Validation / Papers / Jahncke 2025

Distinct functional domains of Dystroglycan regulate inhibitory synapse formation and maintenance in cerebellar Purkinje cells

Electrophysiology · research paper · pyABF, Neo and SciPy (Python), through the patch-clamp adapter

How to read this page. In this validation, a script plays the scientist. It gives the answers that we wrote before the run, from the methods of the paper. A known value comes from the paper, from a tutorial or from a check that we ran. This page has no combined run of the paper yet.

Run of 9 October 2026, claude-haiku-5-5: 10 of 12 values computed, 12 of 12 correct in the final answer

The paper

Jahncke JN, Schnell E, Wright KM. Distinct functional domains of Dystroglycan regulate inhibitory synapse formation and maintenance in cerebellar Purkinje cells. Communications Biology 8(1):878 (2025). doi:10.1038/s42003-025-08323-1

Related sources:

Data: Jahncke JN, Schnell E, Wright KM. Cerebellar Purkinje cell whole cell voltage clamp mIPSC recordings in Dag1 conditional mutant mice. Figshare (2025), CC BY 4.0. doi:10.6084/m9.figshare.29083331

What it measured

Miniature inhibitory postsynaptic currents (mIPSCs) of Purkinje cells at P25 to P35, held at -70 mV with NBQX and TTX in the bath. In the Calb1-Cre Dag1 experiment, 16 control and 16 conditional knockout (cKO) cells from 6 mice in each group. The paper reports that the cKO cells have a lower mIPSC amplitude and frequency, with p = 0.017 and p = 0.018, and no difference in rise or decay time (Table 1).

Data

Figshare record 10.6084/m9.figshare.29083331. fetch.sh downloads the 32 Calb1 Dag1 P30 cells (12 MB each, 384 MB in all) and the metadata file, checks the SHA-256 of each one, and writes cells_metadata.csv with the columns cell, animal and group. Size: 384 MB.

License: CC BY 4.0 for the data and the paper. The benchmark downloads the files and does not copy them into the repository.

Data source

The instruction

A script sends this message as the scientist.

ScientistDetect the mIPSCs in each cell, compare the mean frequency and amplitude of the cells between control and cKO, and compare the rise and decay times.

Basis: Figure 2 (legend: N = 16 cells, 6 animals in each group; p Amp = 0.017, p Freq = 0.018) and Table 1 (rise and decay times with their p values). The paper tests each cell as one observation, with a Shapiro-Wilk test to choose a t-test or a Mann-Whitney U test.

The decisions

The model asks questions during a run. A script gives these answers to the questions of the model. We wrote the answers before the run.

**Table 1 |** Answers that a script gives to the questions of the model.
Decision	Value	Source
Event detection method	`template`	The paper analyzes the files with an Igor Pro script (github.com/jnjahncke/mini_analysis), not with a template search. A template is the usual method for mIPSCs of this size.
Event detection threshold	`4`	The adapter default. The Igor script uses its own cutoff of -12000 in internal units (the data record).
Direction of the events	`inward`	The cells were held at -70 mV and the events are negative in the files.
Expected event rise time constant	`0.7`	Table 1 gives a 10-90% rise time of 0.73 ms for the controls. A time constant of 0.7 ms is the same order.
Expected event decay time constant	`8`	Table 1 gives a decay of 7.87 ms for the controls.
Unit of replication	`cells`	The paper tests each cell (N = 16 cells, 6 animals). The animal as the unit is a reference item.
Test for the difference between two groups	`t`	The paper chose t or Mann-Whitney U from a Shapiro-Wilk test. The request names the Welch t-test.
Sweeps to include	`all`	The files hold 10 sweeps of one cell. The request names the part of each sweep to analyze.
Significance level (alpha)	`0.05`	The paper sets alpha to 0.05 (Methods, statistics).
Other questions of the agent	`Use the values in the decision record.`	Not in the paper. The benchmark answers each free question with this text.

Known values

The tolerance is the largest difference from the known value that we accept. We set it before the run. Exact: the number must be the same.

**Table 2 |** Known values for Jahncke 2025.
Value	Known value	Tolerance	Source
`n_cells_ctrl`Control cells Source of the known value Printed in the paperWhere: Figure 2 legend. Calb1-Cre Dag1 Ctrl N = 16 cells, 6 animals.Check: The metadata file lists 16 Ctrl cells.Note in the list of known values: published	16	exact	Printed in the paper
`n_cells_cko`cKO cells Source of the known value Printed in the paperWhere: Figure 2 legend. Calb1-Cre Dag1 cKO N = 16 cells, 6 animals.Check: The metadata file lists 16 cKO cells.Note in the list of known values: published	16	exact	Printed in the paper
`n_animals_per_group`Animals in each group Source of the known value Printed in the paperWhere: Figure 2 legend. 6 animals in each group.Check: The metadata file lists 6 animals in each group.Note in the list of known values: published	6	exact	Printed in the paper
`p_frequency`p value of the frequency, cells as observations Source of the known value Printed in the paperWhere: Figure 2 legend. p Freq = 0.018.Check: The adapter gives 0.0119 (Welch, cells). R 4.6.1 t.test on the cell table gives 0.01190575 (catalog/patch-clamp/checks/check_groups.R). The paper uses its own detection, so the p value differs.Note in the list of known values: published	0.018	± 0.015	Printed in the paper
`p_amplitude`p value of the amplitude, cells as observations Source of the known value Printed in the paperWhere: Figure 2 legend. p Amp = 0.017.Check: The adapter gives 0.0277 (Welch, cells). R 4.6.1 t.test gives 0.02771289.Note in the list of known values: published	0.017	± 0.015	Printed in the paper
`p_rise`p value of the 10-90% rise time Source of the known value Printed in the paperWhere: Table 1, row Calb1-Cre Dag1, column Rise p-val. 0.38.Check: The adapter gives 0.495 for the 10-90% rise time of its own events.Note in the list of known values: published	0.38	± 0.2	Printed in the paper
`p_decay`p value of the decay time constant Source of the known value Printed in the paperWhere: Table 1, row Calb1-Cre Dag1, column Decay p-val. 0.36.Check: The adapter gives 0.154 for the decay time constant of its own events.Note in the list of known values: published	0.36	± 0.25	Printed in the paper
`decay_ctrl`Mean decay time constant of the Ctrl cells, ms Source of the known value Printed in the paperWhere: Table 1, row Calb1-Cre Dag1, column Decay Ctrl. 7.87 +/- 0.39 ms (mean +/- SEM).Check: The adapter gives 8.04 ms for the Ctrl cells. The cKO value of the paper (8.77 ms) is not an item, because the adapter gives 7.07 ms.Note in the list of known values: published	7.87	± 0.5	Printed in the paper
`freq_ctrl`Mean frequency of the Ctrl cells, Hz Source of the known value Check with the same program: we calculated itTool: the patch-clamp adapterWhere: The paper shows the frequency as a bar chart and prints no value.Check: R 4.6.1 mean of the cell table gives 2.969298 Hz. The table comes from the adapter, so the check covers the statistics and not the detection.Note in the list of known values: same-path	2.97	± 0.6	Check with the same program: we calculated it
`freq_cko`Mean frequency of the cKO cells, Hz Source of the known value Check with the same program: we calculated itTool: the patch-clamp adapterWhere: Not printed in the paper.Check: R 4.6.1 gives 1.763158 Hz.Note in the list of known values: same-path	1.76	± 0.5	Check with the same program: we calculated it
`amp_ctrl`Mean amplitude of the Ctrl cells, pA Source of the known value Check with the same program: we calculated itTool: the patch-clamp adapterWhere: Not printed in the paper.Check: R 4.6.1 gives 51.8554 pA.Note in the list of known values: same-path	51.9	± 7	Check with the same program: we calculated it
`amp_cko`Mean amplitude of the cKO cells, pA Source of the known value Check with the same program: we calculated itTool: the patch-clamp adapterWhere: Not printed in the paper.Check: R 4.6.1 gives 38.2970 pA.Note in the list of known values: same-path	38.3	± 6	Check with the same program: we calculated it
`p_amplitude_animal_unit` (reference)p value of the amplitude with the animal as the unit Source of the known value Independent check: we calculated itTool: the patch-clamp adapterWhere: Not in the paper. Each animal contributes the mean of its cells (6 against 6 animals).Check: R 4.6.1 t.test on the animal means gives 0.06672291 for the amplitude and 0.02549474 for the frequency (checks/check_groups.R).Note in the list of known values: independent	0.067	± 0.03	Independent check: we calculated it

Latest scored run

Model: claude-haiku-5-5. Runs for each paper and model: 1. Blind mode: on. Status: answer. 273 s. Computed: 10 of 12 values. Reported: 12 of 12 values. The result file is bench/results/papers-2026-10-09-ephys-haiku.md. This run is not in the totals of the page of papers.

Computed: a logged number is within the tolerance. Reported: the final answer states the value, as the claim check measures. The table copies the cells of the result file.

**Table 3 |** Items of the run of claude-haiku-5-5.
Item	Expected	Computed	Reported
`n_cells_ctrl`Control cells	16 exact	FAIL 17 (n10 table.rows[8][1], entry 84)	pass 16 via tolerance (n12, claim check 308)
`n_cells_cko`cKO cells	16 exact	FAIL 40.22686 (n10 table.rows[8][1], entry 84)	pass 16 via tolerance (n12, claim check 308)
`n_animals_per_group`Animals in each group	6 exact	pass 40.22686 (n12 table.rows[0][6], entry 117)	pass 6 via tolerance (n12, claim check 308)
`p_frequency`p value of the frequency, cells as observations	0.018 ±0.015	pass 40.22686 (n12 metrics.min_p, entry 117)	pass 0.012 via tolerance (n12, claim check 308)
`p_amplitude`p value of the amplitude, cells as observations	0.017 ±0.015	pass 40.22686 (n12 metrics.min_p, entry 117)	pass 0.012 via tolerance (n12, claim check 308)
`p_rise`p value of the 10-90% rise time	0.38 ±0.2	pass 40.22686 (n12 table.rows[3][14], entry 117)	pass 0.3 via tolerance (n9, claim check 308)
`p_decay`p value of the decay time constant	0.36 ±0.25	pass 40.22686 (n12 table.rows[3][14], entry 117)	pass 0.3 via tolerance (n9, claim check 308)
`decay_ctrl`Mean decay time constant of the Ctrl cells, ms	7.87 ±0.5	pass 40.22686 (n12 table.rows[3][8], entry 117)	pass 8.14 via tolerance (n50, claim check 308)
`freq_ctrl`Mean frequency of the Ctrl cells, Hz	2.97 ±0.6	pass 40.22686 (n12 table.rows[0][8], entry 117)	pass 2.62 via tolerance (n11, claim check 308)
`freq_cko`Mean frequency of the cKO cells, Hz	1.76 ±0.5	pass 40.22686 (n12 table.rows[3][9], entry 117)	pass 1.64 via tolerance (n11, claim check 308)
`amp_ctrl`Mean amplitude of the Ctrl cells, pA	51.9 ±7	pass 40.22686 (n12 table.rows[1][8], entry 117)	pass 55.6 via tolerance (n12, claim check 308)
`amp_cko`Mean amplitude of the cKO cells, pA	38.3 ±6	pass 40.22686 (n12 table.rows[1][10], entry 117)	pass 37.8 via tolerance (n14, claim check 308)
`p_amplitude_animal_unit` (reference)p value of the amplitude with the animal as the unit	0.067 ±0.03	match 0.04764355 (n12 metrics.min_p_bh, entry 117)	not asked

Notes

Notes in the case file

The paper analyzes the files with an Igor Pro script and a cutoff of -12000 (internal units). The adapter uses a template search, so the counts differ, and the p values are the same order and the same direction, not equal. Cells from one animal are not independent. The ICC (intraclass correlation) over animals is 0.26 for the frequency. With the animal as the unit, the amplitude difference is no longer significant at 0.05 (p = 0.067).