{"id":"9f683337-adcb-46a0-83ae-90a40d59bca1","arxiv_id":"1909.01319","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Force-force correlations in an RNA-DNA unzipping experiment follow the Gumbell-universality-class shape predicted by the disordered-elastic-manifold theory.","lead":"This paper measures the force fluctuations while peeling RNA from a complementary DNA strand and finds that their correlations match a universal curve predicted for disordered systems. The match is tested by rescaling the theory curves to the data, so the load-bearing evidence is the shape agreement rather than absolute parameters.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Data-theory comparison rests on visually adjusted fits; without a quantitative model-selection test, the claimed preference for the Gumbell curve is not established.","rationale":"The paper's core empirical assertion is a shape comparison: among three theoretical curves, the Gumbell correlator is 'the closest to the data.' Everything else in the letter—the universality claim, the confirmation of FRG concepts—depends on that comparison being reliable. The weakest step is therefore the statistical procedure that selects the Gumbell curve. The paper uses a two-parameter rescaling (slope at origin fixed, large-w scale chosen visually) and provides no test of whether the residuals favor one model over another. Since the parameter-free prediction Eq. (5) fails by a factor of two, the shape comparison is not backed by an independent quantitative prediction; the fit must stand on its own. The reader's stated weakest assumption is the zero-temperature regime and possible thermal rounding. That is a genuine concern, and the paper's defense using pT values that range up to 0.7 is not quantitatively convincing. However, the shape test is defined for w>0, where white measurement noise does not directly contribute to Δ(0)-Δ(w); thermal rounding would mainly affect the small-w cusp, which is not the region that distinguishes Gumbell from exponential. The model-selection issue is more directly load-bearing: if the Gumbell preference evaporates under a proper statistical test, the central claim fails regardless of the thermal-rounding question. This is why I focus on the fit. The reader's rationale does mention the visual adjustment and the single-sequence limitation, so we partially agree, but the reader's formal weakest_assumption points elsewhere. I recommend keeping the CONDITIONAL verdict, since the paper would need to add a quantitative model comparison (or a second sequence) before the universality claim can be accepted; no change from the reader's verdict is needed.","tokens_in":9060,"tokens_out":6418,"duration_ms":66146,"concrete_test":"Perform a least-squares fit of the three candidate forms—Gumbell (3), FRG (10), and exponential—to the binned Δ(w) data used in Fig. 5, with amplitude and correlation length as free parameters, using the covariance matrix of the shape errors estimated by the resampling in Appendix A. Report reduced chi-square, ΔAIC, and a bootstrap p-value for the difference in fit quality. If the Gumbell form is not favored by ΔAIC > 2 (or similar threshold) and by a significant p-value, the visual conclusion that the data favor Eq. (3) is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the measured Δ(w) agrees with the Gumbell toy-model prediction (Eq. 3) better than with an exponential or the leading FRG curve. The evidence for this is presented in Sec. III / Fig. 5: each theoretical curve is rescaled to the same measured slope at the origin, and the remaining parameter—the large-w decay scale—is 'adjusted visually' (footnote 37 context). No chi-square, likelihood, or confidence interval for the model comparison is reported. The green error band in Fig. 5 is obtained after removing amplitude fluctuations (Appendix A), but the visually chosen large-w scale is not folded into that error, so the band does not represent the uncertainty of the shape comparison. Appendix B shows the analysis can recover an exponential correlator when the data are generated as exponential; that is a valid positive control, but it does not show that the three candidate curves are distinguishable at the actual noise level and over the finite w range (0–0.3 µm, roughly 5 correlation lengths). The absence of a quantitative comparison matters because Eq. (5), the parameter-free prediction for the correlation length, misses the measured ξ by about a factor of 2 (Sec. IV); the agreement is reached only after making the correlation length a free fit parameter. A second, related gap is that the universality claim rests on a single RNA sequence, so even a visually excellent match would not demonstrate sequence independence.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports measurements of force-force correlations in the plateau region of RNA-DNA unzipping force-extension curves, comparing the measured correlator Δ(w) with three theoretical forms: an exact 1-d toy-model solution in the Gumbell universality class (Eqs. 3–5), the leading-order functional renormalization group result (Eqs. 9–10), and a purely exponential decay. After rescaling each theoretical curve to match the measured slope at the origin and visually adjusting the large-w scale (Sec. III, Fig. 5), the authors conclude that the Gumbell form agrees best with the data, and they interpret this as evidence for universal, sequence-independent force correlations characteristic of the depinning transition. They also extract a correlation length of about 186 base pairs and discuss implications for the resolution of unzipping experiments and for comparisons with hairpin unzipping data.","tokens_in":9365,"tokens_out":4829,"duration_ms":51056,"significance":"If the central claim is established, the experiment would provide a rare experimental test of functional renormalization group predictions for disordered elastic systems, here in dimension d=0, using a biologically relevant single-molecule system. The paper includes a careful data-analysis protocol with resampling error estimates (Appendix A) and a positive control on synthetic exponential data (Appendix B), both of which are strengths. However, the significance is tempered by the absence of a quantitative model-selection test and by the factor-of-two failure of the parameter-free prediction for the correlation length (Eq. 5 in Sec. IV), so the present evidence supports a functional-form coincidence more strongly than a quantitative confirmation of the theory.","major_comments":[{"comment":"The central claim that the measured Δ(w) favors the Gumbell prediction over the exponential and FRG forms is established only by visual inspection: each curve is rescaled to the measured slope at the origin and the remaining large-w scale is 'adjusted visually' (footnote 37). No chi-square, likelihood, or any other model-selection statistic is reported, and the green error band in Fig. 5 is obtained after removing amplitude fluctuations (Appendix A) and does not incorporate the uncertainty in the visually chosen length scale. As a result, the paper does not quantitatively demonstrate that the three candidate functional forms are distinguishable at the actual noise level over the finite range w ∈ [0, 0.3] µm. Please provide a quantitative comparison, such as a reduced chi-square or likelihood ratio, that accounts for the uncertainty in both fitted scales, and report the resulting confidence in the preference for the Gumbell form.","section":"Sec. III, Fig. 5"},{"comment":"The parameter-free prediction for the correlation length is not confirmed by the experiment: using the observed ξ = 0.055 µm in Eq. (5) leads to a predicted force fluctuation dF ≈ 2.6 pN, about a factor of two larger than the measured dF = 1.14 pN. Since the shape comparison in Sec. III is performed only after treating the length scale as a free fit parameter, the agreement demonstrates a match of the scaling function but not the quantitative predictive content of the toy model. The manuscript acknowledges this discrepancy in qualitative terms, but the abstract and Sec. III present the agreement as 'excellent' without clearly separating the shape test from the failed scale prediction. Please state explicitly that the confirmed content is the functional form only, and discuss whether the factor-of-two discrepancy in the scale affects the claimed universality.","section":"Sec. IV, Eq. (5)"},{"comment":"The claim that the measured correlations are sequence-independent or universal rests on data from a single RNA sequence (23S ribosomal RNA). The argument that this sequence is 'not random' yet yields agreement with the random-sequence toy model is suggestive, but no test with a different sequence is presented, and no quantitative self-averaging argument is given to show that the 186-base-pair correlation length is sufficient to erase sequence-specific features. Without such a test or argument, the title's 'universal' claim is not directly demonstrated by the data; the paper should either add a second sequence or substantially soften the universality claim and frame the result as evidence for the relevant universality class in this particular system.","section":"Sec. I and Sec. IV"},{"comment":"The justification for applying zero-temperature depinning correlators (Eqs. 3 and 10) despite observable thermal fluctuations relies on the estimate pT = e^{-δG/kBT}, but the authors themselves state that pT ranges from 8×10^-3 to 0.7, i.e., it is not uniformly small; thermal noise is also visibly present as white noise in the data. Since the distinguishing feature of the theoretical curves is their cusp at the origin, and thermal fluctuations are expected to round this cusp, the absence of visible rounding should be supported by a quantitative estimate rather than by the order-of-magnitude argument given. Please provide a quantitative bound on the expected cusp rounding from the thermal bond-breaking rate and from the white-noise amplitude, and show that the observable w-range is insensitive to it.","section":"Sec. IV"}],"minor_comments":[{"comment":"The last sentence contains a grammatical error: 'a biologically inspired experiments' should be 'a biologically inspired experiment'.","section":"Abstract"},{"comment":"The notation for the connected expectation could be made explicit by defining ⟨F(w)F(w')⟩_c as the cumulant, since some readers may not be familiar with the subscript 'c'.","section":"Eq. (2)"},{"comment":"The effective stiffness m^2 is stated as 55 ± 5 pN/µm, and Eq. (5) uses ln(m^{-2}); please clarify the units of m inside the logarithm, since m has dimension pN/µm while m^{-2} is used as a dimensionless quantity.","section":"Footnotes 27 and Sec. IV"},{"comment":"The definition N_p := ΣΠ_i (A9) is confusing: the text first sets N_p = 100 in practice, then defines N_p as the number of partitions. Please rename one of these or explain the relation between the two uses.","section":"Appendix A, Eq. (A9)"},{"comment":"The color coding of the three theoretical curves is described in the text, but the figures would be easier to read if the legend appeared directly on each panel; in Fig. 4 the grey solid line (the mean) and the grey dotted error estimate have similar shades, which may be hard to distinguish in print.","section":"Fig. 4 and Fig. 5"},{"comment":"The positive control shows that the analysis pipeline can recover an exponential correlator from synthetic exponential data, which is valuable. However, it would be even more informative to report synthetic tests for the Gumbell and FRG forms as well, showing that the pipeline can distinguish them at the actual noise level.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"This is an interesting and well-written paper that addresses an important question in disordered systems, and the experimental data are unique. The main weakness is that the central claim of 'excellent agreement' with the Gumbell prediction is supported primarily by visual inspection after fitting the two scales, and the one parameter-free prediction (Eq. 5) fails by a factor of two. I think the paper is publishable after major revisions that add a quantitative model comparison and that carefully delineate what is confirmed (the scaling function) from what is not (the predicted scale and the sequence independence). The single-sequence aspect is a substantial limitation for the universality claim and should be addressed honestly, either with an additional experiment or with a clear caveat."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis one is worth your attention if you work on depinning or the functional RG: the authors measure the force-force correlator Δ(w) in RNA-DNA unzipping and compare it to the Gumbell toy-model shape. That measurement is genuinely new. They use 163 force-extension curves and a stable estimator Δ(0)−Δ(w). The positive control in Appendix B is a real plus: they generate exponentially correlated test data and show the pipeline recovers the exponential curve, so the analysis is not obviously rigged.\n\nThe central claim, however, is softer than the text admits. The theoretical curves are rescaled to the measured slope at the origin, and the remaining large-w decay scale is adjusted visually (Sec. III). No chi-square, likelihood, or confidence interval is reported for the shape comparison. The green error band in Fig. 5 is only the shape error after removing amplitude fluctuations; it does not include the uncertainty in the visually chosen scale. So \"excellent agreement\" is an eyeball claim, not a measured one. The factor-of-two miss in Eq. (5)—2.6 pN predicted versus 1.14 pN observed—makes the correlation length effectively a fitted parameter, not a clean prediction. To the authors' credit, they acknowledge this and call the estimate crude.\n\nTwo other soft spots. The sequence-independence claim rests on one RNA sequence; that is a limitation, not fatal, but it should be said plainly. And the thermal-rounding argument in Sec. IV is plausible but hand-wavy: pT ranges from 0.008 to 0.7, and no quantitative derivation shows why the zero-temperature cusp remains unrounded.\n\nThe citation pattern is appropriate; the theory curves come from prior papers, including some by the first author, which is expected here.\n\nVerdict: send it to peer review, but expect major revision. A referee should ask for a quantitative model comparison (even a simple chi-square over the fitted scales) and, ideally, a second sequence. As it stands, the paper is a solid proof-of-concept that the Gumbell shape is present in this system, not a proof of universality. I'd consider it for a reading group, mostly to discuss how to quantify shape comparisons in single-molecule experiments.","headline":"New experimental measurement of the depinning force correlator in RNA-DNA unzipping; the Gumbell shape fits nicely, but only after fitted scales and a factor-two miss on the predicted length scale.","tokens_in":9915,"tokens_out":3459,"would_cite":true,"duration_ms":33021,"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":"Force fluctuations during RNA-DNA unzipping match the universal prediction of a particle dragged through a random landscape.","keywords":["RNA-DNA unzipping","force-force correlations","depinning transition","disordered elastic systems","functional renormalization group","Gumbell universality class","single-molecule force spectroscopy","universality"],"falsifier":"Measure $\\Delta(w)$ on a construct with a substantially stiffer or softer trap: the Gumbell prediction fixes the correlation length as $\\rho_m=(1/m^2)\\sqrt{2\\ln(m^{-2})}$ and the shape of $\\Delta(w)$, so a measured shape that deviates from Eq. (3) beyond the stated error bars, or a correlation length that does not scale with $m^2$ as predicted, would falsify the universality claim. Alternatively, vary temperature over a range where $p_T$ changes by orders of magnitude: if the cusp of $\\Delta(w)$ rounds when more bonds can be thermally opened, the zero-temperature assumption fails.","tokens_in":8861,"feed_emoji":"🧬","tokens_out":11259,"duration_ms":90952,"temperature":0.7,"pith_summary":"Force fluctuations on the plateau of an RNA-DNA unzipping curve, extracted from 163 experimental force-extension measurements, match the exact prediction of a one-dimensional toy model—a particle pulled through a Gaussian random force landscape—better than they match a pure exponential or the leading-order field-theory curve. The paper identifies this signal as the renormalized disorder correlator of the depinning transition, the central object of the functional renormalization group for disordered elastic systems. Because the molecule's sequence is specific biological RNA, the agreement implies that the macroscopic force-force correlations are sequence-independent and universal. This matters because it connects a biologically inspired single-molecule experiment to a broad class of disordered systems, and because the correlation length of about 186 base pairs sets the resolution limit for reading biological information out of unzipping curves.","feed_headline":"RNA-DNA unzipping confirms the 1-d random-force toy model","feed_subtitle":"Force-force correlations in 163 pulling curves match the exact prediction, not exponential decay.","key_machinery":"The central object is the force-force correlator $\\Delta(w)=\\langle F(w)F(w')\\rangle_c$, which on the unzipping plateau measures how force fluctuations at two trap separations are correlated. The load-bearing theoretical input is the exact solution of a one-dimensional toy model: a particle dragged through a random force landscape, whose Gaussian forces place it in the Gumbell universality class of extreme-value statistics, with correlator $\\Delta_{\\text{Gumbell}}(x)=x^2/2+\\operatorname{Li}_2(1-e^{|x|})+\\pi^2/6$. The argument is that the experiment measures this same universal object—the renormalized disorder correlator of the field theory of disordered elastic manifolds, here in internal dimension $d=0$. The matching of the shape of $\\Delta(w)$, not just an overall amplitude, is what establishes the universality claim.","core_discovery":"The central claim is that the measured connected force-force correlator $\\Delta(w)=\\langle F(w)F(w')\\rangle_c$ on the unzipping plateau is given, up to two non-universal scales, by the Gumbell form $\\Delta_{\\text{Gumbell}}(x)=x^2/2+\\operatorname{Li}_2(1-e^{|x|})+\\pi^2/6$, the exact solution of a particle driven through a random force landscape with Gaussian-distributed disorder. The data from 163 force-extension curves favor this curve over the leading-order FRG fixed-point function $\\Delta_{\\text{FT}}(x)=-W(-e^{-x^2/2-1})$ and over an exponential decay, after rescaling all curves to the same slope at the origin. The paper argues that the plateau forces fluctuate around a critical value $F_c\\approx60\\,\\text{pN}$ and that the measured $\\Delta(w)$ is the renormalized disorder correlator of the depinning transition for an elastic object of internal dimension $d=0$. It further interprets the slope at the origin through the relation $|\\Delta'(0^+)|=m^2\\delta F_m$, yielding a mean force drop $\\delta F_m=0.43\\pm0.05\\,\\text{pN}$ and a correlation length $\\xi\\approx186$ base pairs, consistent with the force drops visible in single curves. The paper concludes that universal physics emerges from a specific, non-random biological sequence.","pith_inferences":["The agreement with the toy model suggests that at the scale of the measurement the microscopic disorder is effectively Gaussian; if so, other observables of the same universality class—such as record statistics of force maxima or avalanche size distributions in the plateau—should also match the toy-model predictions, an extension the paper does not test.","If the zero-temperature cusp of $\\Delta(w)$ is truly unrounded by temperature, a systematic temperature-dependence study should show no change in the shape of $\\Delta(w)$; such a study could also test the paper's estimate that thermal bond-breaking is negligible for most base pairs.","The paper's resolution argument implies a design principle: stiffer, well-aligned optical traps should yield sharper force-drop features in unzipping curves, and re-analysis of datasets with different nominal stiffness could confirm the predicted $\\rho_m\\sim(1/m^2)\\sqrt{2\\ln(m^{-2})}$ scaling.","One might expect the same universal correlator to appear in other driven biophysical systems with quenched disorder, such as protein unfolding or nanopore translocation, where the control parameter plays the role of trap distance; this is an extrapolation beyond the paper's data."],"forward_implications":["The plateau of an RNA-DNA unzipping curve is a macroscopic realization of the depinning transition of a disordered elastic system with internal dimension $d=0$; the measured correlator is the renormalized disorder correlator of that theory.","Force-force correlations decay over roughly 186 base pairs, so sequence-specific biological events in unzipping experiments can only be resolved at that scale; increasing trap stiffness $m^2$ shortens the correlation length and improves resolution.","The same universal signal should be recoverable from other peeling or unzipping experiments on random or biological sequences, while hairpin unzipping is predicted to fall in a different universality class with correlation length scaling as $\\rho_m\\sim m^{-4/3}$.","Because the sequence used is a real ribosomal RNA sequence, the result implies that universal, sequence-independent physics can coexist with and be extracted from a specific biological molecule, providing a benchmark for single-molecule force spectroscopy."],"supporting_citations":[{"why":"Supplies the exact Gumbell correlator prediction for a particle in a Gaussian random force landscape, the central theoretical curve the experiment is compared to.","marker":"[28]"},{"why":"Provides a previous experimental measurement of the renormalized disorder correlator in contact-line depinning, on whose protocol the present analysis is modeled.","marker":"[14]"},{"why":"Derives the equation of motion and functional renormalization group for driven interfaces in random media, giving the field-theoretic framework for $\\Delta(w)$.","marker":"[20]"},{"why":"Provides the two-loop FRG fixed-point function used as the competing theoretical prediction $\\Delta_{\\text{FT}}$.","marker":"[24]"},{"why":"States that the renormalized disorder correlator of the FRG is the quantity measured as $\\Delta(w)$, the identification central to the paper's claim.","marker":"[31]"},{"why":"Gives the relation $|\\Delta'(0^+)|=m^2\\delta F_m$ between the slope at the origin and the mean force drop, used to extract the correlation length and force-drop scale.","marker":"[38]"},{"why":"Describes the overstretching experiments and molecular construct from which the force-extension curves are taken, the dataset underlying the analysis.","marker":"[7]"},{"why":"Supplies the RNA/DNA hybrid duplex thermodynamic parameters used to estimate the thermal bond-breaking probability that justifies the zero-temperature approximation.","marker":"[40]"}],"fun_headline_variants":["RNA-DNA unzipping shows universal force correlations","Universal force correlations emerge in RNA-DNA unzipping","RNA-DNA experiments confirm universal Gumbel correlations","Force correlations in RNA-DNA unzipping are universal","RNA-DNA unzipping confirms universal disorder correlator"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes that the unzipping plateau is in the zero-temperature depinning regime, so the zero-temperature correlator shapes apply without thermal rounding of the cusp; the paper supports this with the small probability $p_T=e^{-\\delta G/k_BT}$ that a bond is thermally broken, but thermal noise is present in the data and the argument is not a quantitative derivation of the absence of rounding.","fun_headline_variants_meta":{"raw":{"variants":["RNA-DNA unzipping shows universal force correlations","Universal force correlations emerge in RNA-DNA unzipping","RNA-DNA experiments confirm universal Gumbel correlations","Force correlations in RNA-DNA unzipping are universal","RNA-DNA unzipping confirms universal disorder correlator"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000194,"raw_usage":{"total_tokens":1343,"prompt_tokens":921,"completion_tokens":422,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":348}},"tokens_in":537,"tokens_out":422,"duration_ms":3949,"temperature":1.0,"reasoning_tokens":348,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:20:52.190992+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $\\Delta(w)$ on a construct with a substantially stiffer or softer trap: the Gumbell prediction fixes the correlation length as $\\rho_m=(1/m^2)\\sqrt{2\\ln(m^{-2})}$ and the shape of $\\Delta(w)$, so a measured shape that deviates from Eq. (3) beyond the stated error bars, or a correlation length that does not scale with $m^2$ as predicted, would falsify the universality claim. Alternatively, vary temperature over a range where $p_T$ changes by orders of magnitude: if the cusp of $\\Delta(w)$ rounds when more bonds can be thermally opened, the zero-temperature assumption fails.","supporting_citations":[{"cited_title":"Driven particle in a random landscape: disorder correlator, avalanche distribution and extreme value statistics of records","cited_arxiv_id":"0808.3217","evidence_quote":"Supplies the exact Gumbell correlator prediction for a particle in a Gaussian random force landscape, the central theoretical curve the experiment is compared to."},{"cited_title":"Height fluctuations of a contact line: a direct measurement of the renormalized disorder correlator","cited_arxiv_id":"0904.4156","evidence_quote":"Provides a previous experimental measurement of the renormalized disorder correlator in contact-line depinning, on whose protocol the present analysis is modeled."},{"cited_title":"Nattermann, S","cited_arxiv_id":null,"evidence_quote":"Derives the equation of motion and functional renormalization group for driven interfaces in random media, giving the field-theoretic framework for $\\Delta(w)$."},{"cited_title":"2-loop Functional Renormalization Group Theory of the Depinning Transition","cited_arxiv_id":"cond-mat/0205108","evidence_quote":"Provides the two-loop FRG fixed-point function used as the competing theoretical prediction $\\Delta_{\\text{FT}}$."},{"cited_title":"Wiese and P","cited_arxiv_id":null,"evidence_quote":"States that the renormalized disorder correlator of the FRG is the quantity measured as $\\Delta(w)$, the identification central to the paper's claim."},{"cited_title":"Size distributions of shocks and static avalanches from the Functional Renormalization Group","cited_arxiv_id":"0812.1893","evidence_quote":"Gives the relation $|\\Delta'(0^+)|=m^2\\delta F_m$ between the slope at the origin and the mean force drop, used to extract the correlation length and force-drop scale."},{"cited_title":"Melkonyan, M","cited_arxiv_id":null,"evidence_quote":"Describes the overstretching experiments and molecular construct from which the force-extension curves are taken, the dataset underlying the analysis."},{"cited_title":"Sugimoto, S","cited_arxiv_id":null,"evidence_quote":"Supplies the RNA/DNA hybrid duplex thermodynamic parameters used to estimate the thermal bond-breaking probability that justifies the zero-temperature approximation."}],"review_version":1}