{"id":"43fd29ee-495d-4d30-8f55-d14713f5c6fe","arxiv_id":"2505.08452","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Escape-time stereometry on a chip measures molecular size and shape from single-molecule trapping durations, enabling mass, affinity, and conformational analysis.","lead":"This paper describes a microchip that measures the size and shape of single molecules in solution by timing how long they stay in tiny pockets. It can distinguish small mass differences, measure binding strengths, and detect insulin in serum through a shape change in the insulin receptor.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (1) is simulated only for spheres; its extension to non-spherical molecules assumes isotropic tumbling, which is questionable when h1 approaches molecular length, and every Ds-based inference depends on it.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing point: rotational averaging of non-spherical molecules into a single bounding-sphere diameter Ds. This is the right focus because the paper's central claim is a quantitative mapping from t_esc to (rH, Ds), but the mapping is validated by BD simulations of spheres only. No simulation or analytical derivation supports substituting the geometric miniball diameter for a non-spherical molecule under strong confinement. The assumption matters most where confinement is strongest, which is precisely where the method claims its greatest sensitivity and where the headline conformational claims live: DNA rise per basepair, nanostructure shape, riboswitch states, and IR-ECD compaction. The paper has genuine independent support: sphere-based calibration reproduces known protein rH, the inferred DNA rise per basepair agrees with crystallographic values, and oxDNA/SAXS comparisons support the nanostructure Ds values. These checks do not, however, test the rotational-averaging assumption, because they either involve globular proteins or use the same assumed Ds model when calibrating and fitting. The finite escape times measured for 60 bp DNA in shallow slits are a concrete hint that the miniball assignment is not literally correct there. A rigid-body simulation is the decisive, inexpensive check. Since the issue is a testable model assumption rather than a demonstrated error, the appropriate verdict remains CONDITIONAL: the paper should be accepted if the rigid-body simulation validates Eq. (1) up to L/h1 ≈ 0.8, and revised if it does not.","tokens_in":52177,"tokens_out":7218,"duration_ms":79873,"concrete_test":"Run rigid-body Brownian dynamics simulations (e.g., Ermak-McCammon with hard-wall excluded volume) for spherocylinders of length L = 5-25 nm and diameter about 2 nm in the exact trap geometry (h1 = 20-70 nm, h2 = h1 + 300 nm, pocket diameter 550 nm), including full 3D rotational diffusion and the same time-averaged detection (t_exp = 5 ms). Compute t_esc as a function of L/h1 and compare with Eq. (1) using Ds = L and rH from the Tirado cylinder model. Also record the steady-state orientation distribution in the slit versus the pocket. If t_esc deviates by more than the roughly 1% measurement precision for L/h1 > 0.5, Eq. (1) is not valid for the non-spherical analytes, and all Ds-based structural inferences would need to be re-derived using an orientation-averaged effective height.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Equation (1) is established by Brownian Dynamics simulations of spherical particles only (S5.1, with Ds = 2rH). The extension to non-spherical molecules rests on the claim in the main text, after Eq. (1), that because rotational diffusion is faster than translation, a molecule 'sweeps out a sphere' of diameter Ds, the geometric minimum bounding-sphere diameter. This assumption is not derived or simulated for rods or ellipsoids, yet it underlies every Ds inference in the paper: DNA rise per basepair, DNA nanostructure shape, SAM-IV riboswitch conformational states, and the IR-ECD compaction claim (Delta Ds ≈ 1.5 nm). In the actual measurement regime the assumption is suspect: dsDNA of 30-60 bp has length 10-20 nm in slits with h1 approximately 20-25 nm, and IR-ECD has Ds ≈ 17 nm in slits with h1 approximately 30-40 nm. Hard walls forbid isotropic rotation, so the relevant entropic excluded height is an orientation-dependent quantity between the rod diameter and the rod length, not the miniball diameter. Indeed, assigning Ds = L for 60 bp DNA in h1 ≈ 25 nm makes the denominator h1 - Ds near zero and would predict a much larger t_esc than the finite values reported, an internal inconsistency suggesting that the effective Ds is not the geometric bounding-sphere diameter. If this assumption fails, the absolute values of Ds and the inferred differences between conformational states are systematically biased.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a microfluidic 'escape-time stereometry' (ETs) method. Singly labeled molecules diffuse through nanoslits with periodic cylindrical pockets; the average time to leave a pocket (t_esc) is argued to obey Eq. (1), t_esc = A rH (h2 - Ds)/(h1 - Ds) + t0, where rH is the hydrodynamic radius and Ds is the diameter of the minimum bounding sphere. The authors calibrate A, t0, and h1 using Brownian Dynamics simulations of spheres and globular proteins of known rH, then use Eq. (1) to infer rH and Ds for dsDNA/dsRNA, DNA nanostructures, a SAM-IV riboswitch, and the insulin receptor ectodomain. The paper also measures binding affinities and kinetics for DNA hybridization, DNA-protein, protein-protein, and aptamer-insulin interactions, and demonstrates a diagnostic readout for insulin in serum based on ligand-induced compaction of IR-ECD.","tokens_in":52574,"tokens_out":8760,"duration_ms":92226,"significance":"If the central mapping holds, this is a broadly applicable, high-throughput solution-phase method for molecular size, shape, conformation, and interaction thermodynamics at single-molecule sensitivity, with clear clinical potential. The paper's strengths are its extensive validation strategy: Eq. (1) is tested by BD simulations for spheres, rH values for globular proteins agree with HYDROPRO and 2f-FCS, DNA rise-per-basepair values match crystallographic expectations, DNA nanostructure Ds values agree with oxDNA, riboswitch inferences are benchmarked against cryo-EM structures, and measured Kd values fall within a factor of 2-3 of literature values. The method is not circular: A and t0 come from simulation, h1 is calibrated with independently known protein radii, and downstream results are checked against multiple independent techniques. However, the central extension from spherical to non-spherical molecules rests on an assumption that is not directly validated, which limits the weight of the shape-related claims.","major_comments":[{"comment":"The central mapping Eq. (1) is validated by BD simulations only for spheres (S5.1), yet the paper extends it to non-spherical molecules by asserting that rapid isotropic rotation makes a translating molecule 'sweep out a sphere' of diameter Ds equal to its minimum bounding-sphere diameter. This assumption underlies every shape and conformation inference in the paper, including DNA rise per basepair, nanostructure Ds, riboswitch conformational states, and the IR-ECD compaction claim. The physical situation is not obviously compatible with the assumption: hard-wall confinement in slits with h1 comparable to molecular length restricts isotropic rotation, so the effective excluded height should be an orientation-dependent quantity between the short and long molecular axes. I request a concrete test of the assumption, either by BD simulations of rigid spheroids/cylinders with the aspect ratios and slit heights used here, or by an explicit orientational averaging model. Without such a test, the absolute values of Ds and, more importantly, the inferred differences in Ds between conformational states are not firmly established.","section":"Main text, after Eq. (1); Section S5.1; Sections S5.4-S5.7"},{"comment":"The DNA/RNA rise-per-basepair fits are performed with Ds = b*nbp and with both A and b treated as free parameters (S5.4). For the longest constructs (56-60 bp, L about 19-20 nm) in slits with h1 about 25 nm, the denominator h1 - Ds becomes small. Under the fitted parameters, the escaping time predicted for a rod with Ds equal to its full length is considerably larger than the measured t_esc values for 60 bp DNA reported elsewhere in the manuscript (e.g., Fig. 4B shows about 26 ms for 60 bp in a comparable device). This suggests that either the effective Ds is not the geometric minball diameter, or the fitted parameters absorb this inconsistency. The authors should report the measured and fitted t_esc versus n_bp data with residuals for the specific devices used, and should test the fit's robustness by fixing A from a non-DNA calibration and inferring Ds independently for each DNA length.","section":"Section S5.4, Fig. 3A, Fig. S11"},{"comment":"The Kd determination for IR-ECD binding insulin uses t_av as a proxy for bound fraction with the statement that t_av is 'proportional to the amount of IR-ECD in the liganded state.' However, Section S7.1 derives a nonlinear relation between m2 and t_av (Eq. S28). For the small t_av changes reported (a 7% decrease), the linear and nonlinear mappings may differ by only a small amount, but the quoted Kd values (2-12 nM in PBS, about 100 pM in human serum) are used as quantitative claims. The authors should state explicitly the range of validity of the linear approximation and quantify how much the inferred Kd shifts when Eq. (S28) is used instead of a simple linear interpolation.","section":"Section S7.10, Fig. 7E, Eq. (S28)"}],"minor_comments":[{"comment":"The heading 'Determination of rH and Ds relation for the SAMI-IV riboswitch' contains a typo: 'SAMI' should be 'SAM-IV.'","section":"Section S5.2 heading"},{"comment":"Please state explicitly the units used in Eq. (1). As written, A has units s/(molecular-length unit), and rH must be expressed in the same unit for the product A*rH to have units of time. The text uses rH in nm in some places and the calibration values in s/micrometer, which can confuse readers.","section":"Eq. (1) and its usage throughout"},{"comment":"The reported fit parameters for A have large fractional uncertainties (alpha = 0.16 +/- 0.34 s/nm, beta = -2.35 +/- 0.60, gamma = 0.265 +/- 0.062 s/micrometer). Because A enters every absolute determination of Ds and rH, the paper should include a statement on how these systematic uncertainties propagate into the final Ds and rH values, beyond the statistical uncertainties shown in Figs. 3 and 7.","section":"Section S5.1, Eq. (S11)"},{"comment":"The abstract claims detection 'down to two carbon atoms,' while the main text reports a demonstrated 25 Da difference (about two carbons) and a 'theoretical ability' to detect about 10 Da. The wording should be adjusted so that the demonstrated resolution is not overstated as a routine capability.","section":"Abstract and Fig. 2B"},{"comment":"The discussion of concentration inaccuracy is useful but could be more explicit: the statement that Kd is 'relatively insensitive to [A]0' should be accompanied by the parameter ranges over which this holds, particularly for the HLA and serum measurements where [A]0 is around 0.1 nM or lower.","section":"Section S7.11"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is scientifically ambitious and covers a wide range of applications. The central concern is the unvalidated extension of the sphere-based calibration to non-spherical molecules; I believe this can be addressed with additional BD simulations or an explicit orientational model, and should not require a complete reworking of the experimental claims. The paper would be strengthened by reducing the number of forward-looking statements in the abstract and by more clearly separating demonstrated capabilities from theoretical extrapolations."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth taking seriously. Escape-time stereometry (ETs) builds on the group's earlier entropic-trap work, but the new elements are substantial: a purely entropic high-salt regime, extraction of both rH and Ds from measurements at multiple slit heights, single-molecule escape-time spectra, and a wide set of applications from DNA rise per basepair to insulin detection in serum. The validation is unusually thorough: Eq. (1) is calibrated with BD simulations for spheres, rH for proteins agrees with HYDROPRO, DNA/RNA rise per basepair comes out close to crystallographic values, DNA nanostructure Ds matches oxDNA, and Kd values fall within a factor of 2-3 of literature across six orders of magnitude. The paper is also honest about its limitations, including h1 calibration uncertainty and the absence of a good apo-state structure for the insulin receptor.\n\nThe main soft spot is the extension from spheres to non-spherical molecules. Eq. (1) is simulated only for spheres; for rods and ellipsoids the paper assumes isotropic rotational diffusion sweeps out a sphere of diameter Ds, the minimum bounding-sphere diameter. That assumption is questionable when h1 approaches molecular length, as it does for 60 bp DNA (length ~20 nm in h1 ~25 nm) and for IR-ECD (Ds ~17 nm in h1 ~30-40 nm). Hard walls restrict rotation, so the relevant excluded height should be an orientation-dependent quantity somewhere between rod diameter and rod length, not the miniball diameter. The stress-test note correctly flags this: assigning Ds = full molecular length in the denominator of Eq. (1) would predict much larger escape times than observed, suggesting the effective Ds entering the fit is not simply the geometric bounding-sphere diameter. The empirical success of the model (DNA rise per basepair, oxDNA agreement) is reassuring, but it does not prove the assumption is physically correct; it may just mean the effective Ds carries systematic bias while still tracking conformational differences. I would want the authors to run BD simulations for rods and ellipsoids under confinement, or at least provide a theoretical treatment of orientation-averaged excluded height, before fully trusting absolute Ds values. The differences between states (e.g., apo vs. liganded IR-ECD) are probably more robust than the absolute numbers, but even those could be shifted if the model bias is state-dependent.\n\nOther soft spots are minor: no code or raw data are provided, and the two-height inference could carry more explicit uncertainty propagation, though the paper does sample over h1 calibration errors. None of this undermines the core claim that ETs measures rH accurately for globular proteins and gives useful, semi-quantitative shape information. The technique demonstrably works, and the breadth of applications makes it significant.\n\nThis paper deserves a serious referee. I would ask the reviewers to focus on the non-spherical assumption, require BD simulations for non-spherical particles, and encourage the authors to release their analysis code and representative raw data. My own verdict is conditional: the central platform is sound, but the Ds readout needs stronger physical grounding.","headline":"A broadly validated chip-based technique for measuring molecular size and shape in solution, with the non-spherical 'bounding sphere' assumption as the main soft spot.","tokens_in":53089,"tokens_out":2850,"would_cite":true,"duration_ms":31731,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Escape-time stereometry claims that a molecule's dwell time in nanoscale pockets encodes both its hydrodynamic radius and bounding-sphere diameter, making size, shape, interactions, and conformational change readable from a wide-field…","keywords":["escape-time stereometry","entropic fluidic trap","hydrodynamic radius","bounding-sphere diameter","single-molecule","nanoslit confinement","binding affinity","insulin receptor"],"falsifier":"Take rigid rod-like molecules of known length, such as 30 to 60 bp dsDNA, measure $t_{\\mathrm{esc}}$ in three or more calibrated slit heights, and test whether Eq. (1) with a single fixed $D_s$ (from the helix structure) and an independently measured $r_H$ fits all the data; if the $D_s$ inferred from different height pairs drifts, the effective-sphere assumption fails.","tokens_in":1810,"feed_emoji":"🔬","tokens_out":3994,"duration_ms":84743,"temperature":0.7,"pith_summary":"The paper introduces escape-time stereometry (ETs), a chip-based way to measure the size and shape of molecules in solution by filming how long each molecule lingers in nanoscale pockets built into a nanoslit. It claims the average dwell time follows $t_{\\mathrm{esc}} = A r_H (h_2 - D_s)/(h_1 - D_s) + t_0$, with two molecular unknowns: the hydrodynamic radius $r_H$ and the diameter $D_s$ of the bounding sphere. Two measurements at different slit heights are enough to extract both. If the formula holds, one microscope readout can sort molecules by weight across three decades, distinguish two-carbon differences among small molecules, follow reactions in real time, quantify binding affinities over six orders of magnitude, and expose conformational changes such as the compaction of the insulin receptor on binding insulin.","feed_headline":"One chip reads molecule size and shape from escape times","feed_subtitle":"Nanoscale pockets trap larger molecules longer; two slit heights yield radius and bounding diameter.","key_machinery":"The entropic fluidic trap: a periodic array of cylindrical indentations of total height $h_2$ in a parallel-plate slit of height $h_1$, where a molecule's residence time is amplified by the ratio $(h_2 - D_s)/(h_1 - D_s)$. The load-bearing identity is Eq. (1), $t_{\\mathrm{esc}} = A r_H (h_2 - D_s)/(h_1 - D_s) + t_0$, which converts measured dwell times into the pair ($r_H$, $D_s$). The prefactor $A$ and offset $t_0$ are fixed by Brownian-dynamics simulations of spherical particles, and $h_1$ is calibrated with globular proteins of known $r_H$; the same equation then serves for all molecular species.","core_discovery":"The central claim is that the escape time of a fluorescently labelled molecule from a cylindrical pocket in a nanoslit obeys $t_{\\mathrm{esc}} = A r_H (h_2 - D_s)/(h_1 - D_s) + t_0$, with $A$ a geometry- and viscosity-dependent prefactor fixed by Brownian-dynamics simulation. Because the pocket height $h_2$ and slit height $h_1$ are known, two escape-time measurements in different slit heights give two equations for the two unknowns $r_H$ (the Stokes or hydrodynamic radius) and $D_s$ (the diameter of the smallest sphere enclosing the molecule, which for a rotating non-spherical molecule can be much larger than $2r_H$). The authors take as evidence the agreement of inferred $r_H$ values with structure-based calculations, the recovery of roughly 3.2 Å rise per base pair for B-DNA and 2.3 Å for A-RNA, the resolution of same-mass DNA nanostructures with different shapes, and the ligand-induced compaction of the insulin receptor. They therefore present ETs as a single platform for molecular-weight determination, mixture analysis, affinity and kinetics measurements, and conformational detection in native solution.","pith_inferences":["If the two-height inference is as robust as claimed, ETs could serve as a routine solution-phase screen between size-exclusion chromatography and small-angle scattering, working on samples too dilute or heterogeneous for either.","The single-molecule escape-time spectra suggest a general way to count coexisting conformational states and their abundances without fragile multi-exponential fitting; this could be tested on a two-state folding system whose population ratio is controlled externally.","The claimed sub-1% precision in $t_{\\mathrm{esc}}$ implies that longer single-molecule trajectories could resolve mass differences below one carbon atom; a direct test would be a homologous series of small molecules measured in one calibrated chip.","If $D_s$ values inferred by ETs match $D_{\\mathrm{max}}$ values from small-angle scattering on the same constructs, the method could provide a cheap, high-throughput constraint for validating structural models and for machine-learning structure prediction."],"forward_implications":["A single minute of imaging can return $r_H$ and $D_s$ for thousands of individual molecules, giving molecular-weight, shape, affinity, and conformational readouts in native buffer without tethering or strong fields.","Because the readout responds to $D_s$ rather than mass alone, two species with nearly identical diffusion coefficients become distinguishable once the slit height is chosen close to $D_s$.","Measuring escape times at two different slit heights yields both $r_H$ and $D_s$, and with three or more heights the same data can resolve closely spaced conformational states and map them onto ellipsoidal models.","Binding affinities spanning roughly $10^{-11}$ to $10^{-4}$ M, together with on- and off-rates, follow from the same platform because the bound fraction shifts either a resolved escape-time component or the mean escape time.","Conformational compaction can dominate the escape-time change on binding, as claimed for the insulin receptor, so the method can report ligand-induced shape changes even when the mass of the complex increases."],"supporting_citations":[{"why":"Supplies the wide-field imaging and escape-event analysis pipeline that ETs inherits.","marker":"[3]"},{"why":"Introduces geometry-induced trapping of nanometric objects in a fluid, the physical basis of the pocket trap.","marker":"[24]"},{"why":"Establishes that the trap can be purely entropic for a single molecule, justifying the high-salt operating regime.","marker":"[22]"},{"why":"Provides the Brownian-dynamics simulation method from which the calibration parameters $A$ and $t_0$ in Eq. (1) are obtained.","marker":"[29]"},{"why":"Gives structure-based hydrodynamic radii used to calibrate slit heights and validate inferred $r_H$.","marker":"[14]"},{"why":"Supplies the standard A- and B-form rise-per-base-pair values that the dsDNA/dsRNA measurements reproduce.","marker":"[30]"},{"why":"Describes the two DNA nanostructures and their characterization, the key shape-discrimination test objects.","marker":"[23]"},{"why":"Provides the cryo-EM structure of the SAM-IV riboswitch used as the reference for the solution-state conformational states.","marker":"[37]"},{"why":"Provides the ligand-saturated insulin receptor ectodomain structure and binding context that anchor the compaction analysis.","marker":"[51]"}],"fun_headline_variants":["Escape times reveal molecular size and shape on a chip","Chip measures molecule dimensions via escape time","Escape-time chip discriminates size and shape","Nanopocket escape times sort molecules by size and shape","Pocket escape times size up molecules on a chip"],"cache_read_input_tokens":55168,"weakest_assumption_plain":"The method assumes a non-spherical molecule rotates quickly enough that, while diffusing through the slit, it behaves like an effective sphere of a single diameter $D_s$ (larger than twice its hydrodynamic radius), and that this one $D_s$ controls the entropic factor $(h_2 - D_s)/(h_1 - D_s)$; if real molecules do not sweep out such a sphere under strong confinement, every inferred $D_s$ would be systematically biased.","fun_headline_variants_meta":{"raw":{"variants":["Escape times reveal molecular size and shape on a chip","Chip measures molecule dimensions via escape time","Escape-time chip discriminates size and shape","Nanopocket escape times sort molecules by size and shape","Pocket escape times size up molecules on a chip"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000732,"raw_usage":{"total_tokens":3254,"prompt_tokens":904,"completion_tokens":2350,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":520,"completion_tokens_details":{"reasoning_tokens":2275}},"tokens_in":520,"tokens_out":2350,"duration_ms":15494,"temperature":1.0,"reasoning_tokens":2275,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T21:54:37.855097+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take rigid rod-like molecules of known length, such as 30 to 60 bp dsDNA, measure $t_{\\mathrm{esc}}$ in three or more calibrated slit heights, and test whether Eq. (1) with a single fixed $D_s$ (from the helix structure) and an independently measured $r_H$ fits all the data; if the $D_s$ inferred from different height pairs drifts, the effective-sphere assumption fails.","supporting_citations":[],"review_version":1}