{
 "dataset": {
  "archive": "dandi",
  "id": "000945",
  "label": "DANDI:000945"
 },
 "page": "datasets/dandi/000945/",
 "generated": "2026-11-05T07:41:23+00:00",
 "caution": "A failure to reproduce is not a finding that the paper is wrong. It may be a failure of the attempt.",
 "outcome_counts": {
  "reproduced": 1,
  "partially_reproduced": 1,
  "not_reproduced": 11,
  "blocked": 9
 },
 "outcome_counts_with_outside_data": {
  "reproduced": 0,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "outcome_counts_elsewhere": {
  "reproduced": 0,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "outcome_counts_links": {
  "reproduced": 0,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "attempts": [
  {
   "id": "000945-ramachandran-2025-01",
   "dataset": {
    "archive": "dandi",
    "id": "000945",
    "version": "0.250213.2212",
    "doi": "10.48324/dandi.000945/0.250213.2212"
   },
   "key": "dandi:000945",
   "label": "DANDI:000945",
   "dataset_version": "0.250213.2212",
   "paper": "10.1088/1741-2552/adbb1f",
   "paper_title": "Parameter-dependent cell-type specific effects of transcranial focused ultrasound stimulation in an awake head-fixed rodent model",
   "date": "2026-10-31",
   "kind": "agent",
   "accountable": "CatalystNeuro",
   "agent": {
    "harness": "Claude Code",
    "model": "claude-opus-5-5",
    "protocol": "pilot-13"
   },
   "runner": {
    "platform": "GitHub Actions",
    "url": "https://github.com/catalystneuro/encore/actions/runs/41243437657"
   },
   "tier": "runs",
   "deprecated": "",
   "deprecated_on": null,
   "entry": "000945-ramachandran-2025",
   "entry_title": "Cell-type specific responses to tFUS in awake and anesthetized rat S1 (Ramachandran et al. 2025)",
   "description": "Tests the findings of Ramachandran, Gao, Yttri, Yu and He (2025) on dandiset 000945, which holds sorted units and trial times of awake and anesthetized rat somatosensory cortex recordings during transcranial focused ultrasound at five pulse repetition frequencies and three duty cycles. It compares time-locked and delayed responses of regular and fast spiking units across parameters and states, and the onset latency of delayed responses. The sham condition and the second batch of rats are not in the dandiset.",
   "notebooks": [
    {
     "path": "reproduce_ramachandran_2025.ipynb",
     "title": "Reproducing Ramachandran et al. (2025) from dandiset 000945",
     "executed": "notebooks/000945-ramachandran-2025/reproduce_ramachandran_2025.html"
    }
   ],
   "outcomes": [
    {
     "claim": "fig3-cell-type-sorting",
     "statement": "Units are sorted into fast spiking units (FSU) and regular spiking units (RSU) by k-means clustering of the initial phase and after-hyperpolarization phase durations of their action potentials; in one animal's recordings this gave 28 FSU and 32 RSU, 60 in total.",
     "source": "Figure 3B and 3C; Methods 2.5",
     "quote": "K -means clustering of detected neurons from one animal's recordings, showing 28 fast spiking units (FSU) and 32 regular spiking units (RSU), for 60 total.",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Both the reproduced and the partially reproduced outcome of the criterion require the clusters to separate on the initial phase and after-hyperpolarization durations, and the files hold no waveforms, so K1 and K2 cannot run and neither outcome can be reached. With the dataset's labels (K0) two single recordings (R1) give all three counts within tolerance, BH527_30_200 with 28 FSU, 34 RSU and 62 units (given as measured) and BH508_4500_200_2 with 28 FSU, 30 RSU and 58 units; under Rep2 and Rep3 only BH527_30_200 does. No rat's series matches (R2; totals 77 to 616). The match is no more than chance, as 22 of 59 recordings have a total within 60 +- 6 and a binomial model with the pooled FSU share (0.505) expects 3.84 recordings to match all three counts, against 2 observed. FSU have the higher median baseline rate in 46 of 59 labelled recordings (8.60 against 5.84 Hz over units).",
     "deviations": [
      "Cell types are the dataset's labels (K0); that they are the paper's k-means output is assumed, and is consistent with the baseline rates without being shown.",
      "The other claims depend on this one for cell types. They use the K0 labels, as the claims file's header provides when the dataset carries labels, and do not need the waveform features whose absence blocks this claim, so they are tested."
     ],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig4-rsu-time-locked-both-states",
     "statement": "RSUs show time-locked responses to tFUS, an increase in firing rate during the 67 ms of stimulation, in both awake and anesthetized recordings.",
     "source": "Results 3.1, paragraph 1; Figure 4C and 4E",
     "quote": "We observed time-locked responses from RSU in both awake and anesthetized recordings, with an observable increase in firing rate during tFUS stimulation",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The criterion compares the 1500 Hz, 200 us RSU time-locked responses with the sham sessions of each state, and the dandiset holds no sham recording. The awake half already fails the condition that the 1500 Hz mean be above 100%, as RSU at 1500 Hz average 98.1 to 98.7% of baseline awake under every reading (n = 227 or 303), so the claim could not be reproduced whatever a sham would show; at best it would be partially reproduced, on the anesthetized half (106.4 to 107.0%, n = 78, two rats).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig4hi-fsu-delayed-latency",
     "statement": "FSU delayed responses to tFUS at 1500 Hz PRF begin later in awake rats (mean onset 0.51 s, spread 0.14 s) than under anesthesia (0.24 s, spread 0.08 s), and the onset distributions differ significantly by a Wilcoxon rank-sum test.",
     "source": "Results 3.1, paragraph 1; Figure 4H and 4I",
     "quote": "In the anesthetized model, these delayed responses had an average onset latency of 0.24 s \u00b1 0.08 s, while in the awake model they had an average onset latency of 0.51 s \u00b1 0.14 s ... Comparing the distributions using the Wilcoxon rank-sum test, we found that delayed response onset times varied significantly between awake and anesthetized models",
     "statistics": [
      {
       "test": "Wilcoxon rank-sum test",
       "p": 0.001,
       "p_relation": "<",
       "n": "FSU with a significant delayed response at 1500 Hz PRF, awake and anesthetized (counts not given)"
      },
      {
       "about": "as printed in the Figure 4 legend",
       "test": "Wilcoxon rank sum test",
       "p": 0.0001,
       "p_relation": "<"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Under none of the 24 readings (S1/S2, D1/D2, L1/L2, Rep1/Rep2/Rep3) are awake onsets later at p < 0.001; rank-sum p is 0.018 to 0.92, and in 8 readings the anesthetized mean is the later. Awake means are 0.40 to 0.58 s (within 0.51 +- 0.14), anesthetized means 0.38 to 0.52 s (all outside 0.24 +- 0.08). Measured values are the S1 D1 L1 Rep1 means (anesthetized, awake). 193 to 265 of 306 awake FSU (86% under S1 D1) and 34 or 39 of 50 anesthetized FSU have a delayed response under the paper's rule as the claims file reads it, and latencies spread over 0.07 to 1.39 s in both states, so the rule flags most units and the onsets look largely noise-driven. The anesthetized FSU come from two rats (BH498, BH512). Q1 and Q2 select the same recordings here.",
     "deviations": [
      "The anesthetized FSU come from two labelled rats where the paper had ten; batch 2 is not in the dandiset.",
      "T2 (505 trials) cannot be run; the files hold 500 trials, or fewer in five recordings."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig4h-awake-latency-wider",
     "statement": "Awake FSU delayed-response onset latencies have a wider spread than anesthetized ones and appear to have two peaks.",
     "source": "Results 3.1, paragraph 1; Figure 4H",
     "quote": "The awake model had a wider variance of onset latencies, and appears to have two peaks",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "methods_underspecified",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The claims file's readings give different results, and its header blocks a claim whose readings disagree. The awake SD exceeds the anesthetized SD under every D1 reading (0.384 to 0.396 s against 0.367 to 0.372 s) but is smaller under 10 of 12 D2 readings (0.381 to 0.393 s against 0.393 to 0.406 s) and larger by under 0.001 s in the other two. Under every reading the SDs are within 0.03 s of each other, with no contrast like the paper's 0.14 against 0.08 s. The awake histogram (S1 D1 L1 Rep1) has its largest count in the first allowed bin (61 of 262 at 0.071 s) and several smaller local maxima. Whether the first bin allowed is one starting at or after 67 ms (D1) or one ending after it (D2) is asked of the paper's authors.",
     "deviations": [],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5a-rsu-time-locked-prf-awake",
     "statement": "In awake rats PRF is a significant factor for RSU time-locked responses; 30 Hz and 300 Hz PRF give lower responses than every higher PRF (so 30 Hz is also below 300 Hz), 1500, 3000 and 4500 Hz do not differ, and sham differs from every PRF but 30 Hz.",
     "source": "Results 3.2, paragraph 1; Figure 5A",
     "quote": "Examining the time-locked response from RSU in the awake model using Kruskal\u2013Wallis ANOVA, we find that PRF is a significant factor ... Post hoc tests show that 30 Hz and 300 Hz PRF induced responses lower than all higher PRFs ... The sham condition was significantly different than all but the 30 Hz condition ... 1500 Hz, 3000 Hz, and 4500 Hz showed no significant differences",
     "statistics": [
      {
       "about": "PRF as a factor",
       "test": "Kruskal-Wallis",
       "p": 0.001,
       "p_relation": "<"
      },
      {
       "about": "30 and 300 Hz against each higher PRF",
       "test": "post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      },
      {
       "about": "sham against 300, 1500, 3000 and 4500 Hz",
       "test": "post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "PRF is a significant factor (Kruskal-Wallis p = 3.3e-16, W1 Rep1) but the post hoc pattern differs. The highest mean is at 3000 Hz (108.2%; 1500 Hz 98.5%), and 3000 Hz differs from 1500 and 4500 Hz, so part (4) fails under every reading; part (3) fails under every reading; part (2) holds only under W2 Rep1 and W2 Rep3. Parts (5) and (6) need the sham and were not tested; even if both held, part (1) with at most three others falls short of the four the partial outcome needs.",
     "deviations": [
      "No sham recording exists; parts (5) and (6) are reported as not tested, and a claim with an untested part is not judged reproduced.",
      "Ten awake recordings repeat a condition in the same rat; results are given with them (Rep1), without the `_2` recordings (Rep2) and with the earliest recording of each pair (Rep3)."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5c-rsu-time-locked-prf-anesth",
     "statement": "Under anesthesia RSU time-locked responses depend on PRF as in the awake model, the only difference being that 30 Hz and 300 Hz do not differ significantly.",
     "source": "Results 3.2, paragraph 1; Figure 5C",
     "quote": "The anesthetized model had similar statistical results, with the only difference being that no significant difference was observed between 30 Hz and 300 Hz PRF levels",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The Kruskal-Wallis is significant (p = 7.3e-08 W1, 7.7e-07 W2) and 30 and 300 Hz do not differ (part 7), but parts (2), (3) and (4) fail under both readings. 1500 Hz is the peak (107.0%) and differs from 4500 Hz; 30 and 300 Hz are not below 3000 or 4500 Hz. Even if the two sham parts held, part (1) with three others falls short of the five the partial outcome needs. With two rats PRF and rat are confounded, as the 30 Hz group is BH498 alone and the 1500 Hz peak comes from BH512 (110.4% against 101.7% in BH498, whose means span 100.0 to 101.7%).",
     "deviations": [
      "Two labelled rats (BH498 at five PRFs, BH512 without 30 Hz) where the paper had ten; BH521's anesthetized recordings have no labels.",
      "Parts (5) and (6) need the sham and were not tested."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "rsu-time-locked-anesthesia-factor",
     "statement": "Anesthesia is a significant factor in the RSU time-locked response by two-way ANOVA, and it lowers the average magnitude of RSU responses.",
     "source": "Results 3.2, paragraph 1; Discussion, paragraph beginning 'Although these results demonstrate'",
     "quote": "Two-way ANOVA additionally finds that anesthesia is a significant factor in the time-locked response from RSU ... Anesthesia was observed to decrease the average magnitude of RSU responses",
     "statistics": [
      {
       "test": "two-way ANOVA, anesthesia factor",
       "p": 0.01,
       "p_relation": "<"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Anesthesia is a factor at p = 0.0083 to 0.018 over the twelve readings (A1/A2, W1/W2, Rep1/Rep2/Rep3), below 0.01 only under W1 Rep3, and in the reverse direction under every reading; the anesthetized mean (102.6 to 102.9%) is above the awake mean (100.1 to 100.4%). The anesthetized units come from two rats and the awake units from seven, so the state effect is partly a rat effect; with only BH498 and BH512 it disappears (p = 0.425).",
     "deviations": [
      "Anesthetized units from two rats; awake units from seven."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5bd-fsu-time-locked-no-prf",
     "statement": "FSU time-locked responses show no significant differences across PRF levels, in either the awake or the anesthetized model.",
     "source": "Results 3.2, paragraph 1; Figure 5B and 5D",
     "quote": "FSU in both the awake and the anesthetized model show no significant differences due to PRF levels",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "By the paper's test FSU time-locked responses differ across PRF in both states, Kruskal-Wallis p = 1.8e-12 awake and 5.5e-04 anesthetized (W1 Rep1), with 4 to 6 significant post hoc pairs awake and 2 or 3 anesthetized under every reading. The effects are small; group medians span 98.7 to 103.4% awake and 99.8 to 103.9% anesthetized, and the awake 3000 Hz mean is 106.8%. G2 (with sham) cannot be run. The anesthetized groups come from two rats.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5e-rsu-delayed-prf-awake",
     "statement": "Awake RSU delayed responses (0.5 to 1 s) depend on PRF, are highest at 1500 Hz and 3000 Hz with lower responses at both lower and higher PRFs, and every PRF is significantly higher than sham.",
     "source": "Results 3.2, paragraph 2; Figure 5E",
     "quote": "RSUs in the awake rats had highest responses to 1500 Hz and 3000 Hz ... with a decrease at both higher and lower frequencies, and all conditions found to be significantly higher than sham",
     "statistics": [
      {
       "about": "1500 and 3000 Hz highest",
       "test": "post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      },
      {
       "about": "each PRF against sham",
       "test": "post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "PRF is a significant factor (p = 6e-20 Rep1, 4.5e-15 Rep2, 5.1e-15 Rep3) but the two highest means are 1500 and 30 Hz (106.7 and 106.6% Rep1), and 3000 Hz (102.3 to 103.3%) is below 30 Hz (corrected p < 1e-4); part (2) fails under both of its readings and every repeat reading. Part (3), every PRF above sham, needs the sham and cannot be tested. The criterion gives partially reproduced when (1) and (3) hold without (2), so the outcome would be partial if a sham held and not reproduced otherwise, and the data present do not decide between the two. The claim is therefore blocked as data missing, as fig4-rsu-time-locked-both-states is.",
     "deviations": [],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5f-fsu-delayed-prf-awake",
     "statement": "Awake FSU delayed responses depend on PRF, are strongest at 4500 Hz, differ between 30 Hz and the three highest PRFs and between 300 Hz and the two highest, and at 30 Hz and 300 Hz are not significantly higher than sham.",
     "source": "Results 3.2, paragraph 2; Figure 5F",
     "quote": "FSUs in the awake model show the strongest responses to 4500 Hz PRF, with significant differences between 30 Hz and the highest three PRFs, and 300 Hz and the highest two PRFs ... 30 Hz and 300 Hz were not found to be significantly higher than sham",
     "statistics": [
      {
       "test": "post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "PRF is a significant factor (p = 4.6e-12 Rep1) and 30 Hz differs from 1500, 3000 and 4500 Hz (part 3, tested without direction as the criterion says \"differs\"), but the strongest mean is at 30 Hz (103.5%, 4500 Hz 100.3 to 100.4%), the opposite of the paper's pattern, and 300 Hz does not differ from 4500 Hz (corrected p = 1.0), under every repeat reading. Part (5) needs the sham and was not tested; even if it held, part (1) with two others falls short of the three the partial outcome needs.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5gh-delayed-prf-anesth",
     "statement": "Under anesthesia PRF is a significant factor for the delayed responses (0.2 to 0.7 s) of both RSUs and FSUs, which broadly prefer higher PRFs.",
     "source": "Results 3.2, paragraph 2; Figure 5G and 5H",
     "quote": "PRF was found to be a significant factor for RSU and FSU in both awake and anesthetized settings ... Both RSUs and FSUs in the anesthetized rats had similar sensitivity as their time-locked responses, broadly preferring higher PRFs",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "PRF is not a significant factor for anesthetized delayed responses, Kruskal-Wallis p = 0.16 (W1) and 0.12 (W2) for RSU, 0.16 and 0.18 for FSU. The higher PRFs have the higher pooled mean by under one point (RSU 99.97 against 99.15%, FSU 99.87 against 99.35%). Every rat and PRF mean is within 98.5 to 100.9%, and the 30 Hz group is BH498 alone.",
     "deviations": [
      "Two labelled rats where the paper had ten."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "delayed-anesthesia-factor",
     "statement": "Anesthesia is a significant factor in the delayed response for both RSUs and FSUs by two-way ANOVA, and it lowers the average response magnitude.",
     "source": "Results 3.2, paragraph 2; Discussion, paragraph beginning 'Although these results demonstrate'",
     "quote": "Using two-way ANOVA, we found that anesthesia is a significant factor in the delayed response for both RSU ... and FSU",
     "statistics": [
      {
       "about": "RSU",
       "test": "two-way ANOVA, anesthesia factor",
       "p": 0.001,
       "p_relation": "<"
      },
      {
       "about": "FSU",
       "test": "two-way ANOVA, anesthesia factor",
       "p": 0.01,
       "p_relation": "<"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "With one recording per rat and condition, either without the `_2` recordings (Rep2) or with the earliest of each pair (Rep3), anesthesia is a factor for RSU at p = 4.2e-09 to 1.8e-08 and for FSU at p = 0.0012 to 0.0015 under A1/A2 and W1/W2, and the anesthetized mean is lower in both (RSU 99.7 against 104.3 to 104.6%, FSU 99.7 against 101.2 to 101.3%). With every recording (Rep1) RSU holds (p = 3.6e-06 to 4.3e-06) but FSU gives p = 0.038 to 0.041, which misses the 0.01 bound; that reading gives partially reproduced. Leaving out the recording filed under both BH498 and BH508 gives FSU p = 0.0016 to 0.0019 under Rep2 and Rep3, and 0.048 to 0.052 under Rep1 (above 0.05 under W2), where the RSU part alone still holds and the criterion gives partially reproduced (\"when one cell type holds\"). With every recording the claim is thus partial at best, so the reproduced outcome rests on the assumption. The anesthetized units come from two rats, so the state effect may also be a rat effect. The question is recorded for the paper's authors.",
     "deviations": [
      "Assumed that the paper's analysis used one recording per rat and condition. The paper mentions no repeats, Methods 2.3 fits one recording per parameter set per rat without stating it (\"Recordings for a single rat for each set of parameters were completed on the same day\" says when a rat's recordings were made, so it supports the assumption only weakly), and the dandiset's description says \"Some subjects have duplicate recordings taken a few months after the original ones\". Which recording of a pair is the original is open, as three `_2` recordings are earlier than their pair; both choices (Rep2, Rep3) give the same outcome.",
      "Anesthetized units from two rats."
     ],
     "assumed": true,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig6ad-time-adjacent-batch",
     "statement": "In the five rats recorded awake and anesthetized in the same week, time-locked responses to the low PRF and 1500 Hz differ significantly in both states, with a reduced response under anesthesia.",
     "source": "Results 3.2, paragraph 3; Figure 6A to 6D",
     "quote": "In analysis of the time locked responses, we observed significant differences in response to 300 Hz PRF and 1500 Hz PRF stimulation in both awake and anesthetized conditions ... but with a reduced response under anesthesia",
     "statistics": [
      {
       "test": "Kruskal-Wallis with post hoc Wilcoxon",
       "p": 0.05,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": null,
     "notes": "The second batch of five rats recorded awake and anesthetized in the same week, with a sham, is not in the dandiset. Its ten subject folders hold eight rats named in the identifiers, all of which have recordings of the first batch's series, and no file is from a two-PRF or sham protocol. The session dates show no rat recorded awake and anesthetized within a week at two PRFs only; BH498's awake recordings (5 and 6 March 2024) precede its anesthetized ones (10 March) by four to five days, but at all five PRFs and without a sham.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig6ef-rsu-ispta-rise-then-fall",
     "statement": "Awake RSU mean responses, time-locked and delayed, rise with ISPTA up to around 60 mW/cm2 and then decline.",
     "source": "Results 3.2, paragraph 4; Figure 6E and 6G",
     "quote": "We observe that RSU's in both their time-locked and delayed responses initially increase till around 60 mW cm \u22122 , after which the response declines.",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "methods_underspecified",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The delayed means do not rise to their peak under any reading (106.6% at 30 Hz, 101.2 to 102.8% at 300 Hz, 106.7 to 108.8% at 1500 Hz), so at best the claim is partially reproduced, on the time-locked window. Time-locked RSU means rise from 30 Hz (96.4 to 96.8%) to a peak at 3000 Hz (60.26 mW/cm2, 107.7 to 108.7%) and fall at 4500 Hz (98.0 to 99.5%) under W2 with every repeat reading and under W1 with Rep1 and Rep3, but under W1 Rep2 the 300 Hz mean (98.2%) is 0.1 point above the 1500 Hz mean (98.1%, SEMs 1.1 and 1.3). With one recording per rat and condition the claims file's readings W1 and W2 thus disagree under Rep2, and its header blocks a claim whose readings disagree. How a window off the bin edges was measured is asked of the paper's authors.",
     "deviations": [],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig6fh-fsu-ispta",
     "statement": "Awake FSUs show no time-locked response, and their delayed response increases almost linearly with ISPTA.",
     "source": "Results 3.2, paragraph 4; Figure 6F and 6H",
     "quote": "FSU's show no time-locked response, and their delayed response increases almost linearly with I SPTA .",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The time-locked part fails (fig5bd, awake Kruskal-Wallis p = 1.8e-12). Delayed FSU means are 103.5, 98.7, 101.8, 100.6 and 100.3% from 30 to 4500 Hz (Spearman rho = -0.4, Pearson r = -0.27, Rep1; rho = -0.3 Rep2 and Rep3).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig5a-1500-highest-time-locked-mean",
     "statement": "Among PRFs, 1500 Hz gives the highest mean RSU time-locked response in awake rats, although not significantly above 3000 and 4500 Hz.",
     "source": "Results 3.2, paragraph 4; Figure 5A",
     "quote": "Although it is not statistically significant, we found that 1500 Hz produces the highest mean time-locked response",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The highest mean is at 3000 Hz under every reading (107.7 to 108.7%), significantly above 1500 Hz (98.1 to 98.7%; corrected p = 0.0040 or less).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig7a-rsu-time-locked-dc",
     "statement": "Awake RSU time-locked responses at 1500 Hz PRF are significantly lower at 6% DC than at 60% DC (and, by the Figure 7A legend, than at 30% DC), and every DC is significantly higher than sham.",
     "source": "Results 3.3, paragraph 1; Figure 7A and its legend",
     "quote": "Time-locked responses from RSUs showed a significantly lower response to 6% DC than those to 60%, with all conditions significantly higher than sham",
     "statistics": [
      {
       "test": "Kruskal-Wallis with post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Part (2) does not need the 30% group and fails under every reading, as 6% is the highest group (180.8%; 60% 147.5%) and does not differ from 60% (corrected p = 1.0; 0.67 to 0.91 without a 30% group). Part (1), the Kruskal-Wallis over 6, 30 and 60%, needs the 30% group (p = 0.67 and 0.91 without it, 1.5e-22 or smaller with a substitute, so the substitutes alone decide it), part (4), 6% below 30%, needs it too, and part (3) needs the sham, so parts (1), (3) and (4) are untested. The criterion gives partially reproduced when (1) and at least one of (2) to (4) hold, so partial is reachable if (1) and (3) or (4) held, and the data present do not decide it. The claim is therefore blocked as data missing, as fig7c-rsu-delayed-dc is. The substitute 30% groups are far below both (90.4 to 98.7%). The 6% and 60% responses are largest in BH506 (249.0 and 209.9%); BH521 and BH526 give 113 to 121%.",
     "deviations": [
      "The dandiset has no 30% DC recording of its own. The substitutes (C1, the PRF series' 1500 Hz, 200 us recordings of BH506 and BH526; C2, those of every rat, with each repeat reading) and C0, no 30% group, are shown; a part that needs the 30% group is treated as data missing.",
      "The 6% and 60% recordings come from three rats where the paper had ten."
     ],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig7a-60-highest-time-locked-mean",
     "statement": "By mean values 60% DC gives the strongest RSU time-locked response, though it is only significantly higher than 6%.",
     "source": "Results 3.3, paragraph 1; Figure 7A",
     "quote": "Based on mean values, 60% DC appears to produce the strongest time-locked response, although it is only statistically significantly higher than the 6% condition.",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "The highest mean is at 6% under every reading (176.3 to 180.8%; 60% 147.1 to 147.5%), which does not need the 30% group. With the substitute 30% groups, 60% is significantly above 30% (corrected p < 1e-16).",
     "deviations": [
      "The 30% group is a substitute (C1 or C2) or absent (C0), as for fig7a."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig7b-fsu-time-locked-dc",
     "statement": "FSUs show no time-locked response across DC levels; no significant differences are observed.",
     "source": "Results 3.3, paragraph 1; Figure 7B",
     "quote": "no time-locked response was observed as in the PRF study",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "not_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "FSU time-locked responses differ across DC under every reading, and the 6% against 60% difference does not need the 30% group (Kruskal-Wallis p = 0.0054 without it; p = 3.5e-12 or smaller with a substitute, all three pairs significant). FSU respond strongly at 6% and 60% (means 156.8 and 205.3%, medians 115.0 and 128.6%) and not at the substitute 30% (median 100.8%), most in BH506 (338.6 and 403.3%).",
     "deviations": [
      "The 30% group is a substitute (C1 or C2) or absent (C0), as for fig7a."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig7c-rsu-delayed-dc",
     "statement": "Awake RSU delayed responses prefer 30% DC over 6% and 60%, all three DCs differ (Figure 7C legend), and at 60% DC they do not differ from sham.",
     "source": "Results 3.3, paragraph 1; Figure 7C and its legend",
     "quote": "Delayed responses however preferred 30% DC over 6% and 60% ... and 60% DC showed no significant difference from sham",
     "statistics": [
      {
       "test": "Kruskal-Wallis with post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "data_missing",
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Part (2), 30% above 6% and 60%, needs the paper's 30% DC recordings, which are not in the dandiset, and the substitutes decide it. With the same rats' 1500 Hz, 200 us recordings (C1) the Kruskal-Wallis is not significant (p = 0.21) and no pair differs; with every such recording (C2) the effect is significant (p = 2.2e-04 Rep1, 3.7e-05 Rep2, 1.9e-04 Rep3) and 30% is above 6% and 60%. Part (4), 6% against 60%, also depends on the 30% group through the Bonferroni family; 6% is below 60% at p = 0.024 alone and at corrected p = 0.071 over three pairs. Part (3) needs the sham. The C1 substitutes (BH506, BH526) were made three to five months before the same rats' DC recordings; C2 adds recordings from August 2023 to 6 March 2024, among them BH498's of 5 and 6 March, three to four days before the DC recordings of 9 March.",
     "deviations": [
      "The 30% group is a substitute (C1 or C2) or absent (C0), as for fig7a."
     ],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig7d-fsu-delayed-dc",
     "statement": "Awake FSU delayed responses are significantly higher at 60% DC than at 6%, and at 6% DC they do not differ from sham.",
     "source": "Results 3.3, paragraph 1; Figure 7D",
     "quote": "Delayed responses from FSUs showed a significantly higher response to 60% DC over 6% ... with 6% DC found to be not significantly different from sham",
     "statistics": [
      {
       "test": "Kruskal-Wallis with post hoc Wilcoxon",
       "p": 0.001,
       "p_relation": "<",
       "corrected": "Bonferroni"
      }
     ],
     "kind": "finding",
     "inferred": "",
     "outcome": "partially_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_ramachandran_2025.ipynb",
     "notes": "Parts (1) and (2) hold under every reading, without a 30% group included, so they do not depend on the substitute; the Kruskal-Wallis is significant (p = 0.0051 without a 30% group, 0.011 C1, 4.5e-07 to 1.4e-06 C2) and 60% is above 6% (p = 0.0051 alone, corrected p = 0.015 over three pairs). Part (3), 6% against sham, cannot be tested, so the claim is partially reproduced, the outcome the criterion gives when (1) and (2) hold without (3). The difference is small and both groups are below baseline (99.6% and 98.2%).",
     "deviations": [
      "No sham recording; part (3) is reported as not tested.",
      "The 30% group is a substitute (C1 or C2) or absent (C0); it enters only the Kruskal-Wallis."
     ],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "sham-no-auditory-response",
     "statement": "The off-target sham at 1500 Hz PRF produced no significant response, so the activation is not attributed to an auditory or skull-conduction confound.",
     "source": "Discussion, paragraph beginning 'One potential confound'; Methods 2.3",
     "quote": "In this study we additionally performed a sham condition delivering ultrasound off target to the front of the skull, and did not observe any significant response.",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "methods_underspecified",
     "notebook": null,
     "notes": "The criterion blocks the claim, as the paper names no test of the sham against baseline. The dandiset also holds no sham recording, so the sham's responses cannot be shown either.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    }
   ],
   "findings": [
    {
     "kind": "data_missing",
     "statement": "The sham condition and the second batch of five rats (awake and anesthetized recordings in the same week at a low PRF and 1500 Hz) are not in the dandiset. Four claims cannot be tested for this reason (fig4-rsu-time-locked-both-states, fig5e-rsu-delayed-prf-awake, fig6ad-time-adjacent-batch, and sham-no-auditory-response, which its criterion also blocks), fig7a-rsu-time-locked-dc needs the sham as well as the 30% DC recordings, and every comparison with sham in the others is left out.",
     "found": null,
     "status": null
    },
    {
     "kind": "data_missing",
     "statement": "Only part of the first batch is here. The awake PRF series covers seven rats, the anesthetized PRF series three (BH498 at five PRFs, BH512 without 30 Hz, BH521 at three PRFs, unlabelled), and the awake DC series three rats at 6% and 60% only. The DC series' own 30% recordings are missing; the only 1500 Hz, 200 us recordings are the PRF series', those of the same rats made three to five months earlier and those of the other rats from August 2023 to 6 March 2024, three to four days before the DC recordings of 9 March, and the claims on RSU time-locked and delayed responses across DC (fig7a-rsu-time-locked-dc, fig7c-rsu-delayed-dc) cannot be tested without them. The paper describes ten rats for each series.",
     "found": null,
     "status": null
    },
    {
     "kind": "data_missing",
     "statement": "The files hold no waveforms, voltage traces or electrode positions, so the cell-type sorting (k-means on the initial phase and after-hyperpolarization durations) cannot be repeated or checked.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_error",
     "statement": "Eleven files are second copies of other recordings. The five anesthetized BH498 recordings are deposited under BH498 without labels and under subject folder BH497 with labels; four anesthetized BH512 recordings under BH512 without labels and under BH514 with labels; BH512_1500_200_anes2 is a third copy of the BH512 1500 Hz anesthetized recording, sorted differently and dated two months later; BH508_30_200_2 is the subject of the next finding. Trial times, per-unit spike counts and first spike times are identical within each of the nine pairs; the third BH512 copy shares the trial times and first spike times, but not the spike counts (905,554 against 926,883). The identifiers name BH498 and BH512.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_error",
     "statement": "BH498_30_200 (sub-BH498) and BH508_30_200_2 (sub-BH508) share all 203 trial times and 18,860 spike times; they are one recording sorted twice and filed under two rats, with session start times 6.5 hours apart (6 March 2024 17.58 and 7 March 00.23). Which rat it belongs to cannot be established from the files.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_error",
     "statement": "The dandiset's description says the repeat recordings were \"taken a few months after the original ones\", but three of the ten `_2` recordings start before their unsuffixed pair (BH498_1500_200_2 and BH498_4500_200_2 one day before, BH508_1500_200_2 two months before), so the suffix does not mark which recording is the repeat.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_error",
     "statement": "The labelled copies of anesthetized recordings (sub-BH497, sub-BH514) are described as \"Awake S1 Stimulation by tFUS\". Every file gives the age P24W, where the paper gives 7, 9 and 10 months for the three series and the dandiset's description 8 to 10 and 12 months. Electrode columns carry placeholder descriptions (\"my description\") and the location \"brain area\".",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_insufficient",
     "statement": "The PRF, tone-burst duration and anesthesia state of a recording are coded only in the file's identifier. The trial table's stop_time (always 2.2 s after start_time) is not described, and the files do not say which trials of the 505 delivered the 500 are.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_insufficient",
     "statement": "Fifteen anesthetized files have no cell-type labels, among them all of BH521's and BH512's DC recordings, which cannot enter any cell-type analysis.",
     "found": null,
     "status": null
    },
    {
     "kind": "other",
     "statement": "Five recordings hold fewer than the 500 trials the Methods and the dandiset's description give (203, 238, 328 and 339 trials, all from BH508 or the BH498/BH508 duplicate).",
     "found": null,
     "status": null
    },
    {
     "kind": "methods_unavailable",
     "statement": "Were the repeat recordings (BH498, BH506, BH508) part of the paper's analysis, and if not, which recording of each pair was? Methods 2.3 describes one recording per parameter set per rat. The FSU part of delayed-anesthesia-factor depends on it (p = 0.0012 to 0.0015 with one recording per condition, 0.038 to 0.041 with all), and fig6ef-rsu-ispta-rise-then-fall depends on which recording of a pair is kept. This attempt assumed one recording per rat and condition.",
     "found": null,
     "status": null
    },
    {
     "kind": "methods_unavailable",
     "statement": "Was a response window that does not fall on the 35.7 ms bin edges (0 to 67 ms, 0.2 to 0.7 s) measured from exact spike times or from the PSTH bins that overlap it? The two give different results for fig6ef-rsu-ispta-rise-then-fall, which is blocked until this is known.",
     "found": null,
     "status": null
    },
    {
     "kind": "methods_unavailable",
     "statement": "For the delayed-response latency, is the first bin allowed after the 67 ms of ultrasound the first bin starting at or after 67 ms, or the first ending after it? The two give different results for fig4h-awake-latency-wider, which is blocked until this is known.",
     "found": null,
     "status": null
    }
   ],
   "reconciliation": null,
   "verifications": [
    {
     "date": "2026-10-31",
     "kind": "agent",
     "by": "Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook)",
     "reviewed": "The handover of the first run, the executed notebook with the conclusions and summary removed, the helper module and the claims file, with the paper under paper/. The verifier also checked the archive's asset list.",
     "agreed": 17,
     "of": 22,
     "disagreements": [
      {
       "claim": "fig3-cell-type-sorting",
       "outcome": "blocked",
       "reason": "The files hold no waveforms, so the k-means features and the cluster separation the criterion requires cannot be computed. Using the deposited labels, 2 of 59 recordings fall within the count tolerances; this is weak evidence, since recording totals run from 31 to 127 units. One of the two, BH508_4500_200_2, is a 12-month repeat that Rep2 would drop, and no base rate is given for a match among 59 recordings."
      },
      {
       "claim": "fig4h-awake-latency-wider",
       "outcome": "blocked",
       "reason": "Awake SD > anesthetized SD under D1 but not under D2, and the claims header blocks a claim when its readings disagree. Every SD is about 0.37 to 0.41 s, with no contrast like the paper's 0.14 vs 0.08. The notebook overrides the claims file's rule for disagreeing readings."
      },
      {
       "claim": "fig6ef-rsu-ispta-rise-then-fall",
       "outcome": "blocked",
       "reason": "The delayed window fails under every reading; 30 Hz (106.6) is above 300 Hz (101.2). The time-locked window holds under W2 but fails under W1 Rep2 by 0.1% (98.2 vs 98.1), so the window readings disagree. At best it is partial."
      },
      {
       "claim": "fig7c-rsu-delayed-dc",
       "outcome": "blocked",
       "reason": "The paper's own 30% DC recordings are not in the dandiset. The substitute 30% condition decides the outcome; C1 is not reproduced (KW p = 0.21) and C2 is partial. Part (4), 6% vs 60%, fails under every reading (corrected p = 0.071). The notebook adopts C1 by its own choice, where the honest outcome for 30%-dependent parts is data_missing."
      },
      {
       "claim": "fig7d-fsu-delayed-dc",
       "outcome": "partially_reproduced",
       "reason": "Parts (1) and (2) hold under every reading (6% vs 60% corrected p = 0.015). Part (3), the sham comparison, cannot be tested, and a criterion with an untested part should not reach \"reproduced\"; the header only covers a missing anesthetized sham. The effect is tiny, 98.2% vs 99.6%, both below baseline."
      }
     ]
    },
    {
     "date": "2026-10-31",
     "kind": "agent",
     "by": "Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook)",
     "reviewed": "The full executed notebook of the first revision, with its conclusions and summary, the helper module, the claims file and the attempt record, with the paper under paper/.",
     "agreed": 20,
     "of": 22,
     "disagreements": [
      {
       "claim": "fig5e-rsu-delayed-prf-awake",
       "outcome": "blocked",
       "reason": "The criterion gives partially reproduced when (1) and (3) hold without (2). Part (1) holds, part (2) fails and part (3), every PRF above sham, is untested, so if a sham held the outcome would be partial. Elsewhere the notebook settles a claim only when the missing parts cannot change its outcome, and fig4-rsu, whose best outcome is also partial, is blocked as data_missing. fig5e should be blocked as data_missing too, or the record should explain why not. The verifier agreed with not reproduced in round 1, and calls that its error."
      },
      {
       "claim": "fig7a-rsu-time-locked-dc",
       "outcome": "blocked",
       "reason": "Partial needs (1) and at least one of (2) to (4). Part (2) fails. Part (1), the Kruskal-Wallis over 6, 30 and 60%, and part (4), 6% below 30%, need the missing 30% group, and part (3) needs the sham, so all three are untested. The conclusion \"whatever those parts would show, the claim cannot be reproduced\" rules out reproduced, not partial. Cell 13's own rule, that a claim whose outcome depends on a 30% part is blocked as data_missing, applied consistently blocks fig7a. The verifier agreed with not reproduced in round 1, and calls that its error."
      }
     ]
    },
    {
     "date": "2026-10-31",
     "kind": "agent",
     "by": "Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook)",
     "reviewed": "The two outcome changes proposed after round 2 (fig5e-rsu-delayed-prf-awake and fig7a-rsu-time-locked-dc) and the notebook analysis behind them, with the claims file and the attempt record.",
     "agreed": 2,
     "of": 2,
     "disagreements": []
    }
   ]
  }
 ],
 "entries": [
  {
   "name": "000945-ramachandran-2025",
   "type": "reproduction",
   "title": "Cell-type specific responses to tFUS in awake and anesthetized rat S1 (Ramachandran et al. 2025)",
   "description": "Tests the findings of Ramachandran, Gao, Yttri, Yu and He (2025) on dandiset 000945, which holds sorted units and trial times of awake and anesthetized rat somatosensory cortex recordings during transcranial focused ultrasound at five pulse repetition frequencies and three duty cycles. It compares time-locked and delayed responses of regular and fast spiking units across parameters and states, and the onset latency of delayed responses. The sham condition and the second batch of rats are not in the dandiset.",
   "datasets": [
    "dandi:000945"
   ],
   "papers": [
    "10.1088/1741-2552/adbb1f"
   ],
   "derived_from": [],
   "notebooks": [
    {
     "path": "reproduce_ramachandran_2025.ipynb",
     "title": "Reproducing Ramachandran et al. (2025) from dandiset 000945",
     "executed": "notebooks/000945-ramachandran-2025/reproduce_ramachandran_2025.html"
    }
   ]
  }
 ]
}