon selection without requiring a separate molecular length sensor.","pith_inferences":["Editorial inference: the model's symmetry breaking is driven in part by domain size, since the longer neurite has a larger total potential drop and more vesicle capacity; running the same two-neurite computation with potentials and pool capacities normalized per unit length would separate genuine dynamical competition from a static bias.","Editorial inference: the same soma-pool coupling should apply to other bidirectional cargo systems sharing a resource pool, such as mitochondrial or mRNA distribution in dendrites, so the mechanism may be a general principle for competitive cellular transport.","Editorial inference: because the growth-cone pool is treated as a well-mixed compartment, adding explicit exocytosis and endocytosis delays or stochastic vesicle numbers could reveal whether the metastable plateau and sudden drop survive noise; this is testable by stochastic simulation."],"forward_implications":["If the model is right, a neurite that has crossed a critical length will automatically receive a larger share of soma-derived vesicles, providing a transport-level explanation for the experimental finding that a neurite must exceed a minimal length to become an axon.","Nearly equal initial lengths should not break symmetry by themselves; the simulations with $\\Omega_2=[0,0.9]$ show almost no growth-cone asymmetry, implying a threshold length gap is required for reliable axon specification.","Oscillations in growth-cone vesicle pools should appear during polarization, matching observed cycles of neurite extension and retraction and giving a dynamical signature that live imaging could look for.","The coupling through the soma means interventions that change soma capacity, retrograde return, or pool size should shift which neurite wins, making transport rates a control point for axon selection."],"supporting_citations":[{"why":"Supplies the experimental observation that a neurite must exceed a minimal length to become an axon, the phenomenon the model's length-driven symmetry breaking is built to explain.","marker":"[11]"},{"why":"Documents critical-length behavior in developing hippocampal neurons, anchoring the model's threshold interpretation.","marker":"[15]"},{"why":"Reports vectorial cytoplasmic flow toward the future axon before axon formation, the biological precedent for preferential vesicle transport.","marker":"[3]"},{"why":"Reviews how intracellular transport is polarized toward the nascent axon and frames the question the model addresses.","marker":"[32]"},{"why":"Proposes an alternative diffusion-based neurite length-sensing mechanism, giving the comparison that motivates a transport-based explanation.","marker":"[38]"},{"why":"Provides the asymmetric simple exclusion process basis for the lattice dynamics with size exclusion.","marker":"[10]"},{"why":"Supplies the formal derivation of the macroscopic cross-diffusion equations from the discrete lattice model.","marker":"[34]"},{"why":"Establishes that membrane expansion during axon growth requires vesicle insertion, justifying the growth-cone pool as a measure of growth potential.","marker":"[27]"}],"fun_headline_variants":["Vesicle transport turns length edge into growth edge","Bidirectional vesicle flow selects the longer neurite","Longer neurite's vesicle supply fuels its own growth","Symmetry breaks when vesicle transport favors longer neurite","Length gap tips vesicle race toward longer neurite"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model's result assumes that the transport potential and vesicle capacity are proportional to neurite length, so the longer neurite starts with a larger share of the shared vesicle supply; if those quantities were instead held equal per neurite, the predicted advantage of the longer neurite could vanish.","fun_headline_variants_meta":{"raw":{"variants":["Vesicle transport turns length edge into growth edge","Bidirectional vesicle flow selects the longer neurite","Longer neurite's vesicle supply fuels its own growth","Symmetry breaks when vesicle transport favors longer neurite","Length gap tips vesicle race toward longer neurite"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000882,"raw_usage":{"total_tokens":3751,"prompt_tokens":829,"completion_tokens":2922,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":445,"completion_tokens_details":{"reasoning_tokens":2853}},"tokens_in":445,"tokens_out":2922,"duration_ms":21798,"temperature":1.0,"reasoning_tokens":2853,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:55:18.157213+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same two-neurite simulation with the anterograde and retrograde potentials scaled by each neurite's own length, for example $V_a(x)=1.75x/L$, and with equal pool capacities per neurite; if the longer neurite no longer maintains a higher growth-cone pool concentration, the predicted length advantage is an artifact of giving longer domains larger total potential drops rather than a property of the transport dynamics.","supporting_citations":[{"cited_title":"Dotti and Gary A","cited_arxiv_id":null,"evidence_quote":"Supplies the experimental observation that a neurite must exceed a minimal length to become an axon, the phenomenon the model's length-driven symmetry breaking is built to explain."},{"cited_title":"Goslin and G","cited_arxiv_id":null,"evidence_quote":"Documents critical-length behavior in developing hippocampal neurons, anchoring the model's threshold interpretation."},{"cited_title":"Bradke and C","cited_arxiv_id":null,"evidence_quote":"Reports vectorial cytoplasmic flow toward the future axon before axon formation, the biological precedent for preferential vesicle transport."},{"cited_title":"Neuronal polarization: From spatiotemporal signaling to cytoskeletal dynamics","cited_arxiv_id":null,"evidence_quote":"Reviews how intracellular transport is polarized toward the nascent axon and frames the question the model addresses."},{"cited_title":"A diffusion-based neurite length-sensing mechanism involved in neuronal symmetry breaking","cited_arxiv_id":null,"evidence_quote":"Proposes an alternative diffusion-based neurite length-sensing mechanism, giving the comparison that motivates a transport-based explanation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the asymmetric simple exclusion process basis for the lattice dynamics with size exclusion."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the formal derivation of the macroscopic cross-diffusion equations from the discrete lattice model."},{"cited_title":"Pfenninger","cited_arxiv_id":null,"evidence_quote":"Establishes that membrane expansion during axon growth requires vesicle insertion, justifying the growth-cone pool as a measure of growth potential."}],"review_version":1}