{"id":"9d2c2837-46c8-46b5-b6e2-9b166b43f635","arxiv_id":"2510.10061","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"A microfiber's own surface defects scatter light into an interference pattern that reveals single protein molecules and their dynamics label-free.","lead":"Using a thin optical fiber as both light guide and sensor, the authors watch single protein molecules land, move, and respond to sound without fluorescent labels. The potential payoff is label-free single-molecule imaging that avoids photobleaching and label artifacts, plus an acoustic sensor at the molecular scale.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Spot-to-molecule mapping is unproven: no single-step photobleaching or dilution series, and 0.05 aM detection conflicts with diffusion-limited arrival; central 38 dB/single-molecule claims therefore lack support.","rationale":"The reader's weakest assumption—that each bright spot is a single molecule rather than a scattering artifact or impurity—is indeed the load-bearing point. The paper's fluorescence validation is only a single time-correlated onset, not a per-molecule demonstration; the photobleaching behavior described is more consistent with a small ensemble than a single fluorophore. I add a sharper quantitative tension: at 0.05 aM, diffusion-limited transport to the microfiber in a quiescent 500 µL volume yields only ~0.1 molecule arrivals over the reported observation window, so the claimed detection of multiple binding events cannot be explained by bona fide BSA molecules unless an unstated transport or concentration mechanism operates. This directly undermines the sensitivity claim and, because the SNR comparison is not derived from a quantitative model, also undermines the 38 dB claim. Given these gaps, rejection remains justified; the missing controls and analysis prevent verification of the central claims. The proposed single-step photobleaching colocalization assay would settle whether spots are single molecules, and replicate counting at 0.05 aM would test the diffusion-limited sensitivity claim. No independent support—machine-checked proofs, code, or raw data—is provided to offset these concerns.","tokens_in":9363,"tokens_out":6401,"duration_ms":61551,"concrete_test":"Perform a dual-channel single-step photobleaching assay on sparsely adsorbed Rhodamine-B-labeled BSA: record simultaneously until complete photobleaching and require that every differential scattering spot selected for kinetic analysis is colocalized with a fluorescence spot that disappears in a single step while the scattering spot persists. Repeat at the claimed 0.05 aM in at least three independent 500 µL trials and compare the observed spot count with the diffusion-limited expectation (~0.1 molecules per 2000 s). If spots show multi-step bleaching or no fluorescence colocalization, the single-molecule and 0.05 aM claims fail.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—38 dB SNR gain, 0.05 aM sensitivity, and single-molecule dynamics—rests on identifying each differential bright spot as one BSA molecule. Section 2.2 reports only correlated onset of scattering with ensemble Rhodamine-B fluorescence, not stepwise photobleaching, intensity quantization, or a systematic dilution series with replicates. A single fluorophore should photobleach in one step; the paper instead says fluorescence 'rapidly photobleached after 63 seconds,' which suggests a small aggregate or ensemble rather than a single molecule. No protein-free control is quantified beyond a citation to Supplementary Fig. S2. Additionally, at 0.05 aM the solution contains ~15 molecules in 500 µL. For a quiescent 2.9 cm × 1.3 µm fiber, the diffusion-limited arrival rate of BSA (D ≈ 6×10⁻¹¹ m²/s) is about 4×10⁻⁵ s⁻¹, i.e. ~0.1 molecules over 2000 s. Observing multiple binding events at this concentration would require unstated convection, concentration artifacts, or misassignment of impurities/aggregates. Thus the extraordinary sensitivity and the SNR claim are not supported by the evidence presented.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a label-free interferometric imaging approach on a microfiber, using in-plane scattering from natural surface defects to detect single molecules. It claims an approximately 38 dB SNR enhancement over conventional fluorescence, detection of BSA down to 0.05 aM (~15 molecules in 500 μL), real-time tracking of conformational transitions and binding dynamics, and acoustic-wave modulation at the single-molecule level. The theory is based on coherent superposition of scattered fields from surface scatterers; experiments use a 1.3 μm-diameter fiber with 473 nm evanescent illumination.","tokens_in":9747,"tokens_out":6366,"duration_ms":57952,"significance":"If the central claims were supported, this would be a substantial advance: label-free single-molecule imaging on a simple microfiber platform with high SNR, long-term stability, and acoustic transduction would interest a broad community. The paper has some positive elements: the coherent imaging model in Eq. (1) is standard, and the undersampling simulation in §2.5 correctly predicts the ~3 Hz aliased frequencies from a 20 fps camera for 503–504.5 Hz excitation. However, the experimental evidence does not establish the single-molecule interpretation, the SNR gain is never quantitatively defined, and the 0.05 aM sensitivity claim is inconsistent with diffusion-limited arrival. As written, the manuscript is an interesting proposal rather than a validated single-molecule technique.","major_comments":[{"comment":"The central quantitative claim—'approximately 38 dB enhancement in SNR over conventional fluorescence methods'—is stated without any definition of SNR, noise model, or calculation. Without specifying which quantity is compared (e.g., peak intensity over background fluctuation, detection limit, or frame-to-frame standard deviation), the claim is not testable. Moreover, scattering and fluorescence signals have different physical units and noise statistics; a direct dB comparison requires a common, explicitly defined metric.","section":"Abstract; §2.2, Fig. 1g"},{"comment":"Bright spots are identified as single BSA molecules based on simultaneous onset with Rhodamine-B fluorescence and a single control. No stepwise photobleaching, single-emitter intensity analysis, or concentration-dilution series with replicate statistics is provided. The statement that 'fluorescence signal rapidly photobleached after 63 seconds' (Fig. 2f) does not demonstrate a single fluorophore; it is also consistent with a small aggregate or ensemble. The protein-free control is only cited as Supplementary Fig. S2 and is not quantified, so the spot-to-molecule assignment is not established.","section":"§2.2, Fig. 2"},{"comment":"At 0.05 aM there are ~15 molecules in 500 μL. For a quiescent 2.9 cm × 1.3 μm fiber, the diffusion-limited arrival rate (D ≈ 6×10⁻¹¹ m²/s) is roughly 4×10⁻⁵ s⁻¹, or ~0.1 molecules over 2000 s. Observing multiple binding events at this concentration therefore requires an unstated transport mechanism (convection, electrokinetics, or a concentration artifact). Since the concentration-dependence data in Fig. 2c,d rely on this point, the claimed sensitivity is physically inconsistent unless such a mechanism is described and demonstrated.","section":"§2.2, Fig. 2c–d"},{"comment":"The interpretation that decreasing scattering intensity at low pH reflects BSA unfolding increasing the molecule–surface distance is not uniquely supported. The signal could equally arise from desorption, aggregation, pH-induced refractive-index changes of the solution, or focus drift. No control experiments (e.g., pH change without protein, or independent verification that the protein remains surface-bound) are presented, so the conformational-transition claim is not established.","section":"§2.3, Fig. 3"},{"comment":"The acoustic-wave measurements attribute FFT peaks to modulation of single-molecule binding dynamics. However, the microfiber and the surrounding medium are mechanically coupled to the transducer; the observed intensity modulations and FFT peaks could stem from fiber vibration or changes in the interference background rather than molecular motion. A control without molecules, or on a region away from the molecule, is not reported. The undersampling simulation is standard, but it does not validate the molecular origin of the signal.","section":"§2.5, Fig. 5"},{"comment":"The interference term in Eq. (1) is 2|F_j||F_k|cos(Δφ_jk). The contrast therefore depends on the product of the scattered-field amplitudes, not only on their phase balance. The statement that the high-contrast pattern is 'independent of the intrinsic scattering cross section of individual molecules' is not supported by the equation; |F_j| and |F_k| enter directly. If the intended argument is different (e.g., the ratio of the cross term to the incoherent background), it should be stated explicitly.","section":"§2.1, Eq. (1), Fig. 1d"}],"minor_comments":[{"comment":"Equation (1) is typeset incorrectly (e.g., ',-' and '.') and the symbols are not fully defined. Please rewrite in standard notation and define F_j, h(x), and the convolution.","section":"Eq. (1)"},{"comment":"The error bars in Fig. 5g are not described. Please state the number of repeats and how the standard deviation was computed.","section":"Fig. 5g"},{"comment":"The text says the obtained frequency is 'about three orders of magnitude lower' than the acoustic excitation. For 503 Hz sampled at 20 fps, the alias is 3 Hz, which is about two orders lower; please correct.","section":"§2.5"},{"comment":"The fiber surface is described as plasma-hydroxylated in §4.2 but as 'silanized' in §4.4. Clarify the actual surface chemistry and whether a silane layer was used.","section":"§4.2 and §4.4"},{"comment":"Reference 54 appears to duplicate Reference 6 and has a title that does not match the cited topic. Please verify and correct.","section":"Reference list"}],"recommendation":"reject","confidential_remarks":"The manuscript's headline claims are not supported by the evidence. The most serious issue is the 0.05 aM detection claim, which is inconsistent with diffusion-limited transport unless an unstated mechanism exists; the SNR gain is undefined; and the single-molecule assignment lacks the usual controls (stepwise photobleaching, intensity quantization, dilution statistics). These are load-bearing problems that would require substantial new experiments and possibly a reframing of the claims. I would not recommend requesting major revision unless the authors supply such data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me give you the short version first. The platform is real and somewhat original: a microfiber's evanescent field plus interference from intrinsic surface defects, used to image protein adsorption and acoustic response. That's worth a look. But the paper's central quantitative claims — the 38 dB SNR gain, 0.05 aM detection, and the assignment of each bright spot to a single BSA molecule — are not supported by the data as presented. The stress-test note is right, and the reader's REJECT is defensible.\n\nWhat's good: the experimental arrangement is simple and low-cost; hyperspectral imaging and the dual-channel scattering/fluorescence setup are thoughtful; the acoustic under-sampling explanation (503 Hz at 20 fps) is internally consistent. Section 2.5's FFT peaks tracking drive frequency is the cleanest result in the paper.\n\nThe soft spots are concentrated where the load-bearing claims live. The 38 dB SNR figure appears without definition, noise model, or calculation. Eq. (1) is generic coherent-imaging theory and doesn't show contrast independent of molecular scattering cross section — the cross term scales with |F_j||F_k|. More importantly, the spot-to-molecule mapping is unproven. The fluorescence validation is ensemble-level: no single-step photobleaching, no intensity quantization, no dilution series with replicates. The fluorescence signal 'rapidly photobleached after 63 s' actually argues against a single fluorophore — a single RhB should show one discrete step. And the 0.05 aM claim collides with diffusion: ~15 molecules in 500 µL, and a quiescent fiber would see far less than one arrival over 2000 s. The paper offers no convection or concentration artifact control. The pH-unfolding interpretation (Section 2.3) doesn't rule out desorption or refractive index changes.\n\nThese are not minor. The core selling point — label-free single-molecule sensitivity at 0.05 aM with 38 dB SNR — is exactly what is unsupported. The paper would need raw data, an SNR definition, a proper dilution series, and controls against nonspecific scattering and drift before it can carry those claims.\n\nWho should read it: researchers working on label-free scattering imaging and microfiber sensors will find the setup idea worth a look, but should not cite it as evidence of single-molecule sensitivity without replication.\n\nRecommendation: send to peer review — the platform is interesting enough to merit referee time — but expect major revision or rejection if the authors cannot produce the missing controls.","headline":"A genuinely new microfiber interferometric imaging scheme, but the single-molecule and 0.05 aM claims are not backed by the evidence; needs major revision before it can be believed.","tokens_in":10122,"tokens_out":1741,"would_cite":false,"duration_ms":15859,"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":"A bare optical fiber can image single molecules label-free, with signal-to-noise roughly 38 dB above fluorescence.","keywords":["single-molecule detection","label-free imaging","interferometric scattering","optical microfiber","evanescent field","protein dynamics","biosensing","acoustic sensing"],"falsifier":"Run the identical differential-imaging protocol on the same microfiber with a particle-free buffer (e.g., filtered to 0.02 µm) and no protein; if bright spots still appear, or if spot counts do not scale predictably with protein concentration, the single-molecule interpretation collapses.","tokens_in":9273,"feed_emoji":"🔬","tokens_out":4318,"duration_ms":40946,"temperature":0.7,"pith_summary":"The paper proposes that the interference pattern formed by light scattered from natural defects on a microfiber surface can act as a highly sensitive, label-free reporter for single-molecule events. It claims an approximately 38 dB signal-to-noise enhancement over conventional fluorescence, detection of BSA down to 0.05 aM (about 15 molecules in 500 µL), and real-time tracking of conformational transitions, binding/unbinding dynamics, and acoustic stimulation of individual proteins. If correct, this would enable long-term single-molecule imaging without photobleaching and push biosensing toward the molecular scale on a simple fiber platform.","feed_headline":"Single molecules imaged on bare optical fiber","feed_subtitle":"Evanescent-field interference tracks protein binding, unfolding, and acoustic response in real time.","key_machinery":"The central mechanism is coherent interference among light fields scattered by natural defects on the microfiber surface, illuminated by the fiber's evanescent field. The image intensity includes a cross term proportional to the product of two scattered amplitudes times the cosine of their phase difference; a molecule perturbing one field's phase shifts this term, producing a detectable localized intensity change that is independent of the molecule's intrinsic scattering cross section.","core_discovery":"The paper demonstrates that a microfiber supporting a strong evanescent field produces high-contrast interference patterns from in-plane scattering by natural surface defects. When a single molecule binds to the fiber, it locally changes the phase of one scattered field, altering the interference pattern in a way that can be imaged. The authors show this allows label-free visualization of single BSA molecules, with binding kinetics, pH-induced unfolding, distinct binding states (including a 'dancing' state), and acoustic-wave responses all captured in real time—achieving an ~38 dB SNR improvement over fluorescence without plasmonic or microcavity amplification.","pith_inferences":["The 'dancing' state may offer a direct readout of the molecule–surface binding free-energy landscape, a property typically inferred only from ensemble kinetics.","Because the interference contrast does not depend on the molecule's scattering cross section, the method could generalize to weakly scattering analytes, such as small molecules or dielectric nanoparticles.","The undersampling trick could be exploited to sense high-frequency acoustic or mechanical signals using low-cost, low-frame-rate cameras when the excitation frequency is known.","If confirmed by independent single-molecule calibration, the phase-change mechanism could be integrated into silicon photonic chips for compact, arrayed single-molecule biosensing."],"forward_implications":["Single proteins can be tracked label-free for long durations without photobleaching, enabling studies of slow conformational dynamics.","Ultra-low analyte concentrations (down to ~15 molecules in 500 µL) become detectable, potentially enabling early biomarker detection.","Distinct molecular binding states (bound, unbound, and 'dancing') can be classified from intensity time traces, revealing heterogeneity in surface interactions.","The same microfiber can function as a molecular-scale acoustic sensor, with acoustic frequency recoverable via undersampling and filtering.","The technique may extend to other biomolecules or nanoparticles, providing a general platform for label-free single-molecule imaging and biosensing."],"fun_headline_variants":["Label-free single-molecule imaging on a bare microfiber","Evanescent-field interferometry reveals single-protein dynamics","38 dB SNR boost: single molecules seen without labels","Microfiber evanescent field captures single-molecule motions","Single proteins tracked in real time on a bare optical fiber"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"Each bright spot is a single molecule: the interference signal change must be dominated by a molecule-induced phase shift at the fiber surface, not by scattering from buffer impurities or by mechanical or thermal drift of the fiber.","fun_headline_variants_meta":{"raw":{"variants":["Label-free single-molecule imaging on a bare microfiber","Evanescent-field interferometry reveals single-protein dynamics","38 dB SNR boost: single molecules seen without labels","Microfiber evanescent field captures single-molecule motions","Single proteins tracked in real time on a bare optical fiber"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000152,"raw_usage":{"total_tokens":1047,"prompt_tokens":759,"completion_tokens":288,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":503,"completion_tokens_details":{"reasoning_tokens":208}},"tokens_in":503,"tokens_out":288,"duration_ms":3102,"temperature":1.0,"reasoning_tokens":208,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T10:19:54.076979+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the identical differential-imaging protocol on the same microfiber with a particle-free buffer (e.g., filtered to 0.02 µm) and no protein; if bright spots still appear, or if spot counts do not scale predictably with protein concentration, the single-molecule interpretation collapses.","supporting_citations":[],"review_version":1}