{"id":"a2bde1df-f8b1-417d-8b55-6f184d4eae54","arxiv_id":"2507.08258","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":10,"one_line_summary":"A spatial network simulation reproduces calcium spark/quark switching in RyR clusters and shows calsequestrin prolongs release through buffering but creates excess refractoriness through RyR binding.","lead":"This study models calcium sparks in heart cells with a spatial network of ryanodine receptor channels arranged in realistic clusters, alongside the regulatory protein calsequestrin. It shows the clusters behave as switches that release either tiny quarks or full sparks, and that calsequestrin lengthens sparks by buffering calcium while its direct binding to receptors causes a refractory period.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Clamped NSR makes JSR refill artificially fast, so the attribution of excess refractoriness to CSQ binding rather than slow store refilling is not yet established.","rationale":"The central argument rests on reproducing sparks and on the separable roles of CSQ. The paper's strongest evidence is that removing the CSQ binding state removes excess refractoriness, but this removal is performed in a system where the JSR is force-refilled by a clamped reservoir. The same manipulation would also appear to remove refractoriness if the true mechanism were store depletion, because the clamped refill makes depletion short-lived. Therefore the key attribution is underdetermined by the current simulations. The authors are transparent about the simplification, and the rest of the model is plausible and grounded in experimental single-channel data. No internal contradiction or circular fitting was found. The reader's weakest assumption identified the same clamped-calcium limitation; I agree. A single closed-loop simulation would settle whether the claimed CSQ-dependent refractoriness survives realistic calcium cycling. Until then, the correct verdict remains CONDITIONAL, and the reader's verdict should be unchanged.","tokens_in":24921,"tokens_out":4158,"duration_ms":52331,"concrete_test":"Rerun the two-spark protocol of Fig. 13 with a finite NSR compartment: replace Eq. (13) by J_refill = ([Ca2+]NSR - [Ca2+]JSR)/τ_refill with a dynamic [Ca2+]NSR that is depleted by all RyR fluxes in the simulated region and replenished by a SERCA term with a realistic time constant (e.g., Shannon et al. parameters). Compare the recovery curve, i.e., the fraction of first-spark Npeak, and its timescale against both the JSR depletion/refilling curve and the CSQ unbinding time τu = 125 ms. If the recovery tracks the refilling curve or shifts with the SERCA rate, the excess-refractoriness claim in Section IV.C collapses; if it remains CSQ-limited, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section II clamps [Ca2+]NSR at 1000 µM, and Eq. (13) refills the JSR with τ_refill = 5 ms toward this fixed reservoir. Under this clamp, the JSR always recovers within tens of milliseconds after a spark, which is exactly what Fig. 13(j) compares against when arguing that the refractory period is not caused by local JSR refilling. In a real myocyte the NSR is finite, is shared by many release units, and is refilled by SERCA on a slower beat-to-beat timescale; after a large spark the local store may remain partially depleted. Thus the observed excess refractoriness and its attribution to CSQ-bound RyR states, rather than to depleted stores, could be a consequence of the clamped boundary condition. The paper acknowledges this in Section V.C, calling the calcium-clamped setup a simplified system and requesting future work under cycling conditions, but the claim that CSQ-RyR binding controls the refractory period is central to the abstract and Section IV.C. Without a closed-loop calcium-cycling test, the separation between CSQ-dependent refractoriness and store-refilling-dependent refractoriness is not established. This is not an internal inconsistency; it is a load-bearing assumption that remains untested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a stochastic spatial network model of ryanodine receptors (RyRs) on the junctional sarcoplasmic reticulum, with each RyR occupying one of four states (CSQ-unbound/open/closed and CSQ-bound/open/closed). Calcium dynamics in the subspace and JSR are described by diffusion-like coupling between RyRs, efflux to a clamped myoplasm, refilling from a clamped NSR reservoir, and rapid buffering including calsequestrin (CSQ). Through large ensembles of randomly generated cluster geometries, the authors simulate spontaneous and LCC-evoked calcium release events. They report a bimodal distribution of short-lived calcium quarks and longer-lasting calcium sparks, interpret RyR clusters as on-off switches, and study how spark duration and refractoriness depend on RyR transition rates, clamped calcium concentrations, CSQ concentration, and CSQ malfunction. The central conclusions are that CSQ's buffering action prolongs spark duration, while CSQ-RyR binding produces excess refractoriness beyond that explained by local JSR refilling.","tokens_in":25150,"tokens_out":3897,"duration_ms":50591,"significance":"If the claims hold, the model offers a computationally tractable way to incorporate experimentally observed irregular RyR cluster geometry into spark simulations, and it separates two putative roles of CSQ (buffering vs. RyR modulation) in a clean, mechanistic way. The paper is commendable for specifying all model equations and parameters, for using large sample sizes (thousands of cluster geometries and triggers in Section III), and for framing the central predictions in a falsifiable manner. The qualitative agreement with experiments on spark duration trends (Terentyev et al.) and on refractoriness (Brochet et al.) is a useful benchmark. The main limitation is that the clamped-NSR boundary condition may not allow the authors to distinguish CSQ-induced refractoriness from store-depletion refractoriness in a physiologically realistic setting.","major_comments":[{"comment":"The claim that excess refractoriness is caused by CSQ-RyR binding rather than by slow JSR refilling is not established under the model's clamped-NSR condition. With [Ca2+]NSR fixed at 1000 µM and τ_refill = 5 ms, the JSR in Eq. (13) refills very rapidly after a spark, so the comparison curve [Ca2+]SS/100 µM in Fig. 13(j) necessarily shows fast recovery. In a real myocyte the NSR is finite, shared among many release units, and refilled by SERCA on a slower timescale, so local store depletion may persist much longer. The paper acknowledges in §V.C that the clamped setup is a simplified system, but the abstract and §IV.C state the CSQ-RyR interaction controls the refractory period as a general conclusion. A closed-loop calcium-cycling test (or at least a finite NSR compartment) is needed to separate CSQ-dependent refractoriness from store-refilling-dependent refractoriness; otherwise the attribution remains an untested load-bearing assumption.","section":"§IV.C, Eq. (13), Fig. 13"},{"comment":"The interpretation of RyR clusters as on-off switches relies on a clear bimodal separation between quarks and sparks, but the evidence for bimodality in Fig. 6 is partly asserted rather than quantified. The horizontal threshold lines are placed at NRyR/4 (or at Npeak=10 for multicluster cases), and the 2D histograms in Fig. 6(a-i) show broad continua for some cluster sizes, especially the 30-RyR and multicluster panels. Please report a formal separation measure (e.g., a two-component fit or a gap statistic between the quark and spark populations) and show that the bimodality is not an artifact of the chosen threshold. Without this, the central 'on-off switch' claim is less strongly supported than the text suggests.","section":"§III.C, Fig. 6"},{"comment":"The statement that the fast buffering approximation is acceptable 'because the calcium concentration is clamped, making mass conservation irrelevant' is imprecise and could affect the CSQ-related results. Only [Ca2+]myo and [Ca2+]NSR are clamped; [Ca2+]JSR and [Ca2+]SS are dynamic. The rapid buffering approximation in Eq. (15) treats CSQ as an instantaneous buffer without tracking total buffer-bound calcium, which may bias the time course of [Ca2+]JSR depletion and recovery that underlies the spark-duration and refractoriness analyses. Please clarify why neglecting buffer mass conservation does not distort the CSQ buffering effects reported in Figs. 12 and 13, or test the sensitivity of these conclusions to an explicit finite-buffer formulation.","section":"§II.C, Eq. (1), Eq. (15)"}],"minor_comments":[{"comment":"The section heading 'REGULATION OF CLACIUM SP ARKS' contains typos ('CLACIUM', 'SP ARKS'); the same heading style appears with 'sprak' in the text. These should be corrected.","section":"Section IV title and text"},{"comment":"The caption begins 'Pannels (a), (b), and (c)' with a typo; it should read 'Panels'.","section":"Fig. 7 caption"},{"comment":"The caption says 'The line in (c) is restuls by Sato et al.'; 'restuls' should be 'results'.","section":"Fig. 11 caption"},{"comment":"The notation in Eq. (15) is unclear: the term 'KCBCSQn' mixes subscripts and variables. Please write it with explicit multiplication and clarify that KC is the half-saturation constant and BCSQ is the CSQ concentration.","section":"Eq. (15)"},{"comment":"Several references, e.g., Ref. 47 and Ref. 60, have inconsistent formatting ('Y . V' instead of 'Y. V', missing journal title in Ref. 60). A careful copyedit is needed.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper's central claim about CSQ-dependent refractoriness is plausible and well-aligned with prior modeling by Restrepo et al., but the clamped-NSR design makes the central attribution untestable as presented. I would encourage the editor to ask for either a closed-loop finite-NSR test or a careful reframing of the claim as conditional on clamped conditions. The bimodality quantification is also worth requesting because the 'on-off switch' language is central to the abstract."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is a computational modeling paper, not an experimental one. It builds a spatial network of RyRs using realistic irregular cluster geometries from Jayasinghe's super-resolution data, a four-state RyR model with CSQ binding from Restrepo, and sigmoid fits to Cannell's single-channel rates. The genuinely new thing is the combination: explicit cluster geometry plus CSQ states, and the systematic ablation of CSQ's buffer versus binding functions.\n\nThe paper does several things well. The bimodal quark/spark distribution is an emergent outcome of the stochastic simulation, not something fit to spark-level data. The parameter studies are extensive, using thousands of cluster geometries and triggers. The ablation results are clean: removing the CSQ buffer function eliminates the duration prolongation; removing the CSQ binding states eliminates the excess refractoriness. That is a nice, falsifiable decomposition, and the qualitative comparisons to Terentyev and Brochet are reasonable.\n\nThe main soft spot is the clamped calcium assumption. The stress-test note is right: clamping [Ca2+]NSR at 1000 µM with tau_refill = 5 ms makes the JSR refill nearly instantly, so the Fig. 13(j) comparison showing that recovery is not tied to store refilling is comparison against an artificially fast refill. In a real myocyte, the NSR is finite and shared by many release units, and after a large spark local depletion can persist on a slower timescale. Thus the claim that CSQ-RyR binding, rather than store refilling, drives the excess refractoriness is not yet established. The authors acknowledge this in Section V.C, which is honest, but they still put the CSQ-refractoriness claim in the abstract and Section IV.C. This is not fatal to the whole paper—the switch-like behavior and the CSQ buffering results probably survive—but it tempers the central mechanistic claim.\n\nA minor issue: no code or data are shipped, so exact reproduction is harder even though the equations are fully specified. Also, the network coupling parameters (r0, tau_RR) are estimated rather than swept; a sensitivity analysis there would strengthen the model.\n\nWho this is for: people working on stochastic calcium release modeling, cardiac dyad biophysics, and anyone interested in CSQ's dual role. It deserves a serious referee. The model is clearly specified and the ablation logic is sound, but the referee should push on the clamped-NSR assumption, ideally asking for a closed-loop test or a toned-down claim about refractoriness.","headline":"A competent stochastic spatial-network model of RyR clusters that cleanly separates CSQ buffering from CSQ-RyR binding, but the clamped-store setup leaves the central refractoriness claim under-supported.","tokens_in":25776,"tokens_out":1766,"would_cite":true,"duration_ms":22335,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["92C37","92C42"],"pacs":[],"model":"deepseek-v4-flash","headline":"A spatial-network model produces calcium sparks from RyR clusters acting as on-off switches.","keywords":["calcium sparks","ryanodine receptor","spatial network model","calsequestrin","calcium-induced calcium release","refractory period","stochastic simulation","cardiac myocyte"],"falsifier":"The cleanest test is a variant that keeps CSQ's calcium buffering but blocks CSQ-RyR binding: the model predicts long sparks with no excess refractory period, so observing a persistent refractory period in that variant would falsify the paper's assignment of refractoriness to the CSQ-bound state.","tokens_in":24636,"feed_emoji":"⚡","tokens_out":9572,"duration_ms":98766,"temperature":0.7,"pith_summary":"This paper argues that a stochastic spatial network of ryanodine receptors, built from experimentally measured irregular cluster geometries and a four-state model of calsequestrin regulation, can reproduce the two observed classes of calcium release: brief calcium quarks and sustained calcium sparks lasting about 20 ms. The central claim is that each RyR cluster acts as an on-off switch, so the same calcium-induced calcium release mechanism produces either a small short-lived event or a large spark depending on how many receptors are recruited. The authors further claim that spark termination is governed by the falling calcium gradient between the junctional SR and the subspace together with stochastic channel gating, and that CSQ shapes release in two separable ways: buffering lengthens the spark, while CSQ binding to the RyR complex creates the excess refractory period. If the model is right, it separates the buffering and regulatory roles of CSQ and shows how a spatially realistic receptor network can exhibit switch-like calcium signaling without added inactivation mechanisms.","feed_headline":"Model: RyR clusters are on-off switches for calcium sparks","feed_subtitle":"Spatial-network simulations reproduce 20 ms sparks and split CSQ's roles into buffering and receptor regulation.","key_machinery":"The machinery is a spatial network of RyR nodes whose pairwise diffusive coupling is an exponential adjacency matrix $A_{ij}=e^{-r_{ij}/r_0}/\\tau_{RR}$ with $r_0=60$ nm and $\\tau_{RR}\\approx 0.01$ ms estimated from the calcium diffusion constant; cluster sizes follow an experimentally derived power-law distribution and the positions are generated by a self-avoiding random walk with the measured nearest-neighbor statistics. Each receptor has four states, closed/open and CSQ-unbound/bound, with opening rates fitted to a sigmoid of the subspace calcium concentration and CSQ binding rates that depend on luminal calcium. The subspace and junctional-SR calcium concentrations evolve under clamped myoplasmic and network-SR calcium, with a fast buffering approximation for CSQ; a fixed-time-step stochastic algorithm updates channel states and calcium concentrations together. This network geometry supplies the diffusive links through which calcium-induced calcium release recruits neighboring receptors, and the four-state scheme supplies the two separate CSQ actions that the paper then isolates by removing one effect at a time.","core_discovery":"The central discovery is that no cooperative coupling or inactivation term is needed to make a RyR cluster behave like a switch: with only calcium diffusion between receptors as the coupling, the model produces a bimodal distribution of release events, with calcium quarks under 10 ms involving few open channels and calcium sparks of roughly 20 ms in which about half of the cluster's receptors open. Cluster size sets the spark amplitude while the duration stays near 20 ms, and the properties of evoked sparks are largely independent of the LCC trigger type once a spark is initiated. Spark termination is not store depletion; the junctional SR calcium stabilizes near 500 µM and the reduced gradient, together with stochastic gating, closes the cluster, with CSQ binding at low luminal calcium further stabilizing closure. The paper's specific discovery about regulation is that removing the CSQ buffer shortens spark duration but leaves refractoriness intact, whereas removing the CSQ-bound receptor states removes the excess refractory period, so the two roles of CSQ are cleanly separated.","pith_inferences":["A direct test of the on-off switch claim would be to record from a single cluster with known super-resolution geometry and check that the peak-open-channel histogram is bimodal with a gap near one quarter of the cluster size, rather than unimodal.","The clamped-calcium simplification may be the main reason spark duration is so robust: in a beating myocyte, where bulk calcium rises and SR load cycles, the refractory period could be set by store refilling as well as CSQ binding, so the model's assignment of refractoriness to CSQ needs verification under cycling conditions.","The exponential coupling assumption implies that sparse, elongated clusters with large internal gaps should fire less reliably than compact clusters of the same size, a prediction that could be tested by comparing spark probability across clusters of different geometry from the same images.","A mutant CSQ that buffers calcium normally but cannot bind the RyR complex should preserve long sparks while shortening the refractory period, providing a clean experiment that distinguishes the two claimed roles."],"forward_implications":["Below a recruitment threshold a RyR cluster emits only calcium quarks, while above it the same cluster fires a spark involving roughly half of its channels: the cluster is a digital switch.","Spark amplitude is set by cluster size, while spark duration stays near 20 ms across cluster sizes, trigger types, and RyR opening rates.","The excess refractory period after a spark is caused by CSQ binding to the RyR complex, not by slow refilling of the junctional SR, and the model reproduces the measured recovery curves.","Raising CSQ concentration prolongs spark duration through buffering: removing the buffer eliminates the prolongation, while removing the CSQ-bound states eliminates the excess refractoriness.","Dysregulated CSQ, through stuck binding or lost buffering, disrupts the cluster's on-off switching, and raising the SR calcium load only partially restores normal release patterns."],"supporting_citations":[{"why":"Supplies the experimental single-channel RyR opening and closing rates that the model refits in sigmoid form.","marker":"[23]"},{"why":"Introduces the CSQ-bound and CSQ-unbound RyR states and the binding-rate scheme that generates excess refractoriness.","marker":"[50]"},{"why":"Provides the super-resolution cluster geometry and nearest-neighbor statistics used to generate the spatial network.","marker":"[34]"},{"why":"Establishes the clamped-calcium modeling setup and the subspace and JSR volume parameters.","marker":"[16]"},{"why":"Provides experimental evidence that CSQ overexpression prolongs spark duration, which the paper attributes to buffering.","marker":"[40]"},{"why":"Gives the measured calcium-blink and refractoriness recovery curves that the model reproduces.","marker":"[51]"},{"why":"Defines calcium sparks as the elementary release events whose amplitude and timing the model targets.","marker":"[10]"},{"why":"Supplies the LCC-to-RyR coupling latency and trigger duration used for evoked calcium sparks.","marker":"[11]"},{"why":"Shows the adjacency-matrix treatment of RyR clusters that the paper extends to exponential distance coupling.","marker":"[24,25]"}],"fun_headline_variants":["RyR clusters switch on-off without cooperative coupling","Calcium sparks arise from diffusion, not cooperativity","CSQ dual role: buffering and receptor regulation separated","Sparks terminate via calcium gradient, not store depletion","Spatial model reveals RyR switch without coupling"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that calcium in the myoplasm and network SR can be held fixed at 0.1 and 1000 µM while only the subspace and junctional SR evolve, so mass conservation is irrelevant; if the reservoirs were allowed to fluctuate as they do in a beating myocyte, the balance between depletion and refilling could change both spark termination and the refractory period, a simplification the authors flag for future work.","fun_headline_variants_meta":{"raw":{"variants":["RyR clusters switch on-off without cooperative coupling","Calcium sparks arise from diffusion, not cooperativity","CSQ dual role: buffering and receptor regulation separated","Sparks terminate via calcium gradient, not store depletion","Spatial model reveals RyR switch without coupling"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000634,"raw_usage":{"total_tokens":2959,"prompt_tokens":1011,"completion_tokens":1948,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":627,"completion_tokens_details":{"reasoning_tokens":1871}},"tokens_in":627,"tokens_out":1948,"duration_ms":16098,"temperature":1.0,"reasoning_tokens":1871,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:22:46.588423+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The cleanest test is a variant that keeps CSQ's calcium buffering but blocks CSQ-RyR binding: the model predicts long sparks with no excess refractory period, so observing a persistent refractory period in that variant would falsify the paper's assignment of refractoriness to the CSQ-bound state.","supporting_citations":[{"cited_title":"Cohen \\ and\\ author S","cited_arxiv_id":null,"evidence_quote":"Supplies the experimental single-channel RyR opening and closing rates that the model refits in sigmoid form."},{"cited_title":"Györke , author N","cited_arxiv_id":null,"evidence_quote":"Introduces the CSQ-bound and CSQ-unbound RyR states and the binding-rate scheme that generates excess refractoriness."},{"cited_title":"Franzini-Armstrong , author F","cited_arxiv_id":null,"evidence_quote":"Provides the super-resolution cluster geometry and nearest-neighbor statistics used to generate the spatial network."},{"cited_title":"\\ Jiang \\ and\\ author J","cited_arxiv_id":null,"evidence_quote":"Provides experimental evidence that CSQ overexpression prolongs spark duration, which the paper attributes to buffering."},{"cited_title":"Liu , author N","cited_arxiv_id":null,"evidence_quote":"Gives the measured calcium-blink and refractoriness recovery curves that the model reproduces."},{"cited_title":"Asfaw , author E","cited_arxiv_id":null,"evidence_quote":"Supplies the LCC-to-RyR coupling latency and trigger duration used for evoked calcium sparks."}],"review_version":1}